User segmentation is the practice of grouping users together based on common traits they share. These characteristics might include region, paid vs. unpaid customers, users who are spending the most time in your product, or, conversely, users who are spending the least amount of time.
Use this guide to user segmentation to learn how grouping users with similar characteristics can not only help you track product health and performance, but optimize it.
What is user segmentation?
While most website analytics only allow you to track page views, sophisticated product teams leverage product analytics to understand their users’ demographics (what actions or “events” they performed) and demographics (who they are). Because not all users created equal, segmenting users into different groups and tracking their movement between different engagement states is a great way to get a deeper understanding of your users and find opportunities for further growth.
Why should I be segmenting my users?
By segmenting similar users into similar groups and subgroups, product teams are able to:
- Figure out who their most valuable users are and where they come from
- Know where to focus marketing efforts to attract more high-value users
- Identify purchasing trends and predict future purchasing behaviors
- Find underserved customers
- Nurture relationships by offering custom solutions to meet specific needs
- Identify incentives that encourage conversion and repeat business
- Inform both product development and marketing strategy
- Understand pricing sensitivity and create a more profitable pricing strategy
It’s no mystery that consumers have not only grown accustomed to increasingly customized products, content, and services—they’ve come to fully expect them. In order to deliver the kind of tailor-made experience that sets you apart from the competition and creates lasting customer loyalty, you must first understand precisely who your users are.
But knowing who’s using your product is only one piece of the puzzle. Even if you were able to understand each and every one of your unique users, how could you possibly cater to them all? Intuiting what users want based on their individual data or gut feelings is neither a practical nor a productive approach. Engaging with them is a science-powered art that requires tools, time, and proper strategy. That’s where segmentation comes in.
User segmentation is designed to help you scale your efforts efficiently and effectively so that you can understand how similar types of users are collectively engaging—or not engaging—with your product, and tailor their experience accordingly. The better teams are at segmenting their users, the more personalized the user experience will be, which is vital to achieving long-term product success.
What types of segmentation strategies should I be using?
Any piece of user data that companies are able to track can be useful in segmentation, so long as those distinctions are meaningful. Whether they know it or not, your users are constantly conveying vital information about themselves and their relationship with your product, but it’s up to you to use that information to make product decisions.
How do you decide how to segment your users? Some commonly used strategies include:
- Location: How do your users’ behaviors differ across the globe? Segmenting users by location allows teams to determine whether your product is resonating with certain geographies but not others, and can help inform whether different marketing strategies are needed.
- Device or browser type: How do behaviors change across mobile vs. web, iOS vs. Android, Chrome vs. Safari? This type of segmentation will allow you to customize the experience on these platforms wherever possible.
- Lifecycle stage: Segmenting new users, power users, and dormant users allows you to understand what “good” users look like and how many you have. By identifying and adding power users to beta programs, for example, you can figure out if your feature launches are likely to resonate with your broader user base. Segmenting out your dormant users, on the other hand, allows you to get them back into your product by targeting them with re-engagement campaigns.
- Time on website/app: Your users’ time is valuable. If they’re choosing to spend it using your website or app, it’s a powerful indicator that they’re finding value in it. If they’re not spending time in your product, that’s also a useful, albeit different, signal. By segmenting users by the time spent, you can learn which features, factors, and content are correlated with higher engagement to surface new ideas for driving increased usage.
- Plan type: What’s the difference between a user who is a paid subscriber and one who is on a free plan? Or between a user who has purchased once vs. ten times? Conversion is defined differently for every business, but it’s critical to know. By analyzing users that have passed a conversion checkpoint, you can isolate the characteristics shared by your most prized users and use that information to convert even more of them.
- “Negative” behaviors: Not all user behaviors are positive. Sometimes customers churn, abandon their shopping carts, or downgrade subscriptions. By analyzing the events that preceded these actions, such as a service interruption or a new feature release, you can identify areas of friction and improve those parts of your website or app.
A media streaming company, for example, might choose to segment their users like this:
By segmenting their users, the media streaming company can see who their most loyal users are, who’s likely to purchase, who could be spending more and getting more from their product, and who is close to churning. By identifying and understanding each segment, the company will be better equipped to tailor their experience.
Whichever segmentation strategy you choose, keep in mind that it should always tie back to your product metrics, such as purchases or revenue.
Every team will have its own idea of what data is most valuable to capture. An ad-driven online magazine might make money from views and will be quite happy tracking viewer traffic, whereas a CRM platform might care about sales and recurring revenue. It all depends on the business model.
Here are a few types of data that teams might want to capture from their users:
- Demographic data: A user’s social data, gender, birthday, language, location, marital status, or income
- Psychographic data: A user’s interests, beliefs, affiliations, or socioeconomic status
- Behavioral data: A user’s actions such as visits, taps, clicks, time on site/app, or conversions
- Firmographic data: Like demographics, but for businesses: company age, employee count, revenue, industry, location, or business model (B2B or B2C)
- Technographic data: Technologies a company uses, such as CRM, marketing software, or ERP provider
Collecting some or all of these types of data is what allows you to create user profiles, which are needed in order to perform segmentation. The below user profile may look short and simple, but it contains several types of data.
A user profile for a fictional fitness app, for example, might look like this:
- Demographics: Female, aged 25-40
- Psychographics: Loves cycling classes
- Behaviors: Uses the app more than 5 times per week
Using a product analytics platform, the fitness app team can segment users with similar profiles and continue to monitor it over time. Whenever the team thinks about adding new features, they can look at this segment, and measure how it responds to new releases. They can even assign scores and weights to different segments to determine their lifetime value and measure the ROI of marketing campaigns.
Communicating effectively with your users is a data-driven art. Understanding what your users want takes the right metrics and the tools to match.
Learn how to pick the metrics that matter most for your product
Download our acclaimed Guide to Product Metrics todayGet the guide
How can I segment my users?
The user segmentation process consists of four key steps:
1. Collect data about your users
To segment customers, you need to be able to view all of your users in one place. But getting a grasp on that entire audience isn’t always easy. Sites and apps typically aren’t built to analyze themselves, and it can be difficult to track individual users and their journeys—much less compare two different user segments.
A media streaming company, for example, might be able to use its website’s analytics to determine how many users simply visit the site versus how many stay to click an ad or watch a video. But which users are which? What screenflow do those users typically follow? To draw meaningful distinctions, the company needs a way to group visitors by their characteristics and behaviors. Without knowing who, it’s nearly impossible to determine how or why. This is where product analytics come into play.
2. Identify segments that matter most
As we mentioned before, segments are only useful if they’re tied to product management metrics. If teams know what they want their users to do, they should already be able to discern what their first few segments should be.
If the goal is for users to make a purchase, your segments could be: users who have purchased once, users who have purchased repeatedly, and users who have not purchased. If the goal is to increase app usage, you might segment by time spent in-app to isolate power users from occasional visitors.
Here are some of the common metrics teams use to define segments:
- Engagement / usage
- Acquisition source
- Customer lifetime value (CLV)
- Average visits per user
- Customer journey milestones
3. Use an analytics platform to segment your users
- Compare the usage rates of paid and free users
- Measure the retention rate for users acquired from social media
- Determine which actions taken during a free trial make users more likely to convert
- See whether usage increased after a new feature release
4. Make changes based on segments
By comparing segments to each other, or even the total user population, you can identify meaningful similarities and differences between them. This allows you not only to track product health and performance, but optimize it.
For a mobile fitness app, for example, a team might find that its most valuable users are more likely than the average user to finish setting up their profile. The team can then rearrange its onboarding flow to make sure that every new user completes their profile.
A social media app might find that users that download their mobile app are twice as sticky, and create in-app notifications on their desktop website to drive more mobile users.
Segments lead to insights. When insights are met with action, better products get built.