presentation about marketing segmentation

• My presentations

Auth with social network:

Download presentation

We think you have liked this presentation. If you wish to download it, please recommend it to your friends in any social system. Share buttons are a little bit lower. Thank you!

Presentation is loading. Please wait.

CHAPTER 9 Marketing Segmentation, Targeting, and Positioning

Published by Shanon Hicks Modified over 9 years ago

Similar presentations

Presentation on theme: "CHAPTER 9 Marketing Segmentation, Targeting, and Positioning"— Presentation transcript:

UNIT 4 – MARKET SEGMENTATION

Market Segmentation, Targeting, and Positioning with Duane Weaver

Chapter 9 © 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.

Chapter Objectives Marketing Segmentation, Targeting, and Positioning CHAPTER Identify the essential components of a market. Outline the role.

Part Three Target Market Selection and Research Target Markets: Segmentation and Evaluation 7 7.

Target Markets: Segmentation and Evaluation

Learning Goals Learn the three steps of target marketing, market segmentation, target marketing, and market positioning Understand the major bases for.

Customer-Driven Marketing Strategy:

“You cannot be all things to all people”

Chapter 9 Market Segmentation, Targeting, and Positioning

© 2002 Pearson Education Canada Inc. 7-1 principles of MARKETING Chapter 7 Market Segmentation, Targeting, and Positioning for Competitive Advantage.

Copyright  2004 McGraw-Hill Australia Pty Ltd PPTs t/a Marketing: A Practical Approach 5/e by Peter Rix Slides Prepared by:Joe Rosagrata 4-1 Chapter 4.

Chapter 7- slide 1 Copyright © 2009 Pearson Education, Inc. Publishing as Prentice Hall Chapter Seven Customer-Driven Marketing Strategy Creating Value.

Objectives Be able to define the three steps of target marketing: market segmentation, target marketing, and market positioning. Understand the major.

7- 1 Copyright © 2012Pearson Education, Inc. Publishing as Prentice Hall i t ’s good and good for you Chapter Seven Customer-Driven Marketing Strategy:

© 2011 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Levels of Market Segmentation

Chapter 4 Segmenting and targeting markets

Chapter 10 Target Markets: Segmentation, Evaluation, and Positioning

About project

© 2024 SlidePlayer.com Inc. All rights reserved.

• Collections
• Market segmentation

Market Segmentation Presentation Templates

"it's much user-friendlier to double your business by doubling your conversions than to double your traffic." make your marketing segmentation conversation a huge success using our free powerpoint templates and google slides themes. use these slides to put your data to work..

Promote Your Ideas with Free Market Segmentation PowerPoint Templates and Google Slides Themes

We're here to help you, what are market segmentation powerpoint templates.

Market segmentation slides make communicating how your company is positioned inside the market easier. You can categorize and explain large or diversified markets into segments with various needs.

Where can we use these Market segmentation Templates?

Market segmentation slides can be used in presentations to demonstrate how to use the available materials to develop a value offer and obtain an advantage over competitors. Action plans to reach target clients will be presented with all the schematics and brand marketing models.

How can I make a Market segmentation Template in a presentation?

These presentations can be created using a variety of marketing-related themes and images. Additionally, readymade slides that can be quickly changed are available.

Who can use these Market segmentation Templates?

Marketing experts and analysts can utilize market analysis templates to create presentations on advertising campaigns.

Why do we need to use Market segmentation slides?

With the help of these customizable market segmentation visualizations, you can increase your chances of comprehending your target customers.

Where can I find Market segmentation templates for free?

There are a bunch of websites where you can obtain slides for free. Visit Slide Egg immediately to get the most prominent designs for no cost.

Home PowerPoint Templates Market Segmentation

Market Segmentation PowerPoint Templates & Slides for Presentations

Download 100% editable Market segmentation templates for presentations. Market Segmentation templates can be used to segment your customers and audience. Using different types of market segmentation individuals and analysts can target customers based on unique characteristics. Market segmentation can help to driven more effective marketing campaigns, and find opportunities in your market.

8-Step Circular Diagram PowerPoint Template

Hexagonal Quadrants PowerPoint Template

7-Column Diagram Template for PowerPoint

8-Step Layered Process Diagram PowerPoint Template

Layered Thirds Diagram PowerPoint Template

6-Item Wheel PowerPoint Template

Business Industry Analysis Presentation Template

Go-To-Market PowerPoint Template

Sector Wheel Diagram PowerPoint Template

PAM TAM SAM SOM Slide Template for PowerPoint

Market Analysis PowerPoint Template

Go to Market Presentation Template

Use our pre-designed marketing segmentation templates for PowerPoint and Google Slides to present your business’s market division and positioning. Our 100% editable presentation templates on marketing segmentation examples can help to divide broad or diverse markets into groups of segments with different needs.

A market can be segmented based on multiple criteria. Two of the most known segmentations are demographics and psychographic segmentation, but the quality of these approaches is somewhat limited. In order to achieve better results, business professional and marketing specialists can also segment based on behavioral patterns.

The market segmentation method can help to create targeted strategies that are essential to draw the attention who is interested in our services or products.

How can I use market segmentation templates?

Market segmentation templates can be used by marketing professionals and analysts to prepare presentations on Marketing Strategies. Our 100% editable templates for Market Segmentation can be used to speed up the design process at the time of preparing a business presentation with market segmentation analysis.

How to do a market segmentation?

There are several ways to segment a market, in our article How to Do a Market Segmentation the Right Way we analyze some of the ways business professionals and marketing specialists can segment a market.

Download Unlimited Content

Our annual unlimited plan let you download unlimited content from slidemodel. save hours of manual work and use awesome slide designs in your next presentation..

Newly Launched - AI Presentation Maker

Researched by Consultants from Top-Tier Management Companies

AI PPT Maker

Powerpoint Templates

Icon Bundle

Kpi Dashboard

Professional

Business Plans

Swot Analysis

Gantt Chart

Business Proposal

Marketing Plan

Project Management

Business Case

Business Model

Cyber Security

Business PPT

Digital Marketing

Digital Transformation

Human Resources

Product Management

Artificial Intelligence

Company Profile

Acknowledgement PPT

PPT Presentation

Reports Brochures

One Page Pitch

Interview PPT

All Categories

Top 10 Marketing Segmentation Templates to Build Stronger Customer Relationships

Hanisha Kapoor

If you believe market segmentation is a concept for modern business owners or digital age entrepreneurs’, then let us look at the renaissance period. This practice was developed in the 16th-century where retailers could not afford to serve one type of customer. They, therefore, used the windows of their stores to reach and cater to wealthier customers from “riff-raff.”

This concept of targeting a particular group of people gained traction gradually in the business world. And companies now use it to segregate their market into approachable groups based on demographics, needs, priorities, shared interests, and more.

So why is market segmentation important?

Market segmentation helps you curate and send brand messages to the right set of consumers. Having a defined market segment, companies are able to meet the needs of a variety of customers by attracting them with tailor-made campaigns and appealing offers. 

How to segment the market?

• Determine your market : Is there a need for your products and services? What is the current position of your company in the marketplace?
• Segment your market : Decide which criteria (demographic, psychographic, geographic or behavior) you want to use to segment your market.
• Understand your market : Conduct research surveys, polls, etc. Ask questions that relate to the audience.
• Create your customer segments : Analyze the results from your research to figure out your target audience.

Now the question comes, why do you need it?

Market segmentation empowers you to create product offerings for attracting the right customers. Therefore, deploy market segmentation strategies to increase brand loyalty and drive growth.

If you are looking for some resources to carry out the market segmentation for your business, let us show you some ready-made templates to target the right prospects!

Template 1: Market Segmentation Process PPT Template

This professionally designed PPT design will help you highlight the people that will add value to your business. Incorporate this PowerPoint template and understand your customer needs. Download this customizable PowerPoint graphic and focus on specific segments to provide appealing solutions.

Download this template

Template 2: Market Segmentation Solutions PowerPoint Slide

Deploy this ready-to-use PowerPoint slide and segregate prospective buyers into groups with shared interests and needs. Identify target groups and deliver attractive services to the prospects with this actionable PPT template. Incorporate this PowerPoint template and emphasize the profitable resources. Download now!

Grab this slide

Template 3: Market Segmentation and Targeting PPT Template

Figure out and understand your target audience better with this flexible PowerPoint design. Send your message to a specific group of consumers using this dynamic PowerPoint template. Empower your product development cycle by introducing impeccable services in the market. Grab this customizable PowerPoint slide and use it as per your company’s requirements.

Download this slide

Template 4: Market Segmentation PowerPoint Slide

Here is another PPT template for helping you identify the specific characters, needs, and interests of your customers. Level up your marketing strategies and design lucrative campaigns to target potential customers. Download these customizable PowerPoint slides to narrow your market and increase your profits.

Grab this template

Template 5: Market Segmentation Analysis PPT Graphic

Gain a better understanding of your customers’ interests and needs using this visually appealing illustration. Incorporate this professionally designed PowerPoint template and outline a plan to spend your resources, time, and money efficiently on the target group. Grab this actionable PPT slide and enable your company to differentiate its products based on the common dimensions of the market segment.

Template 6: Market Segment Evaluation PowerPoint Slide

Use this actionable PPT template to cater to the needs of your prospects. Comprehend your audience better and market your brand appropriately with this ready-made PowerPoint template. Deploy this customizable PPT graphic and offer better customer service.

Template 7: Market Segmentation Model PowerPoint Template

Incorporate segmentation strategies to attract prospects, develop exciting products, and increase sales. Use this professionally designed PPT template to divide your customers and curate customized marketing strategies that appeal to them. Grab this customizable PowerPoint slide and use it as per your brand requirements.

Template 8: Market Segmentation Model PowerPoint Template

This is another ready-to-use PowerPoint template for helping you prepare a questionnaire to understand your target market better. Use this professionally designed PPT slide and conduct preliminary surveys on your chosen segment. Download this customizable PPT design and analyze the responses from the research to highlight relevant audience for your business.

Template 9: Market Segmentation PPT Slide

Segregate your customers into different categories using this professionally designed PowerPoint template. This PPT slide is perfect for helping you understand the segmentation process with examples. Use this actionable PowerPoint illustration and sort your market.

Template 10: Market Segmentation PowerPoint Template

Extend your market research and seek customers who add value to your business. Incorporate this ready-made PowerPoint slide and determine the best ways to deliver your products and services to your target audience. Outline appealing marketing strategies using this flexible and customizable PowerPoint design.

Take your company’s sales a notch higher by targeting your content towards the right people. Utilize various market segmentation strategies and increase customer engagement by incorporating SlideTeam’s professionally designed PPT templates! 

P.S: Do you want to position your company strategically in the market? Use our exclusive guide featuring segmentation, targeting and positioning PPT templates to do so!

Related posts:

• Top 10 PowerPoint Templates for Comparative Cohort Analysis [Free PDF Attached]

Your Simplest Guide to Segmentation Targeting and Positioning [Professional Templates Included]

• Top 10 Price List Templates To Enlist Services With Costs
• How to Design the Perfect Service Launch Presentation [Custom Launch Deck Included]

Liked this blog? Please recommend us

15‌ ‌Personalized‌ ‌Marketing‌ ‌Templates‌ ‌for‌ ‌Better‌ ‌Customer‌ ‌Engagement‌ ‌

This form is protected by reCAPTCHA - the Google Privacy Policy and Terms of Service apply.

Market Segmentation PPT: Definition, Types, Importance and Mistakes

Market Segmentation PPT: Definition, Types, Importance and Mistakes Free Download: Market segmentation in marketing is the division of prospective customers into groups or segments with comparable needs and reactions to marketing efforts. Market segmentation enables companies to target different client groups that have different perspectives on the overall value of a given commodity or service.

Three factors can typically be used by businesses to distinguish between various market segments:

Within a section, homogeneity or shared requirements

differentiating oneself from other people or organisations

market response, or an equivalent response

For instance, a company that sells athletic footwear might have market segments for long-distance runners and basketball players. Basketball players and marathon runners react to commercials very differently as separate groups.

Table of Content

• Introduction
• Types of Marketing Segmentation 
• Importance of Marketing Segmentation 
• Mistakes of Marketing Segmentation
• Implementation of Marketing Segmentation 

Related Posts

Social media marketing ppt presentation seminar free, biomedical waste management ppt presentation free, monkey and the cap seller story ppt presentation free download, 1210 electrical engineering(eee) seminar topics 2024, 112 iot seminar topics-internet of thing presentation topics 2024.

330 Latest AI (Artificial Intelligence Seminar Topics) 2024

No comments yet, leave a reply cancel reply.

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed .

• Presentations

Market Segmentation Presentation Template

Get to your target audience using this market segmentation presentation template..

The orange and blue tones of this template perfectly contrast each other, so your content is visible and the visuals help you explain and illustrate your key points, so it is also easily readable.

This presentation template is a great tool to analyze the market and present your findings to clients and bosses, but also for making complex marketing strategies seem easy to execute.

• Change colors, fonts and more to fit your branding
• Access free, built-in design assets or upload your own
• Visualize data with customizable charts and widgets
• Add animation, interactivity, audio, video and links
• Download in PDF, PPTX, MP4 and HTML5 format
• Share online with a link or embed on your website

Create a new marketing strategy based on market segmentation with this contrasting presentation template, or look for a better fit among 500+ ready-to-use presentation templates in Visme’s library.

Edit this template with our  Presentation Software

Template Specifications

16 Slides, 1366 x 768 pixels – 16:9 Aspect Ratio (HD Presentation Quality)

Customizable

This template can be fully customized. You can edit content, change image(s), apply custom colors, input your own fonts and logo, and more.

Download as a PDF to share with your recipient or generate a shareable link for online sharing.

Compatibility

Related tags.

• presentations
• presentation

Explore other presentation themes

Create your Presentation using thousands of gorgeous slides in 20+ content categories.

More like this

Nature Background Presentation

Educate your audience on environmental topics or organic products with this nature presentation template.

Comms Pitch Deck Presentation

Share your creative business ideas with investors using this Intercom-inspired pitch deck template.

Evolution of the Projector Presentation

Communicate to and engage with your audience the right way with this educational media presentation template.

HR SWOT Analysis Presentation

Explore small-business HR with this artistic SWOT Analysis presentation template.

Graphic Design - Webinar Presentation

Give your audience a crash course on graphic design using this colorful webinar presentation template.

Communication Skills - Keynote Presentation

Give a show-stopping presentation on the importance of workplace communication with this modern keynote presentation template.

Employee Handbook Interactive Presentation

Transform your employee handbook into a dazzling interactive presentation with this professional interactive presentation template.

Business Case Presentation

Showcase your company's latest case study with this beautiful presentation template.

Risk Analysis Presentation

Walk your audience through conducting a risk analysis with this finance presentation template.

Employee Engagement Presentation

Explain how to boost employee engagement in the workplace using this presentation template.

Business Development Presentation

Create an engaging presentation on your company’s history and showcase its portfolio.

Design Tool Presentation

Use this compelling sales presentation template to highlight your SaaS company and product.

Purple and Black Business Presentation

This all-around, creative business presentation template is a definite head-turner. Get started today.

50 Years After the Moon Landing - Presentation

Design an eye-catching space exploration presentation with this stunning presentation template.

Cybersecurity SWOT Analysis Presentation

Learn about the importance of cybersecurity with this stellar SWOT Analysis presentation template.

Entrepreneurship Webinar Presentation

Reveal the secrets of successful entrepreneurship using this vibrant keynote presentation template.

Free Presentation Templates by Visme

Marketing is all about communication, yes - but it's also about making the best possible impression on your target audience. What you're saying is important, but how you choose to say it is equally so. You can only make one first impression, so you'd better make it the best one that you can - and now, thanks to Visme's presentation templates, it's easier than ever to do precisely that.

Loaded with countless stunning, versatile and totally customizable presentation templates , Visme's presentation software makes it possible to design the engaging, creative collateral that you need without requiring years of design experience under your belt. Visme offers presentation templates for every conceivable industry that you're a part of, making sure that the tools are always available to guarantee that your audience will pay attention to your every word.

Our presentation templates are equal parts colorful, visual, vivid and attractive - but they're also easy to use and even easier to edit, as well. These presentation templates also come with a massive number of free stock images for you to use, guaranteeing that you'll always be able to broadcast your message in exactly the right way at exactly the right moment in your relationship with your audience.

Create Your Presentation

Root out friction in every digital experience, super-charge conversion rates, and optimize digital self-service

Uncover insights from any interaction, deliver AI-powered agent coaching, and reduce cost to serve

Increase revenue and loyalty with real-time insights and recommendations delivered to teams on the ground

Know how your people feel and empower managers to improve employee engagement, productivity, and retention

Take action in the moments that matter most along the employee journey and drive bottom line growth

Whatever they’re are saying, wherever they’re saying it, know exactly what’s going on with your people

Get faster, richer insights with qual and quant tools that make powerful market research available to everyone

Run concept tests, pricing studies, prototyping + more with fast, powerful studies designed by UX research experts

Track your brand performance 24/7 and act quickly to respond to opportunities and challenges in your market

Explore the platform powering Experience Management

• Free Account
• Product Demos
• For Digital
• For Customer Care
• For Human Resources
• For Researchers
• Financial Services
• All Industries

Popular Use Cases

• Customer Experience
• Employee Experience
• Net Promoter Score
• Voice of Customer
• Customer Success Hub
• Product Documentation
• Training & Certification
• XM Institute
• Popular Resources
• Customer Stories
• Artificial Intelligence
• Market Research
• Partnerships
• Marketplace

The annual gathering of the experience leaders at the world’s iconic brands building breakthrough business results, live in Salt Lake City.

• English/AU & NZ
• Español/Europa
• Español/América Latina
• Português Brasileiro
• REQUEST DEMO
• Experience Management
• Brand Experience

Market Segmentation

What is market segmentation?

The benefits of market segmentation, the basics of segmentation in marketing, types of market segmentation, how to get started with segmentation, market segmentation strategy, market segmentation use case examples, ensuring effective segments, common segmentation errors, qualtrics solutions for market segmentation, see how qualtrics strategic brand works, market segmentation: definition, types, benefits, & best practices.

20 min read Market segmentation helps you send the right message, every time, by efficiently targeting specific groups of consumers. Here’s how it works.

By understanding your market segments, you can leverage this targeting in product, sales, and marketing strategies . Market segments can power your product development cycles by informing how you create product offerings for different segments like men vs. women or high income vs. low income.

Read on to understand why segmentation is important for growth and the types of market segmentation to use to maximize the benefits for your business.

Free eBook: How to drive profits with customer segmentation

Companies who properly segment their market enjoy significant advantages. According to a study by Bain & Company , 81% of executives found that segmentation was crucial for growing profits. Bain also found that organizations with great market segmentation strategies enjoyed a 10% higher profit than companies whose segmentation wasn’t as effective over a 5-year period.

Other benefits include:

• Stronger marketing messages : You no longer have to be generic and vague – you can speak directly to a specific group of people in ways they can relate to, because you understand their characteristics, wants, and needs.
• Targeted digital advertising : Market segmentation helps you understand and define your audience’s characteristics, so you can direct your online marketing efforts to specific ages, locations, buying habits, interests etc.
• Developing effective marketing strategies : Knowing your target audience gives you a head start about what methods, tactics and solutions they will be most responsive to.
• Better response rates and lower acquisition costs : will result from creating your marketing communications both in ad messaging and advanced targeting on digital platforms like Facebook and Google using your segmentation.
• Attracting the right customers : targeted, clear, and direct messaging attracts the people you want to buy from you.
• Increasing brand loyalty : when customers feel understood, uniquely well served, and trusting, they are more likely to stick with your brand .
• Differentiating your brand from the competition : More specific, personal messaging makes your brand stand out .
• Identifying niche markets : segmentation can uncover not only underserved markets, but also new ways of serving existing markets – opportunities which can be used to grow your brand.
• Staying on message : As segmentation is so linear, it’s easy to stay on track with your marketing strategies, and not get distracted into less effective areas.
• Driving growth : You can encourage customers to buy from you again , or trade up from a lower-priced product or service.
• Enhanced profits : Different customers have different disposable incomes; prices can be set according to how much they are willing to spend . Knowing this can ensure you don’t oversell (or undersell) yourself.
• Product development : You’ll be able to design new products and services with the needs of your customers top of mind, and develop different products that cater to your different customer base areas.

Companies like American Express , Mercedes Benz , and Best Buy have all used segmentation strategies to increase sales, build better products, and engage better with their prospects and customers.

Understanding segmentation starts with learning about the various ways you can segment your market as well as different types of market segmentation. There are four primary categories of segmentation, illustrated below.

With segmentation and targeting, you want to understand how your market will respond in a given situation, like what causes people to purchase your products. In many cases, a predictive model may be incorporated into the study so that you can group individuals within identified segments based on specific answers to survey questions .

Demographic segmentation

Demographic segmentation sorts a market by elements such as age, education, household income, marital status, family size, race, gender, occupation, and nationality. The demographic approach is one of the simplest and most commonly used types of market segmentation because the products and services we buy, how we use those products, and how much we are willing to spend on them is most often based on demographic factors. It’s also seen as a simple method of predicting future behavior, because target audiences with similar characteristics often behave in similar ways.

How to start demographic segmentation

Demographic segmentation is often the easiest because the information is the most readily available. You can send surveys directly to customers to determine their demographic data, or use readily available third party data such as government census data to gather further information.

Geographic segmentation

Geographic segmentation can be a subset of demographic segmentation, although it can also be a unique type of market segmentation in its own right. As its name suggests, it creates different target customer groups based on geographical boundaries. Because potential customers have needs, preferences, and interests that differ according to their geographies, understanding the climates and geographic regions of customer groups can help determine where to sell and advertise, as well as where to expand your business.

How to start geographic segmentation

Geographic segmentation data again can be solicited from customers through surveys or available third party market research data, or can be sourced from operational data such as IP addresses for website visitors.

Firmographic segmentation

Firmographic segmentation is similar to demographic segmentation, except that demographics look at individuals while firmographics look at organizations. Firmographic segmentation would consider things like company size, number of employees and would illustrate how addressing a small business would differ from addressing an enterprise corporation.

How to start firmographic segmentation

Firmographic segmentation data can be found in public listings for companies and information that the business makes available, as well as trade publications. Again, surveying existing and potential customers can help to build out this data.

Behavioral segmentation

Behavioral Segmentation divides markets by behaviors and decision-making patterns such as purchase, consumption, lifestyle, and usage. For instance, younger buyers may tend to purchase bottled body wash, while older consumer groups may lean towards soap bars. Segmenting markets based on purchase behaviors enables marketers to develop a more targeted approach, because you can focus on what you know they are looking for, and are therefore more likely to buy.

How to start behavioral segmentation

Of all the types of market segmentation, behavioral segmentation is likely best started with the information you have on an existing customer base. Though it can be bolstered by third party market research data, the information you already have on customer purchase and usage behavior will be the best predictor of future behavior.

Psychographic segmentation

Psychographic segmentation considers the psychological aspects of consumer behavior by dividing markets according to lifestyle, personality traits, values, opinions, and interests of consumers. Large markets like the fitness market use psychographic segmentation when they sort their customers into categories of people who care about healthy living and exercise.

How to start psychographic segmentation

Pychographic segmentation relies on data provided by the consumers themselves. Though market research might provide insights on what particular segments are most likely to believe or prefer, psychographic segmentation is best completed with information direct from the source. You can use survey questions with a qualitative focus to help draw out insights in the customers’ own voice.

There are five primary steps to all marketing segmentation strategies:

• Define your target market : Is there a need for your products and services? Is the market large or small? Where does your brand sit in the current marketplace compared to your competitors?
• Segment your market : Decide which of the five criteria you want to use to segment your market: demographic, firmographic, psychographic, geographic, or behavioral. You don’t need to stick to just one – in fact, most brands use a combination – so experiment with each one to figure out which combination works best for your needs.
• Understand your market : You do this by conducting preliminary research surveys, focus groups, polls , etc. Ask questions that relate to the segments you have chosen, and use a combination of quantitative (tickable/selectable boxes) and qualitative (open-ended for open text responses) questions.
• Create your customer segments : Analyze the responses from your research to highlight which customer segments are most relevant to your brand.
• Test your marketing strategy : Once you have interpreted your responses, test your findings by creating targeted marketing, advertising campaigns and more for your target market, using conversion tracking to see how effective it is. And keep testing. If uptake is disappointing, relook at your segments or your research methods and make appropriate changes.

Why should market segmentation be considered a strategy? A strategy is a considered plan that takes you from point A to point B in an effective and useful way. The market segmentation process is similar, as there will be times you need to revisit your market segments, such as:

In times of rapid change: A great example is how the Covid-19 pandemic forced a lot of businesses to rethink how they sell to customers. Businesses with physical stores looked at online ordering, while restaurant owners considered using food delivery services.

If your customers change, your market segmentation should as well, so you can understand clearly what your new customers need and want from you.

On a yearly basis: Market segments can change year over year as customers are affected by external factors that could alter their behavior and responses.

For example, natural disasters caused by global warming may impact whether a family chooses to stay living in an area prone to more of these events. On a larger scale, if your target customer segment moves away from one of your sales regions, you may want to consider re-focussing your sales activities in more populated areas.

At periodic times during the year: If you’ve explored your market and created market segments at one time of the year, the same market segments may have different characteristics in a different season. Seasonal segmentation may be necessary for better targeting.

For example, winter has several holidays, with Christmas being a huge influence on families. This holiday impacts your market segments’ buying habits, how they’ll behave (spending more than normal at this time than any other) and where they will travel (back home for the holidays). Knowing this information can help you predict and prepare for this period.

When considering updating your market segmentation strategy, consider these three areas:

• Acknowledge what has changed: Find out what has happened between one time period and another, and what have been the driving forces for that change. By understanding the reasons why your market is different, you can make key decisions on whether you want to change your approach or stay the course.
• Don’t wait to start planning: Businesses are always adapting to long-term trends, so refreshing market segmentation research puts you in a proactive place to tackle these changes head-on. Once you have your market segments, a good idea is to consider the long-term complications or risks associated with each segment, and forward-plan some time to discuss problem-solving if those issues arise.
• Go from “what” to “why” : Why did those driving forces come about? Why are there risks with your target market? At Qualtrics, we partner with companies to understand the different aspects of target markets that drive or slow success. You’ll have the internal data to understand what’s happening; we help unleash insight into why with advanced modeling techniques. This helps you get smart market segmentation that is predictive and actionable, making it easier for future research and long-term segment reporting.

Where can you use market segmentation in your business? We’ve collected some use case scenarios to help you see how market segmentation can be built out across several departments and activities:

Market and opportunity assessments

When your business wants to enter into a new market or look for growth opportunities, market segmentation can help you understand the sales potential. It can assist in breaking down your research, by aligning your findings to your target audience groups.

For example, When you’ve identified the threats and opportunities within a new market, you can apply your customer segment knowledge to the information to understand how target customers might respond to new ideas, products, or services.

Segmentation and targeting

If you have your entire market separated into different customer segments,  then you have defined them by set criteria, like demographics, needs, priorities, common interests, or behavioral preferences .

With this information, you can target your products and services toward these market segments, making marketing messages and collateral that will resonate with that particular segment’s criteria.

Customer needs research

When you know a lot about your customers, you can understand where your business is connecting well with them and where there can be improvements.

Market segmentation can help with customer needs research (also known as habits and practices research) to deliver information about customer needs, preferences, and product or service usage. This helps you identify and understand gaps in your offerings that can be scheduled for development or follow-up.

Product development

If the product or service you’ve developed doesn’t solve a stated problem of your target audience or isn’t useful, then that product will have difficulty selling. When you know what each of your market segments cares about an/d how they live their lives, it’s easier to know what products will enrich or enhance their day-to-day activities.

Use market segmentation to understand your customers clearly , so that you can save time and money developing products and services that your customers will want to purchase.

Campaign optimization

Marketing and content teams will value having detailed information for each customer segment, as this allows them to personalize their campaigns and strategies at scale. This may lead to variations in messaging that they know will connect better with specific audiences, making their campaign results more effective.

When their marketing campaigns are combined with strong calls to action targeted to the specific segment, they will be a powerful tool that drives your target market segments towards your sales channels.

After you determine your segments, you want to ensure they’ll be useful. A good segmentation analysis should pass the following tests:

• Measurable : Measurable means that your segmentation variables are directly related to purchasing a product. You should be able to calculate or estimate how much your segment will spend on your product. For example, one of your segments may be made up of people who are more likely to shop during a promotion or sale.
• Accessible : Understanding your customers and being able to reach them are two different things. Your segments’ characteristics and behaviors should help you identify the best way to meet them. For example, you may find that a key segment is resistant to technology and relies on newspaper or radio ads to hear about store promotions, while another segment is best reached on your mobile app. One of your segments might be a male retiree who is less likely to use a mobile app or read email, but responds well to printed ads.
• Substantial : The market segment must have the ability to purchase. For example, if you are a high-end retailer, your store visitors may want to purchase your goods but realistically can’t afford them. Make sure an identified segment is not just interested in you, but can be expected to purchase from you. In this instance, your market might include environmental enthusiasts who are willing to pay a premium for eco-friendly products, leisurely retirees who can afford your goods, and successful entrepreneurs who want to show off their wealth.
• Actionable : The market segment must produce the differential response when exposed to the market offering. This means that each of your segments must be different and unique from each other. Let’s say that your segmentation reveals that people who love their pets and people who care about the environment have the same purchasing habits. Rather than having two separate segments, you should consider grouping both together in a single segment.

Market segmentation is not an exact science. As you go through the process, you may realize that segmenting based on behaviors doesn’t give you actionable segments, but behavioral segmentation does. You’ll want to iterate on your findings to ensure you’ve found the best fit for the needs of your marketing, sales and product organizations.

We’ve outlined the do’s , so here are some of the dont’s :

• Avoid making your segments too small or specialized : Small segments may not be quantifiable or accurate, and can be distracting rather than insightful
• Don’t just focus on the segment rather than the money : Your strategy may have identified a large segment, but unless it has the buying power and wants or needs your product, it won’t deliver a return on investment
• Don’t be inflexible : Customers and circumstances change, so don’t let your segments become too entrenched – be prepared to let them evolve.

Market segmentation doesn’t need to be complicated to be effective. We would advise, though, to  get automated from the beginning . Forget spreadsheets – choose  market segmentation software  to measure and streamline your marketing strategy; as you grow, the technology will scale with you.

Innovative features such as Experience iD allow you to build your own customer segments and start personalizing experiences at scale based on the rich insights into your critical customer groups.

If you want to get a feel for your market segmentation upfront, before taking a step towards a streamlined and integrated system, trust us to take you through the research with our Market Segmentation Research service .

Related resources

Market fragmentation 9 min read, behavioral segmentation 20 min read, psychographic segmentation 11 min read, geographic segmentation 14 min read, demographic segmentation 14 min read.

Brand Perception

Brand Sentiment 18 min read

Brand intelligence 12 min read, request demo.

Ready to learn more about Qualtrics?

The No-Nonsense Guide to Market Segmentation (With Tips and Examples)

Marketing to the wrong segment can feel like barking up the wrong tree, or more specifically, barking up tens of thousands of wrong trees. That’s where effective marketing segmentation can bring in some serious value for your business.

Nearly everybody in sales has, at one point or another, heard someone reasoning that simply adding more people to the funnel will improve their sales numbers while preserving their conversion rate.

If you’re a sales rep making 30 calls a day, you might reasonably extrapolate that making 60 calls a day would double your closed deals. Unfortunately, it’s not that straightforward.

Building a sales process can be complicated. What one audience might find valuable might just be noise for another.

Different demographics respond differently to marketing campaigns, and finding the right market segment for your products or services can help you tailor your marketing strategies to be the most impactful they can be.

This guide to marketing segmentation will help you find your target audience and choose the best market segmentation strategies.

Table of Contents

What is market segmentation, what are the benefits of market segmentation, the 4 most common types of market segmentation, 10 other market segmentation techniques you should know.

• Market segmentation strategies
• How to do your own market segmentation

Frequently asked questions about market segmentation

Market segmentation in a nutshell.

Market segmentation is the process of qualifying companies (or people) into groups that respond similarly to specific marketing strategies. This is the first critical step in creating a marketing and sales process tailored to differentiate your business in the market and resonate across multiple demographics.

Market segmentation divides customers into segments based on shared characteristics, behaviors, or other attributes so you can create marketing strategies that appeal to entire groups. Your marketing segmentation strategy will be mainly influenced by what your product is and which types of companies are already buying it.

The history of market segmentation

Wendell R. Smith first coined the expression “market segmentation” in his 1956 publication Product Differentiation and Market Segmentation as Alternative Marketing Strategies . Smith wrote that modern marketing appeals to selective rather than primary buying motives.

In other words, consumers actively contrast products against one another rather than simply purchasing a product to satisfy an immediate need. This realization was the inception of the modern market segmentation we practice today.

Before 1956, there wasn’t a huge market variety, and general stores tended to carry only one or two brands’ versions of the same product. As time passed, more and more emerging brands began offering similar products and thus needed to differentiate themselves through branding and targeting different markets.

It wasn’t enough to just manufacture ketchup. You had to identify your brand as America’s ketchup , kids’ ketchup , or fancy ketchup .

Market segmentation provides several benefits to small teams and enterprises alike, including:

• Bang for your buck: With tailor-made, demographic-specific messages and advertising, companies can more effectively communicate with their audiences, begin boosting their conversion rates, and actually spend less on broad advertising.
• Better conversion rate: The more information you have about your various audiences, the more specificity you can add to your outreach, which will help your prospects convert more easily.
• Customer retention: By marketing toward customers who have already gone through their own buyer’s journey, segmentation makes it easier to keep them engaged and pitch them with occasional upgrades. And with the segment data you’ve captured, you know how to talk to them.
• Expanding your efforts: Segmentation can be a great way to pursue new markets that have something in common with your current markets.  

Want to generate more leads? (Of course you do.)

The  B2B Marketer’s Toolkit  collects 120+ of the best lead generation tips ever published on the Nutshell blog. Download it today!

GET THE TOOLKIT

Market segmentation helps savvy marketers categorize their target customers based on shared characteristics to keep their efforts focused and effective. Below are the 10 most common types of market segmentation: 

• Demographic segmentation
• Psychographic segmentation
• Behavioral segmentation
• Geographic segmentation

1. Demographic segmentation

Demographic market segmentation is the most commonly used form of market segmentation and entails categorizing your market based on age, gender, income, profession, race, religion, education, location, family situation, etc.

Demographic market segmentation examples:

• Switch to the cartoon channel and check out those commercials. Do Nerf guns and neon-colored slime appeal to someone your age? Yeah, us, too— bad example .
• Commercials for vacation homes may target people across ages, genders, locations, and other demographics, but they all appeal to customers with disposable income who are interested in travel.

2. Psychographic segmentation

More specific characteristics are categorized under the umbrella of psychographic segmentation. Less tangible than demographic segmentation, this classification method includes details like lifestyle, personality, beliefs, values, and social class.

This evaluation is essential because two individuals can possess identical demographic information but make purchasing decisions completely differently, thus requiring different marketing.

Psychographic market segmentation examples:

• Health and wellness advertisements might not go a long way with someone who prefers to spend their money on video games and energy drinks, even if they work in the same industry and live in the same apartment building.
• Advertisements for large social gatherings (events, clubs, bars) might not appeal to introverts who would much rather snuggle up with a book than be surrounded by other people.

3. Behavioral segmentation

Another example of market segmentation is behavioral segmentation. At its core, behavioral segmentation is the act of categorizing prospects based on their actions, usually within your marketing funnel. For instance, prospects who visited a landing page for an upcoming event might benefit from receiving a personalized invitation.

Segmenting your market based on behaviors is typically done by marketers within their marketing automation software. Still, any company with a mailing list has already performed behavioral segmentation simply by tracking prospects who have signed up to receive emails.

Behavioral market segmentation examples:

• Sending emails to website visitors who have left items in their cart. “But wait…come back!”
• A retargeting campaign that only displays ads to people who have previously purchased an item.

4. Geographic segmentation

Geographic market segmentation takes into account prospects’ locations to help determine marketing strategies. Although SaaS sales are relatively unaffected, a salesperson of gigantic coats knows to avoid pitching to Arizona residents.

Geographic segmentation variables and examples:

• Climate: Swimwear brands shouldn’t be targeting Alaska residents in January.
• Cultural preferences (based on location): For obvious reasons, McDonald’s in Germany sells beer.
• Population type: A bicycle company may segment its audience differently depending on the population type—rural (mountain bikes, thicker tires, and more durable frames), urban (road bikes, thin tires, and lightweight frames), etc.
• Density: A giant strip mall may require a high density of foot traffic to thrive.

5. Price segmentation 6. Firmographic segmentation 7. Generational segmentation 8. Life stage segmentation 9. Seasonal segmentation 10. Technographic segmentation 11. Needs-Based Segmentation 12. Value-Based Segmentation 13. Usage Rate Segmentation 14. Micro-Segmentation

5. Price segmentation

Price segmentation alters the price of similar products and services sold to different consumer groups. If you ever forced your kids to pretend to be under a certain age to qualify for the “kids eat free” special, then you understand the power and utility of price segmentation.

However, price segmentation can get much more granular. This type of segmentation can be used to identify customers willing to pay more for a particular product or service that they perceive to be more valuable.

Done correctly, price segmentation can capture the maximum revenue for each transaction.

Price market segmentation examples:

• Broad: Senior discount, veteran discount, coupons, etc.
• Granular: Computer processors are priced differently when sold to a company as a part (like inside an iMac) than when sold to a consumer as a standalone product.
• Even more granular: A marketing consultancy may base its prices entirely on the value it can generate for each client’s unique situation.

6. Firmographic segmentation

Instead of categorizing consumers based on age, location, income, etc, firmographic segmentation categorizes companies based on industry, annual revenue, job function, company size, location, status, performance, etc.

For B2B marketers, utilizing firmographic segmentation is non-negotiable to a high-performing marketing strategy.

Just as the demographic segmentation variables can help you form a buyer persona at the consumer level, firmographic segmentation can help you develop a buyer persona at the company level.

Firmographic market segmentation examples:

• Running different ads for different industries—real estate, finance, legal firms, etc.
• A B2B sales team only targeting companies with revenues over $100m.

7. Generational segmentation

Generational segmentation is almost comparable to the “age” variable in demographic segmentation. However, generational market segmentation goes beyond age by considering a particular generation’s preferences, habits, lifestyles, and attitudes.

It’s self-evident that the generations are vastly different. Someone born in the 1960s will likely have experienced a different culture than someone born in the 2000s.

Generational market segmentation examples

• Utilizing more memes on Facebook to target a larger percentage of Millennials.
• Altering your content publishing schedule to mornings to target a more significant percentage of Baby Boomers.

8. Life stage segmentation

Life stage market segmentation is the process of dividing your market based on the life stage of your target audience. For example, someone married with five kids may respond well to an emotional advertisement about convertibles during their midlife crisis.

Life stage market segmentation examples

• Ads about life insurance may not appeal to sophomores in college, but they may appeal to someone who just started a family.
• Someone who just entered the workforce for the first time may be more interested in a new apartment than someone who is retired.

9. Seasonal segmentation

Seasonal segmentation targets people based on their purchasing habits during certain periods of the year. It can include actual seasons (spring, summer, fall, winter), events (Coachella, Super Bowl), and holidays (Christmas, Mother’s Day).

Seasonal market segmentation examples

• A local you-pick berry farm may want to target its ads based on the fruit in season.
• A flower shop specializing in same-day delivery may want to ramp up its ad spend around Mother’s Day, targeting forgetful children.

10. Technographic segmentation

Much like firmographic market segmentation, technographic segmentation only applies to B2B audiences. It’s used to target companies based on the types of technology they’re using.

Whether it’s a customer relationship management (CRM) platform, a website CMS, or a niche-specific software tool, utilizing technographic segmentation can help enhance sales and marketing efforts.

Technographic market segmentation examples

• A company that develops WordPress plugins would have no business targeting companies that use a different CMS, like Wix.
• It would make sense for a SaaS company to target businesses using an app it just integrated with.

Want even more sales and marketing tips from our experts?

Sign up for our free newsletter!

SIGN UP FOR THE NEWSLETTER

11. Needs-based segmentation

The needs-based segmentation method focuses on the underlying needs and motivations of the target audience. This technique requires drilling down below the psychographic, demographic, and behavioral surface to uncover the “why” behind the customer’s actions.

With a needs-based analysis and market segmentation approach, marketers can deliver messaging that resonates with their audience on a much deeper level. When done correctly, needs-based segmentation often results in higher audience engagement, conversions, and customer retention.

Needs-based market segmentation examples

• A financial services company with a recently developed mobile banking app may want to target young adult career starters looking for guidance on managing their finances.
• It would benefit a sports apparel company with a high-performance, sweat-wicking line to market to athletes who seek a balance of functionality and comfort in their sportswear.

12. Value-based segmentation

Once marketing teams have determined a customer’s financial worth to the company, they can employ a value-based market segmentation strategy to define their target market. Knowing how much your customers contribute to your bottom line separates the high-value customers from the others for optimized revenue growth campaigns.

At the heart of this market segmentation technique is the customer lifetime value (CLV), which is an estimate of the total revenue expected from the customer during their relationship with the company. Marketers must consider factors such as purchase history, average order value, and retention rate to determine the CLV.

Value-based market segmentation examples

An airline might divide its target audience into two primary value-based segments: High-value and low-value segments. 

The high-value segment could include frequent flyers, possibly those who travel often for business and purchase more expensive seats and upgrades. Its low-value segment might consist of price-sensitive travelers who travel occasionally and usually opt for discounted options.

These are the marketing strategies the airline might implement for each:

• High-value segment: The airline could offer this segment priority boarding, VIP airport lounge access, and travel miles points to encourage brand loyalty.
• Low-value segment: Setting up an email campaign for this segment with messages about flight promotions, deals, and tips on stretching your travel budget could have a positive impact.

13. Usage rate segmentation

In usage rate market segmentation, marketers group their target audience according to two variables: The degree and frequency with which they interact with the company’s products and services. With this segmentation technique, marketers can create marketing strategies hyper-focused on a particular subset of customer needs and behaviors.

This is another segmentation approach that requires serious data analysis. It delves deep into when and how customers engage with your products and services. Some important metrics to consider are purchase frequency, feature usage, and website engagement.

Usage rate market segmentation examples

• It may make sense for an e-learning platform to offer high-usage students advanced course options, personalized recommendations, and opportunities to interact with instructors.
• A fitness tracker app could create in-app educational campaigns showcasing app functionality and personalized workout recommendations for users who don’t actively engage with the app.

14. Micro-segmentation

Micro-segmentation involves dividing a customer base into even more precise target groups in line with various factors. These factors can include any one or more of the segmentation techniques mentioned above. But, these micro-groups can also be determined through predictive analytics.

Marketers opt for micro-segmentation when they plan to run highly targeted campaigns to ensure they resonate with everyone in the target group. This type of segmentation calls for in-depth data analysis to pinpoint ideal candidates for the target audience.

Micro-market segmentation examples

• To promote a new line of work-appropriate attire, an ecommerce clothing retailer may target the following group: Young professionals between 25 and 35 living in urban areas who have recently purchased business casual clothing.
• A streaming service advertising its broad range of children’s shows and educational programs could consider targeting a group like this: Families with young children who primarily watch shows and other content on tablets.

Market segmentation strategies (and their pros and cons)

Every market segmentation strategy is different, but most of them follow one of two fundamental outlines:

1. Concentration strategy

Concentration strategy is when a company determines that its efforts are best focused solely on a single market segment. This strategy is particularly great for small, growing businesses with a viable use case within a specific market. Focusing on one segment will allow the company to invest more time, energy, and resources into one specific market, which minimizes advertising spend and potentially mitigates wasting efforts across multiple segments.

Concentration strategy is like putting all your cards on the table—if it doesn’t work out, it can end badly. If the market segment hasn’t been properly vetted and turns out to be a bust, all your marketing efforts could be wasted. Be sure to carefully plan and execute thorough market testing before committing your business to a single market segment.

• Pros: High conversion percentages, repeatable marketing practices, less marketing spend
• Cons: “All-or-nothing” growth potential is limited to segment size

2. Multi-segment strategy

Multi-segment marketing, or differentiated marketing, is when a company’s marketing strategies are designed to advertise one product to more than one market segment.

Although apparently “safer” than the concentration strategy, multi-segment marketing is a much larger tax on a company’s marketing spend, as it requires completely different campaigns for each market segment.

However, if a particular segment is highly receptive and converts well, it’s easy to tailor your strategy to market more directly to that segment.

• Pros: Safer, appeals to more consumers, diverse marketing, high growth potential
• Cons: Lower conversion percentages, greater marketing spend

How to do your own market segmentation in phases

Ready to complete market segmentation for your company? Here are three phases to follow during the process that will help you ensure you’re analyzing your markets effectively:

Phase 1: Gather the data

First things first—it’s time to gather data so you can use it to form your market segments. There are many ways to go about it—some people like to buy pre-made lead lists , and others prefer to do their own research. 

Two helpful methods of researching prospects are web forms and surveys. You can place high-quality data behind web forms that require site visitors to submit their name, email address, and other information to access the content. Surveys can get specific information from potential buyers in exchange for tangible rewards, like a gift card or special offer.

If you’re doing your own research, you can frame your searches along the following categories:

• Researching by company size: Size can mean several things, but it is most often measured by the number of employees, number of customers, or overall sales revenue a company claims. Some companies have greater transparency on their websites, which makes reaching out to the correct person much easier.
• Researching by industry: It’s unlikely that your product is applicable across all industries, which is why industry segmentation exists. Industry segmentation will help you ensure you’re not wasting your time by targeting a company without needing your product.
• Researching by location: If you’re offering a location-specific product or service, like landscaping services within the local community, your geographic market segmentation is probably pretty airtight—You probably use handy tools like lead maps, and engage in local marketing wherever possible. For other industries, like IT staffing, your reach might be international. Whatever your product, location is a crucial thing to know about a company because it will help you decide which sales tactics to use and when to send your emails if you’re communicating across time zones, at the least.
• Researching by requirements : This segmentation method entails qualifying companies based on whether they need your products or services. While this definition is straightforward, the process behind making this determination may not be, depending on what you’re offering. You can use Google Maps to look up a company’s HQ if you sell landscaping services. If their office is in a tower in New York City, they probably don’t need any landscaping.

Reporting & analytics in Nutshell

Learn more about Nutshell’s reporting and analytics features

SEE REPORTING FEATURES

Phase 2: Sort the data into segments

There are many ways to sort data. Most involve expensive analysts, marketers, and lots and lots of time. Although the DIY route is faster, it is no substitute for a comprehensive market segmentation strategy.

Assuming time and money are an obstacle—You can approximate your own market segmentation by compiling your data into one single source and running filters on it to group your prospects and companies manually by segment.

Remember, ask yourself the following:

• Is this segment measurable?
• Is this segment large enough to earn a profit?
• Is this segment stable and not going to vanish after a short time?
• Is this segment reachable with my marketing strategies?
• Is this segment homogenous, and will they respond similarly to my marketing strategies?

Phase 3: Plug in your marketing channels

Now that your segments have been firmly established, it’s time to connect the dots and breathe life into your marketing. This means establishing a plan for each of your marketing tools and channels and coming up with real ways to reach your segments with them.

You’ll be attributing different marketing and sales tactics to each stage of your pipeline and determining what sticks. The good news is that your market segments are clearly defined, and you’ll be able to speak to them clearly.

The real challenge is continuously improving your efforts with trial and error to get the best possible conversion rates.

There’s a good, old-fashioned way to map this out quickly and easily:

• Draw your pipeline stages horizontally across a sheet of paper.
• Above each pipeline stage, jot down marketing channels, like Linkedin, emails, or webinars, with blank space in between them.
• Below each marketing channel, write exactly how you will use this tool at this pipeline stage, like “email prospects a link to a recorded webinar.”

Repeat this exercise for each market segment to help establish a concise and repeatable process for marketing to your various audiences. You can fully flesh out your segmented marketing strategy by configuring your sales software and email automation around the outline you’ve created, then make tweaks as needed.

To this end, some CRMs offer reporting and performance tracking , as well as custom reporting , to help you figure out what’s working and what needs to change.

Still have questions about market segmentation? Check out the FAQs below for answers to some common questions:

What are some common challenges faced when implementing market segmentation? 

Here are a few of the challenges you may encounter when implementing your market segmentation strategy:

• Creating segments that are too broad: Your product or service may appeal to several different market segments, but trying to appeal to too many can lead to ineffective marketing and high ad spending.
• Creating segments that are too narrow: The opposite problem can also arise. Small segments might be challenging to quantify and distract from other segments with greater buying power.
• Not being flexible: Just because a particular segment currently buys from you doesn’t mean they always will. Be willing to reevaluate your market segments over time to maximize your marketing spending and revenue.

What are the key factors to consider when selecting a target market segment? 

Five key factors to consider when selecting market segments for your marketing strategies are:

• Whether the segment is measurable
• Whether the segment is large enough to generate a profit for your business
• If the segment is stable and won’t vanish after a short time
• If the segment is reachable by your marketing strategies
• Whether the segment is homogenous and will respond similarly to your marketing strategies

How can you effectively redefine your target market?

If you’ve determined that your target market no longer fits, you can always identify new markets . Here are a few tips for doing so: 

• Identify trends and patterns: Do companies that make a certain amount of annual revenue seem to be shying away from your offerings? If you want to reach those customers, identify any patterns in which products or services they choose instead and strategize for how to provide the value they’re looking for.
• Listen to customer feedback: Your current (or former) customers are valuable sources of feedback. Consider what they say about your product or service and whether you’re meeting their needs. You may be able to identify new opportunities.
• Diversify your marketing channels: Using multiple channels to reach your target market can be a highly effective way to increase your brand’s exposure. Consider diving into new channels like content marketing, email marketing, SEO, and online advertising to drive engagement with your target audience.

Additional resources:

• How to Use Data Insights to Inform Marketing Strategies
• 11 Expert Digital Marketing Communication Tips
• What Is CRM Data & How to Use It for Marketing
• The 30 Best Online Marketing Resources
• Creating and Implementing Effective Sales Strategies
• 7 Types of Sales Leads and How to Close Them
• The Ultimate Guide to Cold Calling
• The Ultimate Guide to Writing Cold Emails That Get Replies

Now, you’ve clearly segmented your demographics, figured out your strategy, and mapped your sales processes tightly to your market segments.

Because of this, you should clearly understand how to talk to your prospects and differentiate your outreach efforts based on the market segment.

The challenges that lie ahead are rooted in constantly adjusting your marketing. That means testing your messages and tactics and measuring your audiences’ responses.

If you’re ready to put your sales and marketing automation into action, get started with a free trial of Nutshell today!

See Nutshell in action

Give our powerful, easy-to-use CRM a try for free for 14 days! Or join a live demo to see Nutshell at work!

START YOUR FREE TRIAL

SEE A LIVE DEMO

Ready to try Nutshell for Free?

Top PPC Lead Management Tips for Your Company

Nutshell Earns First-Ever ‘Leader’ Placement in G2’s Winter 2022 Grid® Report for Email Marketing

How to Secure Your CRM Integrations Effectively

The best sales prospecting tools that integrate with your crm, how to calculate email marketing roi (and why you should).

Join 30,000+ other sales and marketing professionals. Subscribe to our Sell to Win newsletter!

This element relies on Javascript to run. Please enable Javascript in your browser for the best viewing experience. Visit our contact page if you have questions or are in need of assistance.

Market Segmentation

Sep 19, 2014

240 likes | 448 Views

Market Segmentation. Special Topic Mktg 633. Objectives. Definition Reasons for segmentation Bases of segmentation Applications Product Positioning. Definition. The process of dividing all possible users of a product into groups that have similar needs the products might satisfy.

Share Presentation

• segmentation
• segmentation approaches
• extensive segmentation
• select segmentation strategy
• 3 general consumer groups

Presentation Transcript

Market Segmentation Special Topic Mktg 633

Objectives • Definition • Reasons for segmentation • Bases of segmentation • Applications • Product Positioning

Definition • The process of dividing all possible users of a product into groups that have similar needs the products might satisfy.

Historical Development • 50s & 60s: Mass Marketing • 90s: Breaking Up Into Smaller Groups • Extensive Segmentation and targeting smaller consumer groups necessarily produces higher marketing costs

Criteria for Segmentation • Know Your Customers (Scanner Data) • Make What Customers Want. Offer many variations on a single product. • Reach Target Customers. Ex: “smart” cash registers.

Segmentation Approaches • A Priori Approach • Post Hoc Approach

Market Segmentation • Analyze the Consumer/Product Relationship • Essence of developing effective marketing strategies is understanding the special relationship that consumers have with products and/or brands. Concerns how consumers perceive a product as relevant for their lifestyle, salient consequences and values, and self-concepts.

Market Segmentation • Determine Segmentation Bases • Geographic • Demographic • Sociocultural • Affective and Cognitive • Behavioral

Using VALS 2 in Marketing • Who your consumers are? • What are they buying? • Where are they? (geodemographic) • How to effectively communicate with customers? • Why? Relate consumers underlying psychological values to their lifestyles and buying behavior • http://www.sric-bi.com/VALS/l

Sample Questions from VALS 2 • I am often interested in theories • I like outrageous people and things • I like a lot of variety in my life • I love to make things I can use everyday • I follow the latest trends and fashions • Just as the Bible says, the world literally was created in six days • I like being in charge of a group • I have more ability than most people • I must admit that I like to show off • I like trying new things Measured on a 4-point scale from Mostly Disagree to Mostly Agree.

VALS I Typology

VALS 1 • Classifies the American population into 4 general consumer groups • Need-Driven (Sustainers and Survivors) • The poor and uneducated (11%) • Outer-Directed (Achievers, Emulators and Belongers) • Lifestyles directed by external criteria (67%) • Inner-Directed (Societally Conscious, Exeriential and I-Am-Me) • Motivated by personal needs than by expectations of others (20%) • Integrated • Those who are able to combine the best of both outer- and inner-directed values (2%)

VALS 2 • Classifies the American population into 3 general consumer groups and then subdivides into 8 distinctive subgroups • Principle-Oriented • motivated in their choice by their beliefs, rather than by desire for approval • Status-Oriented • guided by actions, approval, and opinions of others • Action-Oriented • propelled by a desire for social or physical, variety, and risk-taking

Example of VALS 2 Segment: Principled • 35-64 years, Married • College graduate • 11% of U.S. adult • Well informed, “info junkie” • Value education and travel • Practical consumers • Prestige, image unimportant • Watch TV news

Market Segmentation • Select Segmentation Strategy • Do not enter • Mass market (mass market strategy) • One segment (concentrated strategy • More than one segment, separate marketing mix for each (differentiated strategy)

Product Positioning • Determining how to differentiate one’s product from those of competitors in the minds of consumers • 1. Position on product attributes • Ex: price (Budget Car Rental) • 2. Position on benefits • Ex: Close-Up’s benefit of fresh breath • 3. Position on usage • Ex: Michelob: special occasion beer • 4. Position on user category • Ex: Wheaties as the “breakfast of champions”

• More by User

Market Segmentation. Consider. the role of segmentation in marketing strategy types of market segmentation in consumer markets B2C industrial B2B criteria & bases for segmenting consumer markets. the segmentation process & basic strategies positioning & repositioning

2.91k views • 50 slides

Market Segmentation. Purpose of this section. 1. Introduce the Concept of the MARKETING PLAN 2. To Define Market Segmentation 3. Present 4 types of market segmentation 4. Aspects of the Canadian market 5. Main types of segmentation in industrial markets.

1.08k views • 49 slides

Market Segmentation. Segmentation, Targeting, Positioning. Markets people and institutions with: willingness to buy purchasing power authority to buy. Types of Markets. Classified by: purchaser reasons for buying the goods Consumer products

1.4k views • 51 slides

MARKET SEGMENTATION

MARKET SEGMENTATION. Learning Objectives. 1. Segmenting the Market 2. Benefit/Purpose and Limitation of market segmentation. 3. Market Segmentation Procedure. 4. Basis for market segmentation.

923 views • 40 slides

Use of advertising specific to customers. Producing goods and services customers want. Target the Market. By not making goods which consumers do not need or want. Avoids waste. Meeting the needs of people more efficiently and effectively. By not marketing to the wrong people.

366 views • 11 slides

Segmentation Using Customer Behavior. Customizing visitor experiences to match the site usage behavior patterns of each visitor or type of visitor is called

377 views • 13 slides

Market Segmentation. Objectives:. Describe the characteristics of markets and market segments. Discuss criteria for successful market segmentation. Discuss alternative strategies for selecting target markets. Market.

506 views • 24 slides

Market Segmentation. (chapter 6) (3 rd semester). Meaning of Market Segmentation. Market segmentation is one of the most interesting and useful tools in market segmentation.

198 views • 0 slides

Market Segmentation. November 11, 2010. Previously…. Definition of Marketing Purpose of Marketing Marketing Concept 7 Functions of Marketing. Channel Management Financing Marketing Information Management Pricing Product / Service Management Promotion Selling.

443 views • 23 slides

Market segmentation

Avi Chemay presents. Market segmentation. Ultimate aim of market segmentation. The aim of market segmentation is to increase sales, market share and profits This is done by better understanding and responding to the desires of different target customers . Process of market segmentation.

753 views • 13 slides

Market Segmentation. Chapter 2. Why would marketers want to target teens?. Impulse buyers Disposable income Building brand loyalty. Identifying Your Consumers. What is market segmentation?. D ividing a broad target market into subsets of consumers who have common needs.

1.07k views • 12 slides

Market Segmentation. AS Business Studies Marketing. Look at the other students in the class what characteristics do you all share?. What kind of products are you interested in?. Market Segmentation.

384 views • 18 slides

Market Segmentation. dividing a particular market into sections, which display similar characteristics or behaviour. Demographics. Demographic Segmentation. Age – different needs (kids, old people) Gender – cosmetics, clothing, All Bar one ( London ) –encourage women’s social drinking

248 views • 14 slides

Market Segmentation. Lifelong Learning 2007 Presented by Greg Marsello. We Must Segment the Market …80/20 Principle. Why Segment the Market?. You serve some, but not all, demographic groups. There is no average participant. Roughly 20% of your customers give you 80% of your operating margin.

543 views • 32 slides

Market Segmentation. Today you will know what market segmentation is. You will understand how and why businesses segment their markets. You will be able to apply this to the Levi’s case study. Market Segmentation (1).

382 views • 7 slides

MARKET SEGMENTATION. Market segmentation is based on fact that markets are heterogeneous & not homogeneous. Homogeneous market refer to a market situation where the prospective buyers of any product are found to be uniform in their needs, habits ,choices, nature, etc. CONT-------.

466 views • 21 slides

Market Segmentation. Market Segmentation. Definition of a Target Market : A sub-group of a larger market chosen as the focal point for a marketing or advertising campaign. Market Segmentation.

669 views • 39 slides

Market Segmentation. Jeremy Kees, Ph.D. Segmentation, Targeting, and Positioning. Market Segmentation Market segment —a group of customers who share a similar set of needs and wants NOT to be confused with targeting and positioning. Our overall intent is to….

870 views • 23 slides

Market Segmentation. Trevor Twyman 200 2 CIM Advanced Virrat. Communications Plan. Trevor Twyman 200 2 CIM Advanced Virrat. SOSTAC Planning system:- S - Situation Analysis ( where are we now ? ) O - Objectives (where do we want to go ? ) S - Strategy ( how do we get there ? )

195 views • 14 slides

Market Segmentation. Robin Purohit VP of Product Management VERITAS Software. Who am I. 13 years experience as developer, architect, strategic planner, PM Telecommunication, Enterprise Computing, Networking, Storage HW and SW 4 years @ VERITAS as PM Currently manage 75 person PM org

195 views • 15 slides

Market Segmentation. The process of dividing a potential market into distinct subsets of consumers and selecting one or more segments as a target market to be reached with a distinct marketing mix. Market segmentation. More precise definition of customers needs and wants.

1.01k views • 40 slides

MARKET SEGMENTATION. What is the market segmentation?

145 views • 12 slides

B2B Customer Segmentation: Unlocking Business Growth

Aug 30, 2024

10 min. read

No two customers are quite alike. But dig deeper and you’ll find that some of your B2B customers share unique characteristics. Knowing these characteristics — a process called B2B segmentation — makes it easier to tailor your marketing and customer journey in a way potential customers will notice.

In marketing, generalized messaging isn’t always effective. General messaging talks to everyone and no one. There’s nothing that makes your customer stop in their tracks and think, “Hey, they really get me!”

That’s where B2B customer segmentation can help. While many marketers associate segmentation with consumer marketing, it also applies to business customers. Here’s how you can use this strategy to reach more customers and drive more sales.

What Is B2B Customer Segmentation?

B2b customer segmentation importance, what is an example of b2b customer segmentation, key differences between b2b and b2c customer segmentation, methods for effective b2b customer segmentation, best practices for successful b2b customer segmentation, implementing b2b customer segmentation in your business.

B2B customer segmentation refers to how you divide your business’s customers into groups, or “segments.” These segments are based on shared characteristics, such as the products they purchase or the lead channels they came through.

For example, a business selling software might have customers in multiple industries. Companies might vary in size (e.g., startups, enterprise customers). Knowing these characteristics, you can tailor your marketing, sales, and product offerings to meet their unique needs. This might have an affordable solution for smaller companies and more advanced features for enterprises, for example.

You might also hear customer segmentation referred to as audience segmentation . Same thing.

Ultimately, B2B segmentation helps you target your efforts more effectively. You can put the right message in front of the right person via the right channel. This also helps you prioritize resources and find growth opportunities within different segments.

The importance of customer segmentation cannot be overstated. Segmentation allows you to tailor your sales and marketing to different types of customers in ways that matter to them. They’re more likely to notice your campaigns when you speak directly to their needs. 

It also gives you a more accurate picture of who your target customers are and what makes them buy from you. Segmenting is powerful for surfacing your competitive advantages, and you can build on these insights to attract more customers.

Want to see how Meltwater Consumer Intelligence can help you create new customer segments? Click here and fill out the form below for a personalized demo!

Let’s dig deeper into our software business example of B2B segmentation. Here are customer segmentation examples to put things into perspective.

Company size

Your customers might include small businesses, mid-size companies, and large enterprises. A B2B customer segmentation strategy might include:

• Smaller storage packages, basic features, and lower monthly fees for small businesses
• More storage, advanced features, and scalable pricing plans for medium companies
• Extensive storage, enterprise-level security features, dedicated customer support, and custom pricing for large companies

Companies of all industries rely on software for many aspects of their business. Knowing which industries you serve can help you better segment your customers and tailor your products to them. 

Segmentation for B2B in this case might include the following:

• Technology firms might need fast, scalable solutions and integrations with other tech tools.
• Healthcare organizations focus on data security and compliance with regulations like HIPAA.
• Schools and universities could be interested in discounted pricing for educational institutions.

In this case, you’re looking at the characteristics that stand out to each type of industry. You want to align your messaging with what they think is important in a software solution.

Geographic location

Segmenting customers by geographic location is a common option for B2C marketing, but it can also apply to B2C. 

Here’s how segmentation for B2B might look:

• Customers in North America may prefer local data centers and 24/7 customer support in their time zone.
• European clients might prioritize GDPR compliance and multilingual support.
• Businesses in the Asia-Pacific region might gravitate to flexible pricing and local partnerships.

Buying behaviors

Another option is to pay attention to your customers’ buying behaviors. Various customer behaviors can indicate their loyalty toward you and their willingness to buy from you in the future and refer you to others. 

B2B customer segmentation examples in this case may include:

• New customers who likely need guidance early and often
• Long-term clients who might be interested in premium features or upsells
• Price-sensitive customers who prioritize cost over features and might respond better to discounts or special offers

Tip: Check out our Personalization at Scale Guide to discover how in-depth customer segmentation can help you create better customer experiences!

There’s some overlap between B2B and B2C customer segmentation, but the differences between the two are worth mentioning.

Customer characteristics

In B2C, your customers are individual consumers. Marketers usually segment customers by demographics (age, gender, income), psychographics (lifestyle, values), or behavior (purchasing habits).

In B2B, other characteristics matter. Businesses are your customers, not the general public. You can segment customers based on things like company size, industry, revenue, or even job function.

The decision-making process

The decision-making process in B2B is often more complex and involves multiple stakeholders.

For instance, a purchase might require approval from a manager, finance department, and even the CEO. Because of longer and more complex buying cycles, marketers usually need to account for the roles and responsibilities of various decision-makers.

In B2C, the decision-making process is usually simpler — it typically involves just one or two people. Segmentation here might focus on individual preferences, emotional triggers, and buying motivations.

Purchase size and frequency

B2C transactions are usually smaller and more frequent. Segmentation focuses on purchase behavior, like how often someone buys a product or their average spend per transaction.

In B2B, transactions are larger in scale and the relationships are long term. Segmentation may focus on the frequency of purchases, contract size, or the potential for repeat business. For example, you might segment based on high-value clients versus occasional buyers.

Relationship focus

Relationships are critical in B2B. Businesses tend to build long-term partnerships with vendors or service providers. Therefore, B2B segmentation may focus on customer loyalty, relationship stage (new vs. long-term client), or the potential for upselling.

While relationships matter in B2C, especially for building brand loyalty, the focus is heavier on attracting new customers and encouraging repeat purchases. Segmentation might target loyal customers, first-time buyers, or those at risk of churning.

The goal of B2B segmentation is to group your customers in a way that allows you to personalize your marketing to specific needs. These three B2B customer segmentation models can help you start uncovering potential customer segments.

Firmographic segmentation

Similar to segmenting by demographics, firmographic segmentation refers to dividing customers by company attributes. This could be company size, industry, revenue, or even the customer’s target customer.

Focusing on firmographics gives you insight into how a company operates and what their needs might be. You can better align your product offerings when you know their needs and preferences.

Tiered segmentation

Many companies offer their B2B services in tiers — multiple levels of service at varying price points. You might market to each tier in different ways.

For example, your lowest-priced tier might cater to small businesses or price-sensitive customers who prioritize value. 

You can also peel back the layers of each tier to find sub-segments of each type of customer. 

Needs-based segmentation

A needs-based approach focuses on finding the specific needs or pain points of your customers. 

One option is to segment customers based on the challenges they face. For example, some businesses might prioritize cost savings, while others focus on improving efficiency or compliance.

Different customers might require different solutions. Segmenting by the type of solution they need — such as basic, advanced, or customized — can help you better meet their needs.

These are just a few types of B2B customer segmentation. As you explore different ways to group your customers, you’ll likely uncover multiple B2B customer segmentation characteristics that may prove valuable in your marketing.

The challenges of B2B customer segmentation can prevent companies from taking full advantage of this opportunity: a lack of marketing data, complex buying cycles, and changing market conditions, for example.

To get the most from B2B customer segmentation, we recommend following these best practices.

Define clear objectives

Take a moment to clarify your marketing goals. Are you trying to increase sales, improve customer retention, or optimize your marketing efforts? Your objectives will guide the segmentation process.

Collect high-quality customer data

Make sure the data you’re using is reliable and up to date. You can check your CRM, sales data, and external sources to learn more about your customers and the relationships between them.

Consumer intelligence platforms like Meltwater can also help you find customer segments. AI-powered tools can find connections between data that might evade the human eye.

Start with simple customer segments

Segmentation doesn’t have to be overly complicated. Start small with basic, broad segments. Test these segments and adjust your approach before moving into more complex segments.

Create personas for each segment

There’s a difference between personas vs segments , and both can help you effectively market your business when done right.

Your segment is a group of customers who share similar characteristics. Of that group, you can create one or more “personas” that represent people within the segment.

Personas are helpful when designing your marketing messaging. The personas should represent the customers within the segment. You can ensure you’re speaking directly to your ideal customer and get rid of any messaging or tactic that doesn’t cater to their needs.

Test and refine your customer segments

As you develop your customer segments, make sure you test how well your marketing performs for each segment. Customer segmentation tools like Meltwater can help you keep the details of each segment in one place and track your efforts.

Customer segmentation for B2B can help you maximize your marketing resources while minimizing the effort to reach the right audience. Meltwater can help you surface segments with ease, including those you might not have thought to explore.

Meltwater is an AI-driven data platform that lets you tune in to your customers across the web. We analyze billions of data points in real time, allowing you to keep tabs on conversations that matter to your business. 

Discover more about your customers’ needs and what they’re saying about your products or industry, and get inside their minds so you can market to them more effectively.

Learn more when you request a demo by filling out the form below!

Continue Reading

What Is Customer Segmentation? Definition & Strategy 2024

7 Top Customer Segmentation Tools

Personalization at Scale: Removing Guesswork from Consumer Research

Ultimate Guide to Market Segmentation & Personas

More From Forbes

20 expert strategies for measuring market size.

• Share to Facebook
• Share to Twitter
• Share to Linkedin

Understanding the size of your product's or service's intended market can help you make more informed business decisions. Whether you're entering a new industry or expanding within an existing one, it is important to stay on top of the opportunities and risks involved.

To help you do this, Forbes Business Development Council members share insights on the best methods for market sizing and why accurate measurement is crucial for long-term success. From social listening to modeling tools to customer interviews, these approaches can help you accurately refine your market analysis.

1. Understand The Serviceable Obtainable Market (SOM)

The Serviceable Obtainable Market (SOM) is the most important market size to focus on. This is the market where your current services really fit well, versus the Total Addressable Market (TAM) where it's an unrealistic total market where your services don't entirely fit. Understanding the SOM will help you understand where to invest your energy and money. - Andrew Davidson , Fastenal

2. Know Your Audience

When stepping into a new arena, it’s crucial to understand your audience, size up the competition and devise innovative strategies to pin down a competitive edge. If you want to make an impact, understanding the market size allows a company to flex its strengths effectively. This approach not only captures market share but also attracts new fans, ensuring a powerful presence in the marketplace. - Matthew Rolnick , Real American Beer

3. Use Social Listening Tools

I like the idea of social listening tools. Monitoring social media and online discussions helps identify trends and customer needs in real time. Nike does this well by using social listening tools to measure market sentiment and size, stay ahead of trends and ensure their products meet market demands. - Bryce Welker , The CPA Exam Guy

Today’s NYT Mini Crossword Clues And Answers For Wednesday, September 4

Mega millions jackpot hits $740 million—here’s how much a winner could take home after taxes, harris will propose $50,000 small business tax deduction for startup expenses: here’s what to know, 4. determine key metrics.

Identify key metrics like the number of potential customers, average purchase value and consumption rate for the target market from sources like government data, industry reports, surveys and so on. Calculate market size by multiplying the number of potential customers, average purchase value and consumption rate. Market sizing is critical for decision-making, planning and resource allocation. - Salice Thomas , Wipro Limited

Forbes Business Development Council is an invitation-only community for sales and biz dev executives. Do I qualify?

5. Assess A Wide Set Of OEMs

Analyze a broad set of OEMs and publicly reported financials and back out the numbers for the market you are trying to measure. Do this to gauge the market size, growth rate and segmentation. Also, benchmark your growth rate to your competitors. If you are not growing minimally at the rate of the OEMs and in line with competitors, your sales and operations plans need to be adjusted for success. - Manoj Tandon , Dark Rhino Security

6. Use Secondary Market Research

Calculating the total addressable market requires research and data analysis using firmographic factors like industry, revenue, size and geographic location. With secondary market research from Forrester or Gartner, you get a better idea of how many potential users meet your market criteria—including industry size. With this, you can focus go-to-market on the most valuable, strategic opportunities. - Elizabeth Kiehner , Nortal

7. Define Use Cases And Aggregate Industry Data

The best way to measure market size is to define the use case, target audience and industry. Identify four major industry publications for benchmarks. Aggregate data from these sources to determine your Total Addressable Market (TAM). This is crucial for ensuring your product's sustainability and growth potential over the next five to 10 years, guiding your GTM strategy and confirming product viability. - Raviraj Hegde , Donorbox

8. Study Sales Data

Analyzing sales data is the method for determining market size, as it gives solid proof of real market performance and customer habits. By studying sales data companies, you obtain insights into market patterns and possibilities. This approach is vital, as it sets standards and guides choices guaranteeing that resources are used efficiently and market prospects are evaluated accurately. - Thasha Batts , Pinnacle Global Network

9. Research TAM And SAM

An aggregation of market research data from different reliable entities (like market research firms, analyst firms and management consulting firms) is a fair indicator of market size. Market size definitions in terms of TAM and SAM are key inputs required to define go-to-market strategy (sales channels, product management, marketing, pricing, service strategy and so on.) - Bindesh Pandey , Comviva Technologies Limited

10. Gauge Metrics At Predetermined Intervals

Market size is a key metric that should be measured before you make any attempt to enter the market, as size dictates efforts and resources in order to be impactful. The best way to go about this is to gauge the total serviceable market size at predetermined intervals over the course of a specified time period and use that to develop a baseline. This will account for any changes due to seasonality or other events. - Mustansir Paliwala , Zomara Group & EQUANS

11. Calculate Willingness To Pay In Adjacent Markets

The best way to measure the size of the market is by understanding the pain point you are trying to solve, the associated willingness to pay and unit economics and multiplying that with a conversion ratio based on the number of deals that are potentially available in the segment. This exercise, repeated for all adjacent markets beyond the core market, can reveal the total and service obtainable markets. - Nimay Parekh , Security Scorecard

12. Use A Top-Down And Bottom-Up Approach

Measuring TAM (total addressable market) is key for any business; in fact, it is even more critical at each customer and prospect level. There are three key ways to get the TAM nailed: 1. Top-down with the number of target customer(s) footprint and average spend they do; 2. Bottom-up with assumed revenue capture with the target customer base and 3. Exhaustive with account-based TAM aggregate. - Bharath Yadla , Workato

13. Hire A Local Market Expert

Get a real market authority, not just by states but by cities. Most market size calculations are not realistic, so they are not actionable with simple statistics like case studies. If leaders cannot guide team members with practical sales strategy, none of the stakeholders or investors will be convinced. The best way is to hire a local citizen, who has understood the market for generations with their neighbors. - Gyehyon Andrea Jo , MVLASF

14. Test The Market With Both AI And Traditional Methods

AI can help you get some good research, but AI is still in the early stage and not always perfect. Testing the market through panels and perhaps discussion groups may be old school compared to AI results, but it can still work and give you additional insights that AI is not able to achieve right now. A combination of both is probably the best way to move forward. - David Strausser , Fonseca Advisers

15. Conduct Qualitative Interviews And Use Modeling Tools

Market sizing is the activity of determining how large or small your product's audience or revenue potential will be. Incorrectly assessing this dynamic could lead your ship into very problematic waters. A few ways to operate efficiently are using statistical modeling tools, conducting in-depth research, assessing competitive analysis, qualitative interviews and employing forecasting consultants. - John Drumgoole Jr , USA Mortgage

16. Build A Business Plan Around An Addressable Market

Building Total Addressable Market (TAM), Serviceable Addressable Market (SAM) or Serviceable Obtainable Market (SOM) models are the easiest way to identify and size the dollar value of a market before you build out a comprehensive business plan. You want to build a business plan around an addressable market that is large, has runway to grow and has fragmented competitors to share market from versus monopolies. - Archana Rao , Innova Solutions

17. Conduct A Broad View Analysis

I believe a top-down analysis or approach can be considered more efficient. Starting from the broader view of the market size through available secondary data sources (like governmental ones) would allow you to segment the consumer base in that specific market quickly and be much simpler to execute. It would also be beneficial for a high-level decision-making process. - Mohamed Madkour , Mastercard

18. Combine Data With Customer Conversations

The best way to measure market size is to mix hard data with real customer conversations. Market reports give you the numbers, but talking to your customers reveals the story behind them. This blend is vital because it helps you see where the real opportunities are, guide your strategy and connect with your audience in a meaningful way. - Richard Lindhorn , VivoAquatics Inc.

19. Gauge Needs Of Target Audience And Competitor Strategies

The simplest (and most cost-effective) method is to multiply the approximate price of the product by the total number of potential customers. It is used to achieve basic business goals: identifying the key needs of the target audience, recognizing the most successful strategies of competitors and learning from them, creating a unique selling point and seeing the potential for expanding the product line. - Dima Raketa , Reputation House

20. Analyze Industry Reports And Government Data

Accurately measuring market size is crucial for strategic decision-making and resource allocation. A comprehensive approach often combines analysis of industry reports and government data with primary research like surveys or interviews. The Total Addressable Market (TAM) framework is particularly useful, helping estimate full revenue potential. - Gunjan Paliwal

• Editorial Standards
• Reprints & Permissions

ArcGIS Blog

Aug 29, 2024

Use Business Analyst's Target Marketing Wizard to find customers in a new area

By Sarah Andrews and Darren Cook and Crystal Harper

Imagine that you own a successful mobile dog grooming business in Lancaster County, Nebraska. You have spent the last decade building up this profitable business and developing a loyal set of customers. You are ready to expand the business to the nearby Omaha, NE area.

In this article, we will demonstrate how to leverage your existing customer data to identify the population segments most likely to utilize your services and pinpoint where to find new customers within these segments.

What is Target Marketing?

Overall workflow.

• Create Segmentation Profiles
• Run the Target Marketing Wizard
• Examine the results

The ArcGIS Pro Business Analyst extension provides powerful tools and workflows that allow businesses to make better, more data-driven decisions. The Target Marketing Wizard is one of these tools. The Target Marketing Wizard analyzes the composition of your market area compared with your customers and allows you to see the types of people you are reaching, as well as people you should be targeting and where they can be found.

We will utilize Esri’s Tapestry Segmentation , which is a market segmentation system that classifies residential neighborhoods in the United States into unique market segments based on demographic and socioeconomic characteristics. It helps businesses and organizations understand their target audiences better by providing insights into the lifestyle preferences, behaviors, and demographic profiles of residents in different neighborhoods.

Read on to learn how to use the Target Marketing Wizard to identify and market to potential new customers.

Before we start our workflow, we need to do some setup and gather our input data.

Data source

The Target Marketing Wizard requires a locally installed Business Analyst data source that contains segmentation data. In this article, we are using the U.S. 2024 dataset .

Customer data

You will need a point layer of your current customers and their locations. This can be an existing layer that you have or can be created from an Excel file with location data. Learn about using Microsoft Excel files in ArcGIS Pro .

Our layer of current customers in Lancaster County is called LancasterCountyCusts.

Existing customer area

You will also need a polygon layer that represents the geographic extent of your customers. This can be an existing boundary layer that you have, can be created from standard geographies, or even created using the Minimum Bounding Geometry geoprocessing tool.

We are using a county boundary called LancasterCountyBoundary, which encompasses all customers in LancasterCountyCusts.

New market area

Finally, you will need a layer that represents the area into which you want to expand. It can be based on standard geographies (counties, zip codes, etc.) or any other polygon layer that you like.

Our layer covers Douglas and Sarpy Counties in Nebraska and is called DouglasSarpyCountiesBoundary. We created it using the Generate Standard Geography Trade Areas geoprocessing tool.

Our workflow consists of three high-level steps.

• First, we will create two segmentation profiles – one for our existing customers and one for the area in which those customers live. This will provide us with the segmentation information about which segments perform best for our business.
• Next, we will run the Target Marketing Wizard with a focus on the new market area into which we are looking to expand. The wizard allows us to focus on our successful segments and create map layers and reports for those segments in relation to our new market area.
• Finally, we will examine the map layers and reports that the wizard created and learn how to interpret them.

1. Create Segmentation Profiles

We need two segmentation profiles to perform the analysis. The customer profile tells us who are our customers by categorizing them by Tapestry segment. The market area profile tells us how many adults or households of that Tapestry segment are in the areas we are looking at. The Target Marketing Wizard puts these two profiles together to determine the best areas to find new customers.

Create a Customer Segmentation Profile

A Customer Segmentation Profile is a distribution of your customers classified into socioeconomic groupings.

Each customer is assigned to a standard geography (block groups in the U.S.), and each of those geographies is assigned to the Tapestry segment with the highest population in that area. This information is used to determine to which Tapestry segment each customer belongs. The customer profile consists of a list of Tapestry segments along with the number of customers in each and percentage relative to total customers. Learn more about Customer profiles .

To create a Customer Profile, you will need a layer that contains your existing customers as point data.

First, open the Generate Customer Segmentation Profile geoprocessing tool. Set the following parameters.

• Customer Layer: your customer layer (our is LancasterCountyCusts)
• Segmentation Base: Total Households
• Output Profile: leave default
• Volume Info Field: leave blank

A Customer Profile is created and stored in your project folder. To view this profile, go to the Catalog pane and open the Business Analyst folder and then the Target Marketing folder. Right click on CustomerProfile.sgprofile. The Segmentation Profile viewer opens.

Here you can see a list of all the Tapestry segments in the customer base, along with the counts and percentages of your customers who live in a block group assigned to that segment. We can see that nearly a quarter of our customers are in the 6A (Green Acres) segment.

Create a Market Area Segmentation Profile

A Market Area Segmentation Profile categorizes the population in a geographic area. Population is categorized at the smallest demographic level supported by your dataset (block groups in the U.S.). Each block group is assigned a dominant Tapestry segment based on population or household count.

Because we are working with two geographic areas in this workflow (Lancaster County, where our customers are found and Omaha, NE, where we want to expand our business), we will need to create the Lancaster County market area segmentation in advance of running the wizard. Learn more about Market areas .

To create a Market Area Segmentation Profile, open the Generate Market Area Segmentation Profile geoprocessing tool. Set the following parameters.

• Input features: a polygon layer that contains your customers (our layer is called LancasterCountyBoundary)
• Segmentation Base: Total Households (this should be the same as we used for the Customer Profile creation)

A Market Area Segmentation Profile is created and stored in your project folder. To view this profile, go to the Catalog pane and open the Business Analyst folder and then the Target Marketing folder. Right click on MarketAreaProfile.sgprofile. The Segmentation Profile viewer opens.

Here you can see a list of all the Tapestry segments, along with the counts and percentages of all households in the area (Lancaster County). Later in the workflow, this will be compared with the values in the Customer Segmentation Profile previously created to identify which Tapestry segments to target.

2. Run the Target Marketing Wizard

Now that we have our segmentation profiles are created, we can run the Target Marketing Wizard to understand which specific areas to target for our new location.

From the Analysis ribbon, click Business Analysis and then click Target Marketing Wizard.

The wizard opens.

Wizard Step One

This step allows us to specify the types of customers that use our services. We will do that using the Customer Segmentation Profile that we created earlier.

Set the parameters as follows.

• Customer Layer: leave blank (since we pre-created our Customer Segmentation Profile, we do not need the customer layer here)
• Target Profile: select the customer profile we created earlier (CustomerProfile.sgprofile)
• Leave all other fields as they are

Click Next.

Wizard Step Two

In this step, we specify both the area into which we want to expand and the market area segmentation profile for Lancaster County, which we created earlier.

The Market Area section is where we define our target area for siting a new location.

The Base Profile parameter will be used along with the Customer Segmentation Profile chosen in Step One to determine the Tapestry segments on which we want to focus.

• Market Area: click Use Polygonal Layer and choose your expansion area layer (DouglasSarpyCountiesBoundary)
• Base Profile: select the market area profile we created earlier (MarketAreaProfile.sgprofile)

Wizard Step Three

Targets are collections of one or more segments that are grouped together based on similar characteristics in your customer base. Sets of targets are consolidated into target groups. Target groups are used to produce powerful maps and reports that help you identify and locate new customers.

The Targets box groups each Tapestry segment by one of four designated targets: Core, Developmental, Niche and Monitor. These targets are based on the percent composition of your customers in each segment and how that compares to the percent of households in that segment in the overall area. The comparison value is stored as an index, with 100 being the average.

We can see that our Core customers come from segments 1D (Savvy Suburbanites), 4A (Workday Drive) and 6A (Green Acres). It seems these types of customers are the most interested in mobile dog grooming services.

Core customers contain segments that make up 4 percent or more of your customers and index at 110 or greater. And index of 110 means that your customer base in this segment is 10% higher than what would be expected based on the general market. These are the customers we want to find in our new area.

Developmental, Niche and Monitor customers score less than Core customers in percent composition or index or both. Learn more about Target groups .

This is visualized in detail in the Four Quadrant Analysis chart in Figure x. This chart displays segments according to index and percentage values, as described above.

For example, the highlighted segment in the image falls into the Core group as it contains 7% of customers and an index of 121 (this segment is using our services 21% more than the overall population).

It is possible to move segments to different targets, but we will leave the defaults for this exercise.

Learn about Four Quadrant Analysis .

Wizard Step Four

The wizard will use the inputs we have supplied along with the four-quadrant analysis to help us find potential new customers in our expansion area and will generate map layers and reports to assist us in this.

In this step, we choose which maps layers we want the wizard to create for us. We are interested in identifying in which smaller areas in our expansion area we will find the most potential new customers. The map layers that will help us are Target Layer and Market Potential. These are explained later in “Examine the results”. Learn about Target marketing map layers .

Set the following parameters.

• Geography Level: US.BlockGroups
• Mapping Layers: select only Target Layer, Market Potential
• Output Location: leave the default

Wizard Step Five

In this step, we choose which reports we want the wizard to create for us. We are interested in learning more about the areas identified in the Market Potential map layer, we will choose the Developmental Marketing Strategies and Market Potential reports. Learn about Target marketing reports .

• Available Reports: select Developmental Marketing Strategies and Market Potential
• Export Format: select Interactive HTML

Wizard Step Six

Review and click Finish.

3. Examine the results

Map layer: target layer (core).

This layer shows where potential customers in segments classified as Core can be found. Based on the results from the Four Quadrant Analysis, our core segments are 1D, 4A and 6A.

The map shows all block groups where the dominant segment (most households) is in our Core Target.

Map Layer: Market Potential

This layer shows the potential number of customers in each block group. Darker colors represent more potential customers. The values are calculated based on the percentage of customers we have in each segment in our original area and the number of households in each segment in our expansion area.

Report: Developing Marketing Strategies

The requested reports are generated as interactive HTML infographics.

The Developing Marketing Strategies report provides an overview of the media that your target customers read, listen to, and watch. It helps guide decisions about how to reach your target segment groups based on the media they consume.

Report: Market Potential Report

The Market Potential Report measures the likely demand for a product or service for your market area at a specific geography level. You can use this report to make decisions about where to offer products and services.

Hover over the different components to see more information and options.

In this article, we walked through using Business Analyst’s Target Marketing Wizard to discover new areas with potential customers similar to our best current customers. The output of the wizard includes powerful maps and reports that help us to find and market to prospects in a new area.

We now have a map (Target Layer (Core)) that shows us where in the Omaha area we should focus our marketing efforts for our business expansion. The Market Potential map tells us how many new customers we can expect to gain in each area – this helps narrow our focus.

The Market Potential report summarizes the data in the map layers so we can understand how many total potential customers are in the Omaha area.

Once we’ve selected our target areas, the Developing Market Strategies helps us decide how to reach these potential customers based on their preferences and habits.

Share this article

Article Discussion:

Related articles.

Out with the old, in with the bold: The new Target Marketing reports are here!

Use the Huff model to affirm site selection in Business Analyst Pro

Esri Data 101: Tapestry Segmentation

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 02 September 2024

Deep learning-based virtual staining, segmentation, and classification in label-free photoacoustic histology of human specimens

• Chiho Yoon   ORCID: orcid.org/0000-0003-3971-3115 1   na1 ,
• Eunwoo Park   ORCID: orcid.org/0000-0001-7954-7911 1   na1 ,
• Sampa Misra 1   na1 ,
• Jin Young Kim 1 , 2 ,
• Jin Woo Baik 1 ,
• Kwang Gi Kim 3 ,
• Chan Kwon Jung   ORCID: orcid.org/0000-0001-6843-3708 4 , 5 &
• Chulhong Kim   ORCID: orcid.org/0000-0001-7249-1257 1 , 2  

Light: Science & Applications volume  13 , Article number:  226 ( 2024 ) Cite this article

8 Altmetric

Metrics details

• Photoacoustics

In pathological diagnostics, histological images highlight the oncological features of excised specimens, but they require laborious and costly staining procedures. Despite recent innovations in label-free microscopy that simplify complex staining procedures, technical limitations and inadequate histological visualization are still problems in clinical settings. Here, we demonstrate an interconnected deep learning (DL)-based framework for performing automated virtual staining, segmentation, and classification in label-free photoacoustic histology (PAH) of human specimens. The framework comprises three components: (1) an explainable contrastive unpaired translation (E-CUT) method for virtual H&E (VHE) staining, (2) an U-net architecture for feature segmentation, and (3) a DL-based stepwise feature fusion method (StepFF) for classification. The framework demonstrates promising performance at each step of its application to human liver cancers. In virtual staining, the E-CUT preserves the morphological aspects of the cell nucleus and cytoplasm, making VHE images highly similar to real H&E ones. In segmentation, various features (e.g., the cell area, number of cells, and the distance between cell nuclei) have been successfully segmented in VHE images. Finally, by using deep feature vectors from PAH, VHE, and segmented images, StepFF has achieved a 98.00% classification accuracy, compared to the 94.80% accuracy of conventional PAH classification. In particular, StepFF’s classification reached a sensitivity of 100% based on the evaluation of three pathologists, demonstrating its applicability in real clinical settings. This series of DL methods for label-free PAH has great potential as a practical clinical strategy for digital pathology.

Similar content being viewed by others

NuInsSeg: A fully annotated dataset for nuclei instance segmentation in H&E-stained histological images

A generalized deep learning framework for whole-slide image segmentation and analysis

Deep learning-enabled virtual histological staining of biological samples

Introduction.

Histopathology, the microscopic imaging of specimens, is the primary source of diagnostic information for optimal surgical management. Histopathology and life-science research use chromatic dyes or fluorescence markers for histochemical staining to visualize tissue and cellular structures 1 , 2 . In particular, hematoxylin and eosin (H&E) staining is the gold standard for microscopic tissue examination in histopathology 3 . However, traditional slide preparation for staining is labor-intensive and error-prone, which presents a dilemma 4 . As the number of items requiring pathological examination increases, additional slides must be produced for staining, but an insufficient sample quantity may cause an inappropriate diagnosis.

Recently, many optical microscopic techniques have been utilized to address the issues of sample preparation and staining quality 5 , 6 , 7 , 8 , 9 . For example, light-sheet microscopy 10 , 11 rapidly images large specimens with intrinsic optical sectioning, but it typically involves additional chemical procedures such as optical clearing and fluorescence dyeing. As label-free imaging modalities, bright-field microscopy (BF) 12 , optical coherence tomography (OCT) 13 , and autofluorescence microscopy (AF) 6 , 14 provide histopathological images with a simplified sample preparation without staining. However, these methods are less able than H&E staining to identify specific biomolecules and have difficulty providing sufficient clinical information. Raman microscopy 15 , 16 and spectroscopic OCT 17 resolve unlabeled biochemical composition with spectral analysis, but have relatively weak signal sensitivity. On the other hand, novel label-free imaging methods have been proposed that can acquire selective images by targeting specific excitation wavelengths. Deep-ultraviolet microscopy (DUV) 18 , photoacoustic microscopy (PAM) 19 , 20 , 21 , 22 , and photoacoustic remote sensing (PARS) 23 use endogenous contrasts to visualize individual chromophores. Among these modalities, PAM is a promising high-sensitivity imaging technology that selectively highlights biomolecules based on optical absorption 19 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 . In particular, DNA/RNA highly absorbs ultraviolet (UV) light, allowing UV-PAM to visualize cell nuclei without staining 32 , and thus this technique has been intensively explored as a label-free histological tool (i.e., photoacoustic histology (PAH)) 33 , 34 . In clinical applications, however, label-free PAH techniques are still challenging to provide color-coded high-resolution histopathological images comparable to familiar H&E-stained whole slide images (WSIs). To solve this challenge, unlabeled images need to be translated into interpretable images that contain sufficient information for clinical diagnosis.

The development of deep learning (DL)-based image processing, including virtual staining and histological image analysis (HIA), has greatly expanded the clinical utilization of label-free images 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 . First, virtual staining allows images obtained from label-free microscopy to mimic the morphological characteristics revealed by various histochemical staining styles. However, traditional DL methods have used supervised learning algorithms that require image pairs 43 , 44 , 45 , 46 , 47 , 48 , which involve a difficult image registration process during data pre-processing. As an alternative, researchers have employed unsupervised image transformation methods such as cycle-consistent generative adversarial networks (CycleGAN) 49 , which are sufficient for network training even with unpaired datasets. By incorporating CycleGAN 50 , 51 , 52 , 53 , 54 , 55 , label-free microscopy can generate virtually stained histological images. However, when images in one domain contain more information than images in the other domain, the cycle consistency in CycleGAN can yield poor reconstruction, and using two generators and two discriminators is memory-intensive and time-consuming. Pérez et al 56 . presented a virtual staining method using contrastive unpaired translation (CUT) 57 by maximizing the mutual information between generated and input images. The CUT model uses patch-wise contrastive learning to achieve better virtual staining quality and is lighter and faster than CycleGAN. However, for application to safety-critical medical data analysis, the black box problem, where the process of deriving conclusions by DL is unknown, requires further investigation. Secondly, DL-based HIA, another DL-based image processing method, is a crucial step in the early stages of histological image diagnosis. DL-based HIA uses an automated analysis system that enables objective evaluation and reduces the cost of diagnosis. Numerous DL-based HIA tasks have been proposed, including image classification 58 , 59 , object or lesion identification 60 , and nuclei segmentation 61 . However, most DL-based HIAs primarily rely on conventional histopathological images, and they often are incompatible with label-free images. Even when they do work with label-free images, they are typically designed for a single HIA task 45 , 50 , 52 , 62 . There is a clear need for a DL-based HIA that is compatible with label-free images and can provide sensitive and accurate analysis results.

In this paper, we develop a DL-based framework for automated HIA that performs virtual staining, segmentation, and classification in label-free PAH images of liver cancer (Fig. 1 ). PAH images can reveal histological characteristics, but they also present pathologists with relatively less familiar images that can make diagnosis challenging. As a first step, an explainable-CUT (E-CUT) approach is proposed for virtual staining to transform grayscale PAH images to virtual H&E-stained images (Fig. 1a ). For virtual staining, E-CUT uses saliency loss and integrated gradients to not only preserve image content but also visualize saliency information and feature attribution to increase traceability. Next, it performs image segmentation and feature extraction to extract information for further analysis (Fig. 1b ). Features such as the cell area, cell count, and distance between cells are extracted from the segmented images. Finally, a DL-based classification model using a stepwise feature fusion method (StepFF) is proposed, using the combined PAH, VHE, and segmentation deep feature vectors (DFVs) to classify noncancerous and cancerous liver cells. The performance of StepFF is compared with that of traditional H&E (Fig. 1c ). This multi-step DL-based framework not only transforms PAH images to H&E-style ones for clinical applicability but also enables accurate analysis by fusing multiple DFVs. The innovative single framework for virtual staining, segmentation, and classification unfolds several insights as follows: (1) label-free histological images are obtained using UV-PAM with high sensitivity. (2) The explainable DL-based unsupervised virtual staining technique (E-CUT) is devised, which can highlight histologically significant morphologies in the input data and input features that significantly contribute to discriminator prediction. (3) Biological features are clearly extracted from the VHE images using a U-Net-based segmentation technique. (4) The DL-based stepwise feature fusion method (StepFF) is presented by combining the deep feature vectors, which pathologically classifies human hepatocellular carcinoma (HCC) images. The superiority of the proposed system is confirmed by comparing with reported DL-based label-free histology techniques (Table S1 ).

a Virtual staining sequence with explainability to generate VHE images with label-free PAH. PAH photoacoustic histology image; and VHE virtual staining H&E. b Segmentation sequence to generate features: cell area, cell count, and distance. c Classification sequence to classify cancerous or not with PAH, VHE, and segmentation deep feature vectors

Label-free PAH system

PAH images of human liver samples were obtained using a previously developed UV-PAM system 63 . For label-free DNA/RNA-selective imaging, the PAH system employed a pulsed UV laser with a center wavelength of 266 nm and a pulse repetition rate of 20 kHz 64 , 65 (Fig. 2a ). The zoomed-in image in Fig. 2a is a detailed schematic of the PA signal acquisition module. A formalin-fixed paraffin-embedded (FFPE) tissue section was fixed to the tissue holder, and the laser and acoustic beams were simultaneously scanned with a MEMS mirror. After passing through an opto-ultrasound combiner with an acoustic lens, acoustic waves were collected by an ultrasound transducer with a center frequency of 20 MHz. The imaging system has a lateral resolution of ~1.2 μm 63 , and it takes ~35 s to image a field of view of 700 × 1000 μm 2 (i.e., one piece of the PA image), with a step size of 1.0-micrometer per pixel. Figure 2b shows a PAH maximum amplitude projection (MAP) image of the human liver tissue section. A PA whole slide mosaic image was generated by stitching together 123 pieces (with a total area of 10.5 × 8.0 mm 2 ), according to the scanning geometry of the motorized XY stage. The corresponding H&E WSI was obtained by imaging a slice adjacent to the slice used for the PAH image (Fig. 2c ). Compared to the images from the noncancerous region (Fig. 2d ), higher cell densities and larger cell nuclei can be identified in the zoomed-in PAH images acquired in the cancerous region (Fig. 2e ). The PAH images are highly correlated with the traditional H&E images (Fig. 2f, g ). Detailed quantitative analyses will be discussed in the following sections.

a Schematic of the PAH system and the close-up of the signal acquisition module. PD photodiode, OBS optical beam splitter, NDF neutral density filter, BE beam expander, M Mirror, OBJ objective lens, OUC opto-ultrasound beam combiner, DAQ data acquisition, UST ultrasound transducer, AL acoustic lens, and TH tissue holder. b PAH image of human liver tissue. c Corresponding H&E-stained image. Scale bars, 500 μm. H&E, hematoxylin and eosin stained image. d , e Zoomed-in PAH images of noncancerous and cancerous regions, respectively. f , g Zoomed-in H&E images of noncancerous and cancerous regions, respectively. Scale bars, 100 μm

Explainable contrastive unpaired translation (E-CUT) VHE network

After acquiring grayscale PAH images of the human liver tissue section, the proposed unsupervised DL method, E-CUT, was implemented for virtual staining. E-CUT is based on patch-wise contrastive learning and incorporates additional explainable components such as saliency loss and integrated gradients (Fig. 3a ). Saliency loss continuously tracks both saliency masks from PAH and VHE to assist in resolving singularity issues that may arise during the training phase (Fig. S 1 ). The saliency mask improves explainability by highlighting the important morphology of input data in the virtual staining process and visualizing the model’s ability to preserve structural information 66 , 67 , 68 , 69 . Another one of the explainable components, the integrated gradients, can highlight the most influential features and increase the explainability of the model 70 . The discriminator’s integrated gradients allow the identification of features that are important in the process of determining whether the image generated by the generator is real or fake. Note in Fig. S 1 that as training progresses, the previously randomly emphasized integrated gradients gradually focus on input features around cell nuclear information, which is the information of interest in pathological virtual staining. Subsequently, the virtual staining results of E-CUT were compared with other unsupervised DL methods, such as CycleGAN 49 , explainable CycleGAN 52 (E-CycleGAN), and CUT 57 (Fig. 3 b, c ). The CycleGAN model has a cyclic (bidirectional) structure consisting of two generators and two discriminators that can learn to transform to another domain while preserving the content of the input image. However, CycleGAN is still limited in preserving detailed structural information, so E-CycleGAN incorporates additional explainable components to preserve the more precise structure and increase the explainability of the model (Fig. S 2 ). On the other hand, CUT employs patch-wise contrastive learning, offering better virtual staining results than CycleGAN, with the added advantage of being lighter, i.e., using relatively fewer generators and discriminators. The proposed E-CUT also incorporates additional explainable components in the CUT, which enable better preservation of structural information and increase the explainability of the model.

a Explainable contrastive unpaired translation (E-CUT) network architecture. b Visual comparison of PAH (input), and VHE results with various networks: CycleGAN, explainable CycleGAN (E-CylcleGAN), CUT, E-CUT, and H&E (ground truth, GT). Scale bars, 100 μm. Zoomed-in images scale bars, 50 μm. Black arrows highlight cell nuclei, showing the difference in morphology preservation between different networks. c Quantitative comparison results for different VHE networks with FID and KID scores. Results are evaluated on a total of 100 test tiles. FID Fréchet Inception Distance, KID Kernel Inception Distance

We compared the performance of the four virtual staining models and validated their results against the original H&E images. Figure 3b shows the original PAH (input) and H&E (ground truth) images of the human liver tissue section, and the corresponding VHE results processed with the four virtual staining models for each noncancerous and cancerous case. For a detailed analysis of the staining results, each image is zoomed-in (dotted boxes). Especially in the areas indicated by the black arrows, the VHE results of CycleGAN, E-CycleGAN, CUT, and E-CUT, in that order, show better preservation of the cell nucleus morphology, with staining similar to that in the real H&E image. In particular, the structural aspects of the cell nuclear information in the input PAH images are well preserved in the E-CycleGAN and E-CUT with saliency loss, whereas the overall staining quality (e.g., degree of color and morphology reproduction) is improved in CUT and E-CUT with PatchNCE loss 57 . As a result, E-CUT demonstrates a remarkable capability to effectively follow the morphology of PAH images, thereby contributing significantly to the overall staining quality. Hematoxylin staining is more prominent following the selective visualization of cell nuclei in the PAH images. Including the eosin staining for RBCs, collagen, and smooth muscle, the proposed VHE yields diagnostic histological images. The enhanced staining quality makes the E-CUT results more closely resemble real H&E staining. However, it is important to note that the VHE does not perfectly match the morphology obtained from H&E. As discussed in the previous section, we used two adjacent slides, and the ground truth H&E image is not perfectly registered to the PAH image.

For a quantitative comparison, we applied the Fréchet Inception Distance (FID) 71 and the Kernel Inception Distance (KID) 72 (Fig. 3c ). Both the FID and KID metrics evaluate the performance of the image generation model, calculating the difference between the generated image and the real one. The obtained lower values for both metrics indicate that the distributions of the two data are closer, which implies the staining quality is closer to the ground truth. This finding confirms that E-CUT outperforms than above-mentioned existing models in terms of FID by a large margin (~4 to 17 difference). E-CUT also achieves the lowest KID (0.2451) on the PAH to VHE translation. Notably, E-CycleGAN and E-CUT, which utilize saliency masks, exhibit superior performance to conventional CycleGAN and CUT. Similarly, CUT and E-CUT, employing contrastive learning, outperform CycleGAN and E-CycleGAN. E-CUT has greater stability and dependability because the saliency loss ensures that the extracted saliency mask of the input PAH image stays consistent when transferred to the H&E domain, thus achieving the best FID and KID scores among all virtual staining methods. Since the FID and KID scores comprehensively evaluate color, texture, and structure, these results also indicate that E-CUT is good at reproducing the original color and morphology.

U-Net-based feature segmentation network

The feature segmentation network, illustrated in Fig. 4a , has two primary components: a segmentation module that acquires a segmented image (i.e., cell nucleus mask information) from the input images, and a feature extraction module that extracts structural information about cell nuclei from the segmented images. The segmentation module is based on the fundamental U-Net architecture 73 , comprising a model with contraction and expansion paths, each consisting of four layers. For versatility, the segmentation model was trained and evaluated with a public H&E dataset 74 , 75 , 76 and the results are presented in Table S2 . This trained segmentation model was used to create cell nucleus segmented images from PAH, VHE, and H&E datasets. Afterward, in the feature extraction module, the cell segmented images were analyzed using OpenCV tools 77 (e.g., findContours and minEnclosingCircle) to extract the cell area, cell count, and average intercellular distance. These features are clinically representative of differences between noncancerous and cancerous tissues.

a U-Net-based feature segmentation network architecture with two phases: segmentation and feature extraction. b Segmentation of cell nuclei in PAH, VHE, and H&E images. Scale bars, 100 μm. c 3D scatter plot of each feature in the PAH, VHE, H&E images. The blue and red dots represent noncancerous and cancerous cases, respectively. d Cell area, cell count, and distance features are calculated from the PAH, VHE, and H&E images

Figure 4b shows examples of PAH, VHE, and H&E images and the segmented images for each. In all imaging modalities, the cell nucleus density is significantly higher in cancerous tissues than in noncancerous tissues. For more detailed analyses, the segmented features for PAH, VHE, and H&E were extracted from the segmented images (Table S3 ) and visualized in 3D scatter plots (Fig. 4c ). We marked the averaged values for features in each test tile image. To remove outliers and visualize the correlation between major features, the interquartile range (IQR) was employed 78 . In common, the cancerous tissues (red dots) show higher cell counts, shorter intercellular distances, and higher densities than the noncancerous tissues (blue dots). However, in the PAH images, the noncancerous and cancerous features overlap considerably, making the distinction unclear. In contrast, the VHE and H&E images clearly separate these features. Figure 4d shows the detailed distribution of segmentation features (i.e., cell area, cell count, and distance). Due to its limited resolution and contrast, the PAH image has a greater cell area than the VHE and H&E images. Moreover, the trend of the features distinguishing between cancerous and noncancerous cells in the PAH image appears to be slightly reversed, which is addressed by the VHE. In the cell count and intercellular distance comparisons, both the PAH and VHE images tend to be similar to the H&E images, with a higher cell count and closer distance for the cancerous case. Based on the H&E, the PAH and VHE images show error rates of 14.51% and 6.74%, respectively (Table S3 ), which suggests that the VHE image has similar features to the ground truth H&E. Overall, the segmentation results imply that effective histopathological analysis is possible, because the limitations of images and features in the PAH are addressed by the VHE. However, considering the distribution without IQR (Fig. S 3 ), it is important to note that the segmentation step alone is not enough to effectively classify the type of tissues, and additional steps are required for accurate diagnosis.

Stepwise feature fusion classification network (StepFF)

Although virtual staining in label-free imaging produces VHE images that closely resemble real H&E images, image quality limitations of the source PAH image limit the ability of traditional DL-based HIA techniques to interpret features on VHE. To complement the image information and more accurately classify cancer, we propose StepFF, which integrates multiple DFVs generated in each step (Fig. 5a ). Three DFVs are used for the cancer classification: PAH DFV, VHE DFV, and segmentation DFV. The deep feature extractions from the PAH and VHE images are performed using the ResNet 79 model, which has been identified as the most suitable classification model for VHE image analysis among such well-known CNN models as EfficientNet 80 , Inception 81 , VGGNet 82 , SwinNet 83 , and ResNet 79 (Table S4 ). First, in the deep feature extraction step, the PAH image is utilized to extract 512-dimensional DFV (the default dimension of the output DFV in the basic ResNet). The VHE image obtained in the virtual staining process (Step 1) is utilized to extract 512-dimensional DFV. Subsequently, in the feature segmentation (Step 2), three biological features (i.e., cell area, cell count, and distance), which are basic indicators of clinical evaluation, were extracted from PAH and VHE. Segmentation DFV were optimized by comparing classification scores according to feature combinations (Table S5 ). In this experiment, the best results were obtained when all the segmental features of PAH and VHE were used together, so a 6-dimensional DFV with all the segmental features of PAH and VHE is used as the segmented features for the final classification (Table S6 ). In the final step, all features are changed into 16-dimensional DFVs with fully connected layers to merge them fairly, then fused to classify the cancerous tissue.

a Overall stepwise feature fusion (StepFF) classification network architecture. b Visualization of DL classification results for different source feature inputs. c Cancerous probability outputs of StepFF (0 for noncancerous, 1 for cancerous), and the ground truth. d Comparison of StepFF and pathologists’ classification results

Figure 5b shows a visualization of representative cross-validated DL classification results (accuracy and f1 score), and the details are in Table S6 (accuracy, f1 score, precision, and recall). Among the single-modal results, the classification performance using the VHE (accuracy of 95.60%) is better than that of the PAH (accuracy of 94.80%) as it contains more information (three-dimensional color information). The multi-modal results are better than the single-modal results because they use multiple images or additional segmentation DFV. While VHE shows little performance improvement with the segmentation DFV, PAH performs better. This difference confirms that adding segmentation DFV is effective for the less informative PAH. Also noteworthy is the significant improvement in the classification results for PAH ⊕ VHE (accuracy of 97.20%), a 2.4% improvement over PAH alone, and a 1.6% improvement over VHE alone. Here, combining the undistorted original data (PAH) with the generated virtual image (VHE), which has additional color information, enables more accurate classification. With a combination of PAH, VHE, and segmentation DFVs, the proposed StepFF model achieves the best classification accuracy of 98% and precision of 98.14%, which is comparable to that obtained on H&E images (98.20% accuracy and 97.24% precision).

We also checked the probability of cancer at the level of the entire WSI. The cancerous probability of each tile was color-mapped and then re-stitched to the full slide image size to visualize the cancerous probability (Fig. 5c ). In this experiment, we first used all the dataset tiles (training and test) to find the probability of cancer per tile (1 for cancerous, 0 for noncancerous). We then colored each tile according to the cancer probability, 1 for purple and 0 for blue, and merged the small tiles to the original WSI size while keeping the cancer probability information. For detailed histopathological evaluation, three pathologists quantitatively graded the tile images blindly. A total of 200 test tile images (100 VHE images and 100 corresponding H&E images) were randomly shuffled and graded using the World Health Organization’s histological grading system for HCC 84 . All images distinguished malignancy clearly and morphologically differentiated HCC. The classification used two categories, benign (noncancerous) and malignant (cancerous), according to the presence of HCC. Of the 100 tiles in each of the VHE and H&E test sets, five tiles were excluded if each image had less than 20% tissue surface or only stromal cells (Table S7 ). The Kappa coefficient was 0.979, indicating almost perfect agreement between the pathologists’ responses 85 . Comparing the classification results between StepFF and the pathologists’ grading, StepFF showed a strong correlation with the real H&E grades (Fig. 5d ). Among the 95 tiles, both gradings equally classified 43 benign and 52 malignant tiles. The results show that while classification is still difficult with VHE alone, StepFF can utilize the DFVs of multiple images and segmented features to make judgments that are nearly identical to the pathologists’ assessment based on the real H&E.

Although histological imaging is a routine tool for pathological diagnostics, traditional histochemical staining is laborious and error-prone. To address this, label-free imaging and DL-based HIA have been exploited to highlight oncological features according to intrinsic imaging contrasts. However, conventional single-modal techniques are still insufficient in clinical settings. In this study, we introduce a DL-based framework for automated HIA in label-free PAH. The proposed multi-modal method has three steps: virtual staining, segmentation, and classification (Note S1). First, we present a fast and accurate virtual staining method, E-CUT, that combines contrastive unpaired image transformation and explainable components. E-CUT maximizes the mutual information between the generated (VHE) images and input (PAH) images using only a single generator and discriminator. In addition, the saliency loss and integrated gradients increase explainability, providing improved similarity and traceability during the transformation between the image domains. E-CUT can learn accurate domain mappings and achieve superior performance to traditional virtual staining methods. Second, we demonstrate segmentation for morphological feature extraction. This segmentation analysis provides quantitative metrics for diagnosis from PAH, VHE, and H&E images. The segmentation in VHE shows a distinct distribution of cancerous characteristics that is very similar to that of H&E. Interestingly, in PAH, the noncancerous cell area tends to be larger than the cancerous cell area, as opposed to the VHE and H&E images. Because it was trained with public H&E 74 , 75 , 76 , the segmentation model does not work well with other styles, such as PAH. In the case of cancerous cells with high cell density, it is especially challenging to accurately segment all the cell nuclei. Therefore, the measured cell area in PAH images of cancer tends to be smaller than the actual cell area. For similar reasons, the cell count in cancerous PAH images is underestimated, and the intercellular distance is greater. For the final classification step, we introduce a multi-modal classification method termed StepFF, which uses PAH, VHE, and biological features together for better performance. While scarce information limits the classification performance of single-modal HIA, by integrating DFVs of each step, StepFF achieves remarkable classification performance. Notably, the StepFF’s classification results obtained after excluding the five unsuitable tiles (Table S7 ) show 100% sensitivity to pathologists’ evaluation, demonstrating that the StepFF performs very well in general cases.

While the proposed framework transforms label-free images to H&E-style and provides diagnostic insight, improvements are needed for better analysis. Within the virtual staining stage, the additional employment of a resolution enhancement network can provide much clearer VHE images, allowing further diagnosis by differentiating nuclear atypia in HCC. In the segmentation process, we use a segmentation model trained with public H&E datasets, which limits the segmentation performance. To improve the segmentation performance and enable further evaluation, we plan to obtain the annotation of liver PAH and H&E images. Furthermore, additional transfer learning and data augmentation can address result bias and performance limitations. In particular, the public H&E images can be used as additional training data for our proposed method by generating virtual PAH images. Finally, additional techniques for obtaining attribution maps (such as GradCAM 86 , RISE 87 , Extremal perturbations 88 , etc.) can be used to improve the explainability of the system.

In conclusion, we present a novel, explainable, interconnected DL-based framework for virtual staining, segmentation, and classification of label-free PAH images. This interconnected approach executes the three tasks simultaneously and shares outputs, resulting in improved diagnostic accuracy, time savings, and reduced sample consumption, which can be implemented in intraoperative digital pathology workflows for clinical applications. Furthermore, the multi-modal framework can be generalized across different types of cancer diagnoses and adapted to the digital histopathology of other label-free imaging modalities (e.g., AF, BF, and OCT). We expect the proposed approach to have a clinical impact as a primary histological diagnostic tool.

Data preparation

All histopathological procedures were conducted following regulations and guidelines approved by the Institutional Review Board of POSTECH (approval no. PIRB-2019-E013). For specimen preparation, we harvested human liver tissue with hepatocellular carcinoma, along with adjacent noncancerous tissue. The excised tissue was processed into FFPE blocks. The 10 μm-thick unstained deparaffinized FFPE tissue sections were prepared for PA imaging. PAH images were then obtained with a UV-PAM system that uses an ultraviolet (266 nm) laser for label-free imaging (Fig. 2a ) 63 . Corresponding H&E-stained images were also acquired at approximately the same location as the PAH image acquisition.

For segmentation, training the model requires data with nucleus contour label information for the PAH, H&E, and VHE images. However, obtaining such annotated data is inherently challenging, primarily due to its time-consuming and labor-intensive nature. Acquiring label information for the PAH and VHE images is especially difficult because there is little pathological knowledge to guide segmenting the nucleus contours. Therefore, instead of segmenting the nucleus contour information manually, we quickly trained and tested the model using public H&E datasets containing contour information. A total of four datasets were used to train the segmentation model: CPM-15 74 , CPM-17 74 , Kumar 75 , and TNBC 76 .

Image pre-processing and post-processing

In the pre-processing step for raw PAH images, we conducted contrast adjustment, denoising, and background erasing. For the subsequent DL training processes, WSIs of H&E slides were converted to the same size as PAH images and cropped into smaller image tiles. First, the ×20 H&E images were downsampled to match at the magnification of ×10, which is the scale of the PAH images. The PAH images were then inverted to match the background color of the downsampled H&E image (the background was set to white). For training and testing, WSIs of both H&E and PAH images were cropped into small image tiles of 512 × 512 pixels with 50% overlap. For both virtual staining and classification, these image tiles were divided into training and test sets in proportions of 5:1. On the other hand, for segmentation, we used publicly available datasets 74 , 75 , 76 and cropped them to 224 × 224 pixels for training. All the tiles of PAH, H&E, and VHE images were employed as test data for the segmentation model. A fivefold cross-validation was employed to validate the segmentation and classification results. For testing, we have used altogether different tiles, which had never been used during the training phase. We also organized the test dataset to balance between cancer and non-cancer cases for classification. Additionally, since PAH and H&E images have different numbers of channels and E-CUT requires the same number of channels for the input and ground truth, 1-channel grayscale PAH images were stacked and converted to three-channel PAH images.

The final post-processing step was to stitch the small image tiles to get the original WSI. We merged them considering the 50% overlap, so that the results were summed up, and the overlapping sections were divided by the number of overlapping images.

Explainable contrastive unpaired translation network

Network architectures and training.

We adopted a CUT 57 architecture for the E-CUT model to learn the unpaired image translation between label-free PAH images and corresponding histological images stained with H&E (Fig. 3a ). The generator network for E-CUT, inspired by the ResNet model, consists of downsampling, residual blocks, and up-sampling parts 89 . The downsampling process encodes an input image down to the 9 residual blocks. Each residual block is designed with a skip connection where an input to the block is concatenated to an output of the block, enabling interpretation of the encoding. In the residual block, a padded convolutional layer keeps the image size constant. The residual blocks are followed by up-sampling to decode the representation to match the size of a final output image. For the discriminator network, we utilized a PatchGAN classifier 90 . This patch-level discriminator can determine whether 70 × 70 overlapping patches are real or fake, and it can be used on images of any size in a fully convolutional fashion. The final output of the discriminator is defined as the average of the classification results on all patches.

The patch information from PAH (the input) is trained to transform it into the style of the H&E (the ground truth). In particular, during the training process, our model stores the saliency mask and integrated gradients attribution map at each training step to improve explainability. To obtain the integrated gradients, we approximate the gradient integral of the discriminator model output over the input along the path to compute the importance score for each input feature. In order to obtain the attribution map, a total of 50 steps of approximation are performed using Pytorch’s captum library 91 . Finally, the trained model is used to virtually stain the test data. The training data (PAH and H&E) and virtually stained data (VHE) are reserved for use in later stages (segmentation and classification).

Loss function

To ensure a reliable image translation between PAH images (the source, X ) and H&E images (the target, Y ), it is important to define a loss function. The goal of virtual staining is to transform the input data into the target’s style, color, and shape. However, at the same time, details such as information about cell nuclei should be preserved. Therefore, as the final training loss, we used an equal combination of the adversarial, PatchNCE, and saliency losses (Fig. 3a ).

Adversarial loss ( \({{\rm{l}}}_{{adv}}\) ) minimizes the differences between the output of each network and the target domain image 92 . Contrastive learning with PatchNCE loss ensures that learning proceeds in a way that maximizes the mutual information between the input and output image patches 57 , which are obtained by passing the input and output images through a generator encoder. PatchNCE loss is calculated as the average of \({{\rm{l}}}_{{PatchNCE}}(X)\) on images from domain \(X\) and \({{\rm{l}}}_{{PatchNCE}}(Y)\) on images from domain \(Y\) , where \({{\rm{l}}}_{{PatchNCE}}(X)\) ensures that the input-output patches correspond, and \({{\rm{l}}}_{{PatchNCE}}(Y)\) serves to further prevent the generator from making unnecessary changes.

The saliency loss ( \({{\rm{l}}}_{{Saliency}}(X,Y)\) ) is the L1 loss between \({{\rm{X}}}_{{saliency}}\) and \({{\rm{Y}}}_{{saliency}}\) , and it is employed to preserve similar saliency masks during transformation and to improve explainability by visualizing the saliency mask 52 . In addition to adversarial and PatchNCE losses, saliency loss is used to extract the saliency mask of the input and the generated output to check whether the structural information is well preserved during training and leads to more accurate results. The saliency losses for the source and target domains are obtained by these equations:

The detailed process of obtaining a saliency mask can be seen in Fig. S 4 . To consider only the saliency information, we first convert Y with RGB information to grayscale data by averaging the three-channel information, then apply sigmoid to both X and Y to get the saliency information. In the last step, the image is inversed to make the saliency information equal to 1. The optimal thresholds were obtained manually through experimentation, with 90 as the X threshold , which yields a good saliency mask for both noncancerous and cancerous cases of PAH, and 170 as the Y threshold , which works well for both noncancerous and cancerous cases of VHE (Fig. S 5 ).

Finally, the entire loss function was formulated as

Parameter setting and evaluation metrics

The Adam 93 optimizer, with b1 = 0.5 and b2 = 0.999, was used to optimize the E-CUT network parameters. The model was trained for 400 epochs, with an initial learning rate of 0.0002 for the first 200 epochs and a linear decay to a zero learning rate for the next 200 epochs, with 1 mini-batch setting. During the training phase, we augmented the data with horizontal flips. In terms of time consumption, E-CycleGAN took 73,800 s for the train and 60 s for the test, while E-CUT took relatively less time, ~62,400 s for the train and 33 s for the test. The overall test time of our proposed virtual staining (E-CUT) is much faster than the staining time of a human expert (20–30 minutes).

For a fair comparison, the same configurations were used for other virtual staining models, i.e., CycleGAN, E-CycleGAN, and CUT. We used FID and KID to evaluate the virtual staining quality of the unpaired resultant image tiles. Lower values for both metrics indicate that the distributions of the two data are closer, indicating better virtual staining quality.

To numerically represent the characteristics of the VHE, we segment the cell nuclei information through feature segmentation and present it as three features: cell area, cell count, and mean intercellular distance.

This study employs the most well-known segmentation model, U-Net, which consists of contraction and expansion paths 73 (Fig. 4a ). While expansion paths have several deconvolution layers to upsample data and produce pixel-wise segmentation, contraction paths use convolution layers to produce high-level features in downsampling. Additionally, skip connections restore the spatial information lost during the downsampling. For feature segmentation, we employed four downscaling and four upscaling layers.

In the training phase, we trained the model to segment cell nuclei information using a public dataset, and in the testing phase, we used the trained model to segment the cell nuclei information of PAH, VHE, and H&E images. Finally, the segmented cell nuclei information was analyzed using the OpenCV tool 77 (findContours, minEnclosingCircle) to obtain the cell area, cell count, and mean distance between cell nuclei for each tile. The cell area is the average value of the cell nuclei size, the cell count is the number of cell nuclei, and the mean intercellular distance is the average distance between cell nuclei.

The U-Net-based segmentation model used Adam optimizer with b1 = 0.9 and b2 = 0.999, a learning rate of 0.0001, and a mini-batch size of 64, and was trained for 300 epochs. We used a combination of binary cross entropy and dice losses and allowed the training to stop early, depending on the validation loss 94 . We also used horizontal and vertical flips in the training phase to augment the data. In the last test phase, test time augmentation (TTA) was applied to get more accurate segmentation information. To visualize the results, we used the data visualization package Plotly 95 to plot 3D spatial scatter plots and box plots (Fig. 4c, d ). The total training time for the 5-fold cross-validation of feature segmentation took ~1300 s, and the test took a total of 165 s.

Stepwise feature fusion classification network

In applying our StepFF method, we used the ResNet-18 model 79 as the base model for classification (Fig. 5a ). The 1-channel PAH image and the 3-channel VHE image were processed separately through ResNet-18 and processed into a fully connected layer to obtain 16-dimensional DFV from each. After that, to generate segmentation DFV from the previous feature segmentation results, three features (cell area, cell count, and distance) each from PAH and VHE were normalized with the mean and standard deviation of each segmented feature. Then, the same 16-dimensional DFV was generated using a fully connected layer with the normalized three biological features. Finally, we concatenated the three types of 16-dimensional DFVs into a 48-dimensional DFV and passed them through the last fully connected layer to classify the final non-cancer and cancer information. To compare the DL classification results with different source feature inputs (Table S6 ), we used a single ResNet for a single modality image (e.g., H&E, PAH, and VHE), and two ResNets for multiple modality images (e.g., PAH ⊕ VHE). After using ResNet to obtain DFVs, we used fully connected layers in the same way as StepFF to obtain the final cancer classification results.

For the classification, we used the Adam optimizer to train the model for 1000 epochs, with b1 = 0.9 and b2 = 0.999, with a learning rate of 0.0001 and a mini-batch size of 32. We used focal loss for the imbalance data and allowed the training to terminate early based on the validation loss. We also used horizontal and vertical flips in the training phase to augment the data. The classification results were evaluated in terms of their accuracy, F1 score, precision, and recall. The fivefold cross-validation for StepFF took a total of 6000 s of training time and 16 s of testing time.

For further analysis, we combined the results from each tile to create one large WSI cancerous probability map, which showed the overall cancerous classification results (Fig. 5c ). For this cancerous probability map, both training and test data tiles were used, in the following order. First, for each tile, StepFF’s cancerous prediction result was represented as a value between 0 and 1 and color-mapped (1 to purple, 0 to blue). Then, each tile was combined and reconstructed into the original WSI by using the post-processing method introduced in the image pre-processing & post-processing part.

Pathologists’ evaluation

To compare StepFF’s results with clinical diagnoses, we compared them with three pathologists’ evaluation (Fig. 5d ). We randomly shuffled 100 virtual staining results from StepFF together with a second group of 100 H&E images, then presented them all to three pathologists for evaluation at the same time. The pathologists were asked to determine whether each tile was noncancerous or cancerous according to the World Health Organization’s histological grading system for HCC. Five tiles with tissue coverages of 20% or less, which made the determination difficult, were excluded from the evaluation (Table S7 ). To measure the inter-pathologist agreement, we measured the kappa coefficient 85 , which is -1 for complete disagreement and 1 for complete agreement.

Implementation details

The image pre-processing steps were implemented in MATLAB using R2021a (The MathWorks Inc.). All the virtual staining, segmentation, and classification sequences were implemented using Python, version 3.8.12, and Pytorch, version 1.11.0. We implemented this training and testing on a Linux system with one Nvidia GeForce RTX 3090 GPU, an AMD EPYC 7302 CPU, and 346GB of RAM.

Data availability

The data that support the findings of this study are available on request from first author C.Y. and corresponding author, C.K. The data are not publicly available because it contains information that may violate the privacy of study participants. Supplementary information accompanies the manuscript on the Light: Science & Applications website ( http://www.nature.com/lsa ).

Code availability

The code is available at https://github.com/YoonChiHo/DL-based-framework-for-automated-HIA-of-label-free-PAH-images .

Bancroft, J. D. & Gamble, M. Theory and practice of histological techniques. 6th edn. (Philadelphia: Elsevier, 2008).

Musumeci, G. Past, present and future: overview on histology and histopathology. J. Histol. Histopathol. 1 , 5 (2014).

Article   Google Scholar  

Gurcan, M. N. et al. Histopathological image analysis: a review. IEEE Rev. Biomed. Eng. 2 , 147–171 (2009).

Zhu, E. J. et al. Intraoperatively measured tumor size and frozen section results should be considered jointly to predict the final pathology for lung adenocarcinoma. Mod. Pathol. 31 , 1391–1399 (2018).

Ounkomol, C. et al. Label-free prediction of three-dimensional fluorescence images from transmitted-light microscopy. Nat. Methods 15 , 917–920 (2018).

Zhang, Y. J. et al. Digital synthesis of histological stains using micro-structured and multiplexed virtual staining of label-free tissue. Light Sci. Appl. 9 , 78 (2020).

Article   ADS   Google Scholar  

Cheng, S. Y. et al. Single-cell cytometry via multiplexed fluorescence prediction by label-free reflectance microscopy. Sci. Adv. 7 , eabe0431 (2021).

Visscher, M. et al. Label-free analytic histology of carotid atherosclerosis by mid-infrared optoacoustic microscopy. Photoacoustics 26 , 100354 (2022).

Pleitez, M. A. et al. Label-free metabolic imaging by mid-infrared optoacoustic microscopy in living cells. Nat. Biotechnol. 38 , 293–296 (2020).

Glaser, A. K. et al. Light-sheet microscopy for slide-free non-destructive pathology of large clinical specimens. Nat. Biomed. Eng. 1 , 0084 (2017).

Olarte, O. E. et al. Light-sheet microscopy: a tutorial. Adv. Opt. Photonics 10 , 111–179 (2018).

Selinummi, J. et al. Bright field microscopy as an alternative to whole cell fluorescence in automated analysis of macrophage images. PLoS One 4 , e7497 (2009).

Bouma, B. E. et al. Optical coherence tomography. Nat. Rev. Methods Prim. 2 , 79 (2022).

You, S. X. et al. Intravital imaging by simultaneous label-free autofluorescence-multiharmonic microscopy. Nat. Commun. 9 , 2125 (2018).

Jones, R. R. et al. Raman techniques: fundamentals and frontiers. Nanoscale Res. Lett. 14 , 231 (2019).

Petersen, D. et al. Virtual staining of colon cancer tissue by label-free Raman micro-spectroscopy. Analyst 142 , 1207–1215 (2017).

Xu, C. Y. et al. Spectroscopic spectral-domain optical coherence microscopy. Opt. Lett. 31 , 1079–1081 (2006).

Ojaghi, A. et al. Label-free deep-UV microscopy detection and grading of neutropenia using a passive microfluidic device. Opt. Lett. 47 , 6005–6008 (2022).

Kim, J. et al. Super-resolution localization photoacoustic microscopy using intrinsic red blood cells as contrast absorbers. Light Sci. Appl. 8 , 103 (2019).

Lee, H. et al. Transportable multispectral optical-resolution photoacoustic microscopy using stimulated Raman scattering spectrum. IEEE Trans. Instrum. Meas. 7 , 4502309 (2024).

Google Scholar  

Park, E. et al. Effective photoacoustic absorption spectrum for collagen-based tissue imaging. J. Biomed. Opt. 25 , 056002 (2020).

Kim, D. et al. In vivo quantitative photoacoustic monitoring of corticosteroid-induced vasoconstriction. J. Biomed. Opt. 28 , 082805 (2023).

Martell, M. T. et al. Deep learning-enabled realistic virtual histology with ultraviolet photoacoustic remote sensing microscopy. Nat. Commun. 14 , 5967 (2023).

Park, B. et al. Functional photoacoustic imaging: from nano-and micro-to macro-scale. Nano Converg. 10 , 29 (2023).

Ahn, J. et al. In vivo photoacoustic monitoring of vasoconstriction induced by acute hyperglycemia. Photoacoustics 30 , 100485 (2023).

Choi, W. et al. Recent advances in contrast-enhanced photoacoustic imaging: overcoming the physical and practical challenges. Chem. Rev. 123 , 7379–7419 (2023).

Ahn, J. et al. Fully integrated photoacoustic microscopy and photoplethysmography of human in vivo. Photoacoustics 27 , 100374 (2022).

Cho, S. W. et al. High-speed photoacoustic microscopy: a review dedicated on light sources. Photoacoustics 24 , 100291 (2021).

Park, E. et al. Azimuth mapping of fibrous tissue in linear dichroism-sensitive photoacoustic microscopy. Photoacoustics 31 , 100510 (2023).

Park, J. et al. Quadruple ultrasound, photoacoustic, optical coherence, and fluorescence fusion imaging with a transparent ultrasound transducer. Proc. Natl Acad. Sci. USA 118 , e1920879118 (2021).

Cho, S. et al. An ultrasensitive and broadband transparent ultrasound transducer for ultrasound and photoacoustic imaging in-vivo. Nat. Commun. 15 , 1444 (2024).

Wong, T. T. et al. Label-free automated three-dimensional imaging of whole organs by microtomy-assisted photoacoustic microscopy. Nat. Commun. 8 , 1386 (2017).

Wong, T. T. W. et al. Fast label-free multilayered histology-like imaging of human breast cancer by photoacoustic microscopy. Sci. Adv. 3 , e1602168 (2017).

Kim, D. et al. An ultraviolet-transparent ultrasound transducer enables high-resolution label-free photoacoustic histopathology. Laser Photonics Rev. 18 , 2300652 (2024).

Yang, J. G. et al. Recent advances in deep-learning-enhanced photoacoustic imaging. Adv. Photonics Nexus 2 , 054001 (2023).

Choi, S. et al. Deep learning enhances multiparametric dynamic volumetric photoacoustic computed tomography in vivo (DL‐PACT). Adv. Sci. 10 , 2202089 (2023).

Kim, J. et al. Deep learning acceleration of multiscale superresolution localization photoacoustic imaging. Light Sci. Appl. 11 , 131 (2022).

Jeon, S. et al. A deep learning-based model that reduces speed of sound aberrations for improved in vivo photoacoustic imaging. IEEE Trans. Image Process. 30 , 8773–8784 (2021).

Li, Y. Z. et al. Virtual histological staining of unlabeled autopsy tissue. Nat. Commun. 15 , 1684 (2024).

Bai, B. J. et al. Deep learning-enabled virtual histological staining of biological samples. Light Sci. Appl. 12 , 57 (2023).

Kendall, W. Y. et al. Deep learning classification of ex vivo human colon tissues using spectroscopic OCT. bioRxiv . https://doi.org/10.1101/2023.09.04.555974 (2023).

Zhang, Y. J. et al. Virtual staining of defocused autofluorescence images of unlabeled tissue using deep neural networks. Intell. Comput. 2022 , 9818965 (2022).

Zhang, G. H. et al. Image-to-images translation for multiple virtual histological staining of unlabeled human carotid atherosclerotic tissue. Mol. Imaging Biol. 24 , 31–41 (2022).

Winetraub, Y. et al. OCT2Hist: non-invasive virtual biopsy using optical coherence tomography. medRxiv . https://doi.org/10.1101/2021.03.31.21254733 (2021).

Picon, A. et al. Autofluorescence image reconstruction and virtual staining for in-vivo optical biopsying. IEEE Access 9 , 32081–32093 (2021).

Kaza, N., Ojaghi, A. & Robles, F. E. Virtual staining, segmentation, and classification of blood smears for label-free hematology analysis. BME Front. 2022 , 9853606 (2022).

Boktor, M. et al. Virtual histological staining of label-free total absorption photoacoustic remote sensing (TA-PARS). Sci. Rep. 12 , 10296 (2022).

Rivenson, Y. et al. Virtual histological staining of unlabelled tissue-autofluorescence images via deep learning. Nat. Biomed. Eng. 3 , 466–477 (2019).

Zhu, J. Y. et al. Unpaired image-to-image translation using cycle-consistent adversarial networks. In: Proceedings of the 2017 IEEE International Conference on Computer Vision, 2223–2232. Venice, Italy: IEEE, 2017.

Tsai, S. T. et al. H&E-like staining of OCT images of human skin via generative adversarial network. Appl. Phys. Lett. 121 , 134102 (2022).

Meng, X. Y., Li, X. & Wang, X. A computationally virtual histological staining method to ovarian cancer tissue by deep generative adversarial networks. Comput. Math. Methods Med. 2021 , 4244157 (2021).

Li, X. Y. et al. Unsupervised content-preserving transformation for optical microscopy. Light Sci. Appl. 10 , 44 (2021).

Cao, R. et al. Label-free intraoperative histology of bone tissue via deep-learning-assisted ultraviolet photoacoustic microscopy. Nat. Biomed. Eng. 7 , 124–134 (2023).

Kang, L. et al. Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining. Photoacoustics 25 , 100308 (2022).

Liu, Z. J. et al. Virtual formalin-fixed and paraffin-embedded staining of fresh brain tissue via stimulated Raman CycleGAN model. Sci. Adv. 10 , eadn3426 (2024).

Pérez, J. C. G., Baguer, D. O. & Maass, P. StainCUT: stain normalization with contrastive learning. J. Imaging 8 , 202 (2022).

Park, T. et al. Contrastive learning for unpaired image-to-image translation. Proceedings of the 16th European Conference on Computer Vision (pp. 319–345. Springer, Glasgow, UK, 2020).

Bayramoglu, N., Kannala, J. & Heikkilä, J. Deep learning for magnification independent breast cancer histopathology image classification. In : Proceedings of the 23rd International Conference on Pattern Recognition (ICPR), 2440–2445. Cancun, Mexico: IEEE, 2016.

Coudray, N. et al. Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning. Nat. Med. 24 , 1559–1567 (2018).

Wang, W., Ozolek, J. A. & Rohde, G. K. Detection and classification of thyroid follicular lesions based on nuclear structure from histopathology images. Cytom. Part A 77A , 485–494 (2010).

Mahmood, F. et al. Deep adversarial training for multi-organ nuclei segmentation in histopathology images. IEEE Trans. Med. Imaging 39 , 3257–3267 (2020).

Wu, J. C. et al. Learned end-to-end high-resolution lensless fiber imaging towards real-time cancer diagnosis. Sci. Rep. 12 , 18846 (2022).

Article   ADS   MathSciNet   Google Scholar  

Baik, J. W. et al. Intraoperative label‐free photoacoustic histopathology of clinical specimens. Laser Photonics Rev. 15 , 2100124 (2021).

Stoscheck, C. M. [6] Quantitation of protein. Methods Enzymol. 182 , 50–68 (1990).

Yao, D. K. et al. In vivo label-free photoacoustic microscopy of cell nuclei by excitation of DNA and RNA. Opt. Lett. 35 , 4139–4141 (2010).

Tjoa, E. & Guan, C. T. Quantifying explainability of saliency methods in deep neural networks with a synthetic dataset. IEEE Trans. Artif. Intell. 4 , 858–870 (2023).

Borji, A. Saliency prediction in the deep learning era: successes and limitations. IEEE Trans. Pattern Anal. Mach. Intell. 43 , 679–700 (2021).

Alqaraawi, A. et al. Evaluating saliency map explanations for convolutional neural networks: a user study. In : Proceedings of the 25th International Conference on Intelligent User Interfaces, 275–285. Cagliari, Italy: ACM, 2020.

Saadallah, A., Jakobs, M. & Morik, K. Explainable online deep neural network selection using adaptive saliency maps for time series forecasting. In : Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, 404–420. Bilbao, Spain: Springer, 2021.

Sundararajan, M., Taly, A. & Yan, Q. Q. Axiomatic attribution for deep networks. In : Proceedings of the 34th International Conference on Machine Learning, 3319–3328. Sydney, NSW, Australia: JMLR.org, 2017.

Heusel, M. et al. GANs trained by a two time-scale update rule converge to a local nash equilibrium. In : Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA, 6629–6640, Curran Associates Inc., 2017.

Bińkowski, M. et al. Demystifying MMD GANs. In : Proceedings of the 6th International Conference on Learning Representations. Vancouver, BC, Canada: OpenReview.net, 2018.

Ronneberger, O., Fischer, P. & Brox, T. U-Net: convolutional networks for biomedical image segmentation. In : Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich, Germany, 234–241 (Springer, 2015).

Vu, Q. D. et al. Methods for segmentation and classification of digital microscopy tissue images. Front. Bioeng. Biotechnol. 7 , 53 (2019).

Kumar, N. et al. A dataset and a technique for generalized nuclear segmentation for computational pathology. IEEE Trans. Med. Imaging 36 , 1550–1560 (2017).

Naylor, P. et al. Segmentation of nuclei in histopathology images by deep regression of the distance map. IEEE Trans. Med. Imaging 38 , 448–459 (2019).

Bradski, G. The openCV library. Dr. Dobb’s. J. Softw. Tools 25 , 120–125 (2000).

Vinutha, H. P. et al. Detection of outliers using interquartile range technique from intrusion dataset. In : Information and Decision Sciences. 511–518 (Singapore: Springer, 2018).

He, K. M. et al. Deep residual learning for image recognition. In : Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016, 770–778.

Tan, M. X. & Le, Q. V. Efficientnet: rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th International Conference on Machine Learning . Long Beach, California, USA: PMLR, 2019, 6105-6114.

Szegedy, C. et al. Rethinking the inception architecture for computer vision. In: Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition . Las Vegas, NV, USA: IEEE, 2016, 2818–2826.

Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. In: Proceedings of the 3rd International Conference on Learning Representations . San Diego, CA, USA, 2015.

Liu, Z. et al. Swin transformer: hierarchical vision transformer using shifted windows. In: Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision . Montreal, QC, Canada: IEEE, 2021, 10012–10022.

Bosman, F. T. et al. WHO Classification of Tumours of the Digestive System . 4th edn. (World Health Organization, Geneva, 2010).

McHugh, M. L. Interrater reliability: the kappa statistic. Biochem. Med. 22 , 276–282 (2012).

Selvaraju, R. R. et al. Grad-CAM: Visual explanations from deep networks via gradient-based localization. Proceedings of the 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE, 2017, 618–626.

Petsiuk, V., Das, A. & Saenko, K. RISE: randomized input sampling for explanation of black-box models. Proceedings of the British Machine Vision Conference 2018. Newcastle, UK: BMVA Press, 2018.

Fong, R., Patrick, M. & Vedaldi, A. Understanding deep networks via extremal perturbations and smooth masks. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South): IEEE, 2950-2958.

Johnson, J., Alahi, A. & Fei-Fei, L. Perceptual losses for real-time style transfer and super-resolution. Proceedings of the 14th European Conference on Computer Vision. Amsterdam, The Netherlands: Springer, 2016, 694–711.

Isola, P. et al. A. Image-to-image translation with conditional adversarial networks. In: Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition . Honolulu, HI, USA: IEEE, 2017, 1125–1134.

Kokhlikyan, N. et al. Captum: a unified and generic model interpretability library for PyTorch. Print at https://arxiv.org/abs/2009.07896 (2020).

Mao, X. D. et al. Least squares generative adversarial networks. Proceedings of the 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE, 2017, 2794–2802.

Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations. San Diego, CA, USA, 2015.

Jadon, S. A survey of loss functions for semantic segmentation. Proceedings of the 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB). Via del Mar, Chile: IEEE, 2020, 1–7.

Plotly Technologies Inc. Collaborative Data Science. (Montréal, QC: Plotly Technologies Inc., 2015).

Download references

Acknowledgements

This work was supported by the following sources: Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2020R1A6A1A03047902), NRF grant funded by the Ministry of Science and ICT (MSIT) (2023R1A2C3004880; 2021M3C1C3097624), Korea Medical Device Development Fund grant funded by the Korea government (MSIT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety) (Project Number: 1711195277, RS-2020-KD000008; 1711196475, RS-2023-00243633), Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. RS-2019-II191906, Artificial Intelligence Graduate School Program (POSTECH)), and BK21 FOUR program.

Author information

These authors contributed equally: Chiho Yoon, Eunwoo Park, Sampa Misra

Authors and Affiliations

Departments of Electrical Engineering, Convergence IT Engineering, Mechanical Engineering, Medical Science and Engineering, Graduate School of Artificial Intelligence, and Medical Device Innovation Center, Pohang University of Science and Technology (POSTECH), Pohang, Republic of Korea

Chiho Yoon, Eunwoo Park, Sampa Misra, Jin Young Kim, Jin Woo Baik & Chulhong Kim

Opticho Inc., Pohang, Republic of Korea

Jin Young Kim & Chulhong Kim

Department of Health Sciences and Technology, Gachon Advanced Institute for Health Sciences and Technology (GAIHST), Gachon University, Incheon, Republic of Korea

Kwang Gi Kim

Cancer Research Institute, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

Chan Kwon Jung

Department of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

You can also search for this author in PubMed   Google Scholar

Contributions

C.Y., E.P., and S.M. planned the study and drafted the manuscript. E.P., J.Y.K., J.W.B., and C.K.J. performed the data preparation. C.Y., S.M., and K.G.K. designed and carried out the main framework. C.K. and C.K.J. supervised the project. All authors discussed the results and contributed to the writing.

Corresponding authors

Correspondence to Chan Kwon Jung or Chulhong Kim .

Ethics declarations

Conflict of interest.

C.K. and J.Y.K. have financial interests in OPTICHO, which, however, did not support this work. All other authors declare no conflicts of interest.

Supplementary information

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Yoon, C., Park, E., Misra, S. et al. Deep learning-based virtual staining, segmentation, and classification in label-free photoacoustic histology of human specimens. Light Sci Appl 13 , 226 (2024). https://doi.org/10.1038/s41377-024-01554-7

Download citation

Received : 31 January 2024

Revised : 08 July 2024

Accepted : 24 July 2024

Published : 02 September 2024

DOI : https://doi.org/10.1038/s41377-024-01554-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

• JSE Magazine
• Verify a Broker (New!)
• Find A Broker
• Training Academy
• Virtual Trading Game
• Investor Services
• Podcast | JSE Investor Avenue
• Investment Challenge
• Empowerment Segment
• Special Purpose Acquisition Company (SPAC)
• Standard Listing
• Section 12J
• Green Bonds
• Social Bonds
• Sustainability-bonds
• Actively Managed Certificates
• Exchange Traded Funds
• Exchange Traded Notes
• Structured Products
• Find a Sponsor
• Listing Fee Calculator
• JSE Private Placements
• Market making Index Options
• International derivatives
• eCFD's
• MSCI South Africa Index
• Single Stock Futures
• Dividend Neutral Stock Futures
• Dividend Futures
• Equity Index Futures
• SAVI Squared
• SAVI Top 40
• Currency Futures Prices
• Currency Futures
• Currency Options
• Currency Quanto Futures and Options
• African Currency
• Bond Futures
• Bond Options
• Bond Index Futures
• Short Term interest rate derivatives
• ​Physical Delivery
• Agricultural derivatives
• Energy derivatives
• Metal derivatives
• Become An Equity Trading Member
• Becoming a Debt and Derivatives Market Member
• Become an Accredited Equity Market Software Provider
• Specialist Securities
• Depository Receipts
• Ordinary Shares
• Preference Shares
• Share Instalments
• Real Estate Investment Trusts (REITs)
• Equity Enhanced Order Suite
• Listed Companies and Instruments
• Colocation & Connectivity
• JSE Enterprise Acceleration Programme
• SME Capital Matching Initiative
• JSE Investor Services
• Technical Documents
• JSE Venues & Facilities for Hire
• Services Documentation
• ​Valuations​
• Cash Bonds & ETP Operations
• Derivatives Operations
• Equities Operations
• Bond ETP Risk Management
• Derivative Market Risk Management
• Equity Market Risk Management
• FTSE/JSE Africa Index Series
• FTSE/JSE Fixed Income Index Series
• FTSE/JSE Responsible Investment index series
• ICB Industry
• FTSE/JSE Multi-Asset Composite Index Series
• Global Investor
• JSE Ventures Voluntary Carbon Market
• Clearing and Settlement Operations
• Risk Management
• Knowledge Base
• African Exchanges Linkage Project
• JSE Service Hotlines
• JSE Market Statistics
• Equity Market Data
• Bond Market Data
• Equity Derivatives Data
• Commodity Derivatives Data
• Currency Derivatives Data
• Interest Rate Derivatives Data
• Index Market Data
• Historical Data
• Market Announcements
• Issuer regulation
• Market regulation
• Client Portal
• Market news

FSCA approves JSE Listing Requirements Dealing with Market Segmentation

The Johannesburg Stock Exchange (JSE) is pleased to announce that the Financial Sector Conduct Authority (FSCA) has approved its amendments to the Listings Requirements dealing with Market Segmentation which come into effective on 23 September 2024.

JSE issues apology for the incorrect meta description on broker search

JSE Limited (“JSE) was alerted to an incorrect meta description automatically generated by search engines when searches relating to the JSE in relation to company information is performed. This resulted in the JSE’s awareness campaign relating to a cautionary pop-up on the JSE website regarding investment scams being displayed in search results.

The JSE has implemented technical measures to remedy the incorrect description.

We sincerely apologise for the error and in the event that you require any further information in respect of the remedial actions implemented or the JSE to clarify the incorrect search results, please contact  [email protected] .

Johannesburg, 03 September 2024 – The Market Segmentation Project repositions the JSE’s Main Board into two segments, the Prime and the General Segment. This new structure aims to offer a suitable and efficient level of regulation tailored to the size and liquidity of issuers on the Main Board, while continuing to uphold investor confidence in the market.

The General Segment affords issuers on the Main Board listing with different application of certain provisions of the Listings Requirements. Issuers seeking to apply for the General Segment can submit an application to the JSE from 23 September 2024, with the effective launch date of the General Segment to be communicated.

Once approved, the issuer will be classified under the General Segment. The General Segment affords meaningful regulatory relief to issuers whilst maintaining transparency and disclosure. Some of the benefits include:

• More enabling capital raising measures;
• Significant cost savings;
• Efficient and cost effective financial reporting; and
• Greater flexibility for the boards to manage the business.

“We welcome the FSCA’s approval of the amendments to the JSE’s Listing Requirements in relation to the Market Segmentation Project as we believe it will create a flexible and enabling environment for certain companies listed on the Main Board to raise capital and undertake corporate actions within an appropriate and relevant regulatory framework,” says Andre Visser, Director: Issuer Regulation at the JSE.

The General Segment offers the following, to name a few:

• An automatic annual rolling general authority to issue shares for cash without shareholders ’ approval representing up to 10% of the issuer’s issued share capital;
• A general repurchase authority not requiring shareholders ’ approval, not exceeding 20% in any one financial year;
• A specific repurchase authority not requiring shareholders ’ approval, subject to it not involving related parties and does not exceed 20% in any one financial year;
• Removal of fairness opinions for related party corporate actions and transactions, with more focus being placed on governance arrangements and transparency (agreements open for inspection), and exclusion from voting for related parties and associates;
• Removal of the requirement to release results announcements within three months. Issuers will only be required to prepare an annual report within four months
• Removal of the preparation of pro forma financial information but rather inclusion of a detailed narrative on the impact of the transaction/corporate action on the financial statements;
• Increasing the category 1 percentage ratio to 50% or more (increase by 20%, currently 30%), which increases the category 2 threshold accordingly;
• Requiring only two year audited historical financial information for the subject of a category 1 transaction (currently three years of audited historical information);
• Increasing the small-related party transaction percentage ratio to 3% and less than or equal to 10% (increase from 0,25% and 5%); and
• Increasing the classification of a material shareholder, from 10% to 20%.

Classification into the General Segment is only available to Main Board issuers who are not included in the FTSE/JSE All Share Index.

The JSE remains committed to creating an enabling environment for listed companies and continually assesses its Listing Requirements to ensure they are relevant and applicable to the ever-evolving needs of the market.

Further to the approval of the Market Segmentation reforms, the JSE announced the expansion of its secondary listings framework and has added Tadawul as well as all the Euronext exchanges (Amsterdam, Brussels, Dublin, Paris, Milan, Lisbon and Oslo) to its list of approved and accredited exchanges.

Tadawul and Euronext exchanges are now included in the group of global exchanges recognised for the fast-track process, including the London Stock Exchange, Australian Securities Exchange, New York Stock Exchange, Toronto Stock Exchange, Singapore Stock Exchange and Hong Kong Exchanges and Clearing Ltd. This initiative is part of the JSE's continuous commitment to improving accessibility and efficiency for international companies.

Together, these projects demonstrate the JSE’s proactive approach to regulatory innovation and its dedication to enhancing the attractiveness and effectiveness of the South African capital markets.

The new provisions to the JSE Listings Requirements are available on the JSE’s website:

Announcements regarding Listings Requirements: https://www.jse.co.za/regulation/companies-issuer-regulation

ABOUT THE JSE

The Johannesburg Stock Exchange (JSE) has a well-established history of operating as a marketplace for trading financial products. It is a pioneering, globally connected exchange group that enables inclusive economic growth through trusted, world-class, socially responsible products, and services for the investor of the future. It offers secure and efficient primary and secondary capital markets across a diverse range of securities, spanning equities, derivatives, and debt markets. It prides itself on being the market of choice for local and international investors looking to gain exposure to leading capital markets on the African continent.

The JSE is currently ranked in the Top 20 largest stock exchanges in the world by market capitalisation, and is the largest stock exchange in Africa, having been in operation for 137 years. As a leading global exchange, the JSE co-creates, unlocks value and makes real connections happen. www.jse.co.za  

The JSE takes your privacy seriously in accordance with the POPI Act. We endeavour to only send you relevant information that we think will be of interest to you and the media title you work at. No action is required if you are happy to continue receiving JSE news and information. Should you change your mind at any time in the future, please do let us know. If you do not wish to receive JSE news and information, please respond to this email and we will remove you from future distributions.

Share this article