Using Data-Driven Personas for Enhanced User Segmentation

User segmentation is the practice of dividing potential or existing users into groups that share similar characteristics. As such, user segments go hand in hand with data-driven personas! 

User segmentation is employed as a user understanding technique by nearly every product team. Think of it this way! The ultimate goal of product design or product strategy for that matter is to create products/services that:

  1. are aimed at understanding user needs and pain points as best as possible, 
  2. have essential features to solve the users’ primary needs as best as possible. 

In order to be able to identify user needs and pain points and serve them well, it is essential to have a good understanding of your user/customer base. Both user segmentation and data-driven personas can aid in this understanding.

Imagine a scenario where you have recently developed a product or service and want to understand:

(1) How many people from a certain country use it? 

(2) Frequency of use of your product/service, 

(3) How are people interacting with your product or service, etc. 

An understanding of such user needs and wants can be done through segmenting (i.e.,. dividing your users into homogenous groups), a practice known as user segmentation. 

Identifying user segments is crucial in understanding how people are using your product or service offerings (Source: Moosend)

Example of user segmentation

Take the following example from a cybersecurity firm Avira, based in Germany that segments their user base, which is in the millions and is spread across multiple countries around the world (I have also personally used both their free and paid offerings):

Example of user segmentation in practice. A cybersecurity firm Avira segmenting their users to understand how their products are being used and what functionalities can they introduce (Source: Mixpanel)

Notice from the image how Avira is dividing users into different categories. For instance, they are dividing (or segmenting) their users into two categories in one instance- Free v/s paid users, with the intention being:

  • To understand which features of the product are being used by the free and paid cohorts respectively. 
  • To understand the behavior of free v/s paid users. 
  • How to introduce additional functionalities/features to get free users to convert into paid users. 

The above are just among a plethora of segments that a product team can divide their user or customer base into depending on what they wish to achieve!

The following quote from a scientific paper titled Making Meaningful User Segments from Datasets Using Product Dissemination and Product Impact by the Automatic Persona Generation (APG) team at Qatar Computing Research Institute (QCRI) sums up the value of user segmentation nicely: 


“The identification of user segments is typically aimed at the understanding of a subset of people’s reactions, interactions, uses, etc., based on one or more key performance indicators (KPI), to achieve some goal or objective, such as increasing revenue, increasing market share, or designing future content.”


Moving on, the most common types of user segmentation are:

  • Demographic Segmentation – based on gender, age, occupation etc
  • Geographic Segmentation – based on country, state, or city of residence. Some organizations might even want to segment by specific towns or counties.
  • Technographic Segmentation – based on preferred technologies, software, and mobile devices.
  • Psychographic Segmentation – based on personal attitudes, values, interests, or personality traits.
  • Behavioral Segmentation – based on actions or inactions, spending/consumption habits, feature use, etc.

Data-Driven Personas and User Segmentation

Personas can be understood as humanized profiles of customer segments. 

Depending on different factors such as budget, understanding persona value, and creation, etc., a typical persona profile can contain all or some of the different characteristics mentioned above. 

I would go so far in saying that:

“Personas ‘lift’ segments by providing a much richer qualitative (or quantitative) picture of a typical ‘fictional’ customer within a segment, animating their personality and values”. 

Let’s take a look at the following persona of Indah, who is a busy professional wanting to become more health-conscious, and shop for more organic food products for herself and her family. 

If an organization was wondering what kind of pain points do busy professionals face when doing online grocery shopping, and then develop a product or service to alleviate these pain points, then they would have to:

  • Conduct extensive research of professionals in and around their target market (town, city, country.) 
  • Understand differences between male v/s female personas on their frustrations with online grocery shopping etc. 
  • Flesh out (maybe) a generalized “persona” that can encapsulate all the frustrations and/or pain points in one persona. 

By doing so, a firm would be able to either have multiple personas or a generalized user persona of their user segment (which is a health-conscious busy professional in the above example). 

Not only can a persona represent different types of user segments (e.g. demographic, geographics, behavioral, etc.) but they help give the product team a perspective, which is to say that the designers and developers know the kind of segment for which they are designing the product or service. 

Even though personas are a useful tool in representing these user segments, manual creation of these personas is expensive and time-consuming. For instance, depending on the size of your company, it can take up to 55 to 102.5 staff hours! 

But what if there is a way for you to automatically generate as many personas as you want of your user/customer base that represents different user segments.

The Automatic Persona Generation system (APG) helps you in achieving exactly ALL of the above.

Automatic Persona Generation and User Segmentation

First, a little bit about APG!

APG is a methodology and system for the automatic creation of personas from online analytics and social media data. The purpose is to turn faceless, nameless online analytics data into human representations, i.e., personas. APG is an interactive persona analytics system! 

APG uses the online analytics data which your organization has to identify customer behaviors, generate customer segments, and then enriches these customer segments with gender, age, and nationality-appropriate names and pictures. Customer loyalty rating, customer interests, product interactions, brand sentiment, and segment sizes are represented by the personas, and all of this is done in a privacy-preserving process using only aggregated data. 

Not only are the data-driven personas generated using APG more detailed than a traditional persona example I presented above, but the entire process is automated! 

The data-driven personas (DDPs) generated using the APG are representations of the actual segment of users presented as an imaginary person. 

In practice, the best way of using personas for customer segmentation is to create DDPs using APG. Using this approach, you can create an arbitrary number of personas from your source data; e.g., 5, 10, 50, or 100 personas, in just a matter of hours! 

Here is what you get in a data-driven persona profile created by APG!

A data-driven persona created using APG

All the different characteristics of a user (e.g. demographics, behavioral characteristics, etc). on the basis of which users can be “segmented” are represented in a clear, well thought out way using APG.

For instance, in the above persona, you get all the following information: 

  1. Name – a chronological gender, age, and country appropriate name
  2. Photo – a chronological gender, age, and country appropriate photo
  3. Description – a short description of the persona including online habits
  4. Job – the persona’s most likely job
  5. Education Level – the persona’s most likely educational level
  6. Relationship Status – the persona’s most likely relationship status
  7. Audience Size – the customer segment size that the persona is representing
  8. Gender – the persona’s most likely gender
  9. Age – the persona’s most likely age
  10. Nationality – the persona’s most likely nationality
  11. Loyalty Rating – a three-level rating of the persona’s loyalty to the business
  12. Sentiment Analysis – the persona’s sentiment toward social media content from the business
  13. Social Media Conversations – the social media conversations that the persona has engaged in or viewed
  14. Filter of social media conversations by topic, sentiment, or similarity – filtering of the persona’s social media conversations
  15. Hate filter – a filter to see the personas ‘nice’/’not so nice’ comments
  16. Refresh social media conversations – a refresh option for new social media conversations that the persona has engaged in or viewed
  17. Interests – the persona’s interests as expressed via online engagement
  18. Engagement – the online content, products, or services from the business that the persona has engaged with
  19. Timeline – a chronological graph of the persona’s engagement with the company since the first data collection
  20. Print the profile – ability to print the persona profile

The benefits of using APG for user segmentation include:

  • Data-driven personas are created within minutes
  • The data-driven personas can be updated automatically every month. 
  • The data-driven personas are an actual representation of your user/customer base and can help nail down different user segments. For instance- if you want to find out how many males or females are interacting positively with your product or service (among other questions), you can do so using the APG.

To conclude, I would like to highlight that personas give faces to user segments, which without a persona would be nameless and faceless.

A DDP generated using APG takes it one step further in automatically generating the most accurate personas for your user or customer base which you can then use to segment and understand your users. 

Would you like to learn more?

If this article got you intrigued, read our persona analytics research for more in-depth knowledge on persona development.

If you are interested in learning more about how APG can help identify user pain-points and inform design decisions for your team, contact us!  

Further reading

Jansen, B. J., Salminen, J. O., & Jung, S. (2020). Making Meaningful User Segments from Datasets Using Product Dissemination and Product Impact. Data and Information Management. doi:

An, J., Kwak, H., Salminen, J., Jung, S.G., and Jansen, B. J. (2018) Customer segmentation using online platforms: isolating behavioral and demographic segments for persona creation via aggregated user data. Social Network Analysis and Mining. 8(1), 54.


Scroll to Top