Customer segmentation is a classic marketing concept for good reason.
There’s a high chance that not all of your customers have the same type of business, come from the same part of the world, or even use your product in the same way.
The larger your SaaS empire becomes, the more likely this is to be the case.
If we assume that not all customers are the same, it would be futile to give all of them the same generic product tour (oh hey Intercom!). Or give them all customer support to the same level of depth. Or show them all the same features during secondary onboarding.
More often than not, what one user cares about is of little or no interest to another user.
Intelligent product managers solve this problem by segmenting their users into different cohorts and managing each one according to its individual needs.
But how do you go about segmenting users? What does the segmentation process look like in practice? How do you assess the needs and wants of each user cohort?
This article will give practical answers to these important questions.
- Customer Segmentation is when businesses divide customers into groups according to common characteristics, in order to create better product experiences and more effective market positioning.
- It’s an essential concept for SaaS businesses because you can customize primary, secondary, or tertiary onboarding to a segment’s unique needs. This is much better than giving the same generic experience to all users.
- To segment users, first choose whether you want to segment by demographics, in-app behavior, user attributes, NPS, or account.
- Then assess the needs of your chosen segment by conducting market research. A good customer analytics tool helps a lot with this.
- Make sure you’re reassessing your segments frequently, as they are often subject to change as your business evolves.
- Userpilot offers a huge number of practical ways to segment your customers.
What Is Customer Segmentation?
Here’s our definition of customer segmentation:
The act of dividing your customers into different groups, according to certain common characteristics or behavior, to facilitate better product experiences and more effective market positioning.
You can have as many or as few customer segments (also called “cohorts”) as you wish.
You can also involve as many or as few different criteria in the segmentation process as you wish.
More on the criteria on which you can base the segmentation later in this article.
For now, let’s examine the question of:
Why Is Customer Segmentation Important For SaaS?
Here are 3 reasons why segmenting your customers is an essential exercise:
1. You can customize a user’s onboarding to their needs
As we’ve written numerous times before on this blog, traditional product tours suck.
Classically, a product tour will take ALL the main features of a given product and show them in the SAME way (often in a long, meaningless sequence) to ALL users.
This is a sure way to bore your users. And, honestly, some of them will probably rage-quit before activating. (Source: I may have done this before.)
By contrast, if your SaaS business segments users correctly, you can deliver a different onboarding experience to each different user segment, by means of an interactive product walkthrough.
Follow up with native tooltips that point the user to the minority of features their specific user segment cares about most.
Result: happier users, and higher activation rates for your SaaS business.
2. You can do feature launches aimed at specific user segments
The above logic about the value of segmentation applies just as much to secondary and tertiary onboarding as it does to initial onboarding.
Put another way, when you develop a new feature and point users to it, you’re not necessarily going to point all users to it at once.
It only makes sense to show the new feature to the specific user segment that needs it.
For example, let’s say your customer analytics software highlighted that enterprise customers were consistently asking for more features guaranteeing security compliance.
There’s a good chance that your SME customers have never even heard of security compliance requirements, much less care about it. They’re too busy ensuring their business survives day-to-day.
So when releasing that feature, it would make sense to only highlight it to the enterprise customer segment.
3. You can replicate the behavior patterns of your best users
A common user segment seen in many SaaS companies is the “power user.”
This is the company that refers you to all their friends, uses your platform for 13 hours per day, always answers your NPS surveys immediately with a 10, and… well, you get the picture.
They’re right at the end of the product adoption funnel.
Segmenting power users from other users acknowledges that there is something about a power user that makes them a bit different.
Once you’ve got them in their own specific user segment, you can analyze their behavior and see what you can learn. For example:
- Perhaps you notice that power users activate in half the time that regular users do. Why is that? How can you lead other users to activate faster?
- Perhaps you notice that power users come from a particular geographical area. This is a sign that your sales team probably needs to book more calls with that part of the world.
- Or perhaps you notice that all the power users are social media managers, and none of the other user personas come anywhere close. Again, this is valuable information for positioning, both before and post-sale.
What Does The Customer Segmentation Process Look Like?
If I were to explain how to segment customers to someone who had never done it before, I would break it down into 3 steps:
- Choose a model. What characteristics and behaviors are you going to segment users by?
- Analyze each segment individually. This will help you gauge their individual needs.
- Repeat this process regularly to refine your understanding of your customer segments. As a colleague said in a recent marketing team meeting: “persona work never ends.”
Now let’s look at steps 1 and 2 in more detail.
What Models Can Be Used For Customer Segmentation?
Here are the main criteria types you can use.
1. Segmenting by demographics
Classically, this includes age, gender and income.
More detailed models might also include location, race, nationality, marital status, class, occupation and even lifestyle.
For most SaaS companies, the most useful demographic data point is likely going to be occupation. A product manager and a CEO might use your product in completely different ways, for example.
2. By in-app behavior
This might include:
- Users that have or have not activated.
- Users that have or have not completed some custom event that your developers have defined.
- Users that have or have not gone through a particular UX flow.
- Users that have or have not engaged with a new feature.
The key pattern here is to segment according to whether a user has completed a particular action or not.
3. By user attribute
The difference here from the previous segmentation model is that user attributes are not related to the actions the user performs in-app, but rather to who they are.
Another way to think about user attributes is that they’re more at the account level than the level of in-app behavior.
So this includes things like:
- Which plan the user is on
- Whether they registered themselves or whether they were registered by a company member
- Which language they use
- Which device they use
4. By account
This includes things like:
- New users – who probably need more hand-holding from customer support than other users.
- Power users – whose behaviors you want to understand and replicate.
- Enterprise users – who provide the majority of the cashflow, and have the most complex support requirements.
5. By user sentiment
If you use the NPS framework, this segments users into:
- Promoters – users who love your product and are actively referring it to their friends.
- Passives – users who are so-so about your product; they’ll use it, but they don’t live for it.
- Detractors – users who are in danger of churning, and actively encouraging their friends to churn.
Segmentation can be as simple as focusing on only one of these models at a time, or as complex as mixing and matching criteria from different models.
In general, the larger the SaaS company, the more segments it will have and the more resources it will devote to analyzing those segments in a huge amount of detail.
For smaller companies, 3-4 segments is probably a good starting point.
How Do You Perform Customer Segment Analysis?
Once you’ve identified your customer segments, it’s time to analyze them to figure out what their needs are so that you can serve them better.
There are several sources you can go to in order to find out their needs:
- Customer analytics software has a wealth of data about your customers. Our last article on that matter is well worth reading in full.
- It’s worth scouring the online assets of your customers, such as their website and social media pages.
- Sometimes social media groups that are aimed at businesses similar to your customers can be a good source of insight. Userpilot maintains a Facebook group about product marketing, for example.
- If you think you can get enough responses, sometimes a direct survey through a tool like SurveyMonkey is worth trying. I have done this before myself at a previous business I ran. Alternatively, you can create an in-app survey – here are 31 in-app survey questions you should ask your customers in-app.
- If all else fails, schedule calls with your customers one-on-one and talk to them. Sometimes the old-fashioned solution works best! Since so few businesses take the time to do this, your customer will be flattered.
Once you’ve identified your different customer segments and their respective needs, you’ll be able to understand how you can meet those needs using your product.
Customer Segmentation Examples In Practice
Userpilot offers tons of practical ways to segment users. Let’s look at a few of them together.
1. By NPS score
This is perhaps the easiest way to segment users in Userpilot.
Simply go to NPS under the Evaluate option in the left-hand menu:
From there, click Responses in the menu at the top.
Click on the dropdown that says All Responses, and you’ll be able to sort your users according to whether they’re Promoters, Passives, or Detractors:
If you want to go even more granular, select Users from the left-hand menu, and click Add Filters.
Under the dropdown from Add Conditions, select NPS.
This will give you a ton of different options for selecting NPS score that are more than, equal to, or less than a certain amount. Here I’ve selected scores that are less than 8:
You can now click on the green button that says Save Segment, and Userpilot will remember this segment for future analysis.
2. By demographics
Let’s use Location as an example here.
To sort by location, go to Users on the left-hand navigation again.
Again, click on Add Filters at the top of the screen.
Under the Add Conditions dropdown, select Location.
You’ll get this screen popping up:
Have a play with the country options in order to set the location that you wish to segment by.
Alternatively, hit the + button to add multiple locations, or other types of condition.
3. By user attribute
Before you can sort by user attribute in Userpilot, your developers will need to pass us app-specific information so that you can segment your users by attributes that make sense to your individual business.
Once you’ve done that, segmenting by attribute is incredibly simple.
As with the previous segment types, go to Users in the left-hand menu and click on Add Filters at the top of the screen.
This time, you want to click on User Identification under the dropdown under Add Conditions:
This screen will pop up:
Now, choose the dropdown that says User ID. Lo and behold, all the user attributes you’ve passed us will appear as options for you to choose from.
Once you’ve selected one or more attributes to segment with, click the green button at the top to Save Segment. Easy.
4. By company
Again, go to Users in the left-hand menu, click Add Filters, and this time select Company Identification from the dropdown:
Click the dropdown where it says ID. This will give you various helpful options for segmenting at the company level.
There’s a way to segment by the number of employees a customer has using your product:
This one would be good for identifying enterprise accounts, for example.
Alternatively, you can segment by when customers signed up:
Or when they were last online:
5. By custom event
This is a neat one for segmenting users according to how many of them have completed the actions they need to take in order to activate.
First, your developers will need to define custom events and pass them to Userpilot or – if the event you want to track can be easily attributed to a single click on your frontend – you can also use Feature Tagging (without needing to have your devs pass custom events.)
Next, go to Goals on the left-hand menu. It’s under Analyze.
You’ll see a list of all the custom events and the rate at which they’ve been completed:
To see the users that have completed a custom event, simply click on the relevant event.
Userpilot will then list the users that have completed it, and give you an option to download that user segment as a list. Look for the download icon on the right:
This is very useful if you want to collect email addresses from a certain segment and send them surveys by email, for example.
Alternatively, go to Users, click Add Filters, and select Custom Events from the dropdown.
This method is helpful if you want to segment by whether users have NOT completed a certain event:
6. By experience flow
Sometimes it can be very useful to know which users have interacted with certain parts of your UI. This is especially useful if you’re testing a new feature, for example.
Userpilot make this super easy.
Simply go to Experiences on the left-hand menu:
This will give you a list of all the experience flows you’ve created with Userpilot.
Now select the experience you want to segment by.
In the top navigation, go to Analytics:
And click the red Export button to export a list of the users that have completed this experience.
Alternatively, go to Users, click Add Filters, and select Engagement from the dropdown.
This will give you tons of options to segment according to whether users have seen, engaged with, or completed experience flows that you previously defined within Userpilot.
7. Combining multiple segmentation criterias
It’s possible to use the segmenting feature under Users to combine some of the features we just discussed to produce some really nuanced segments!
Go to Users again and click on Add Filters. Select whatever filter you want this time. I’m going to start with NPS.
The key here is the little + icon on the right, which allows you to create a segment from multiple user criteria. If I do this a few times, I can segment for:
- Users with an NPS score of 9
- Who have also completed the Link Accounts experience
- Who also signed up before the beginning of 2022
- And are also not located in Argentina
As you can see, the only limit here is imagination 🙂
We hope you’ve enjoyed this guide to customer segmentation. It’s a really under-appreciated concept with so many applications in SaaS businesses.
If you take nothing else away from this article, please don’t become one of those businesses that treats all users exactly the same. Your customers will thank you, trust me.
You can use Userpilot to segment your users to your heart’s content. Get a free demo to get started!