We use cookies to optimize experience and functionality.  Learn moreContinue

How to create personas automatically

Buyer personas | 7 MIN READ
how to create personas automatically

In this article, we show how you can create buyer personas automatically for your (and/or your competitor’s) digital assets such as websites and mobile applications. You can use the deeper insights and intelligence thus gained to market, sell and serve your customers better.

Understanding digital users and their interactions

A foundational step for defining and delivering exceptional digital customer experiences for a business is a deep understanding of who its digital users are, their intent and their journeys.

Challenges with current tools

Traditionally, aggregated data from web/mobile analytics tools (such as Google Analytics, Adobe Analytics, Mixpanel, etc.) have been used to understand digital users.

web mobile analytics challenges

Utilizing such data for deep understanding of users and their behavior is often challenging for the following reasons:

  1. Since the number of dimensions and metrics of the aggregated data provided by such tools is large (often in hundreds), abstraction of insights manually is difficult and usually shallow.
  2. Aggregated data often treats the entire audience as a single group. In reality though, there are multiple groups/segments of users, where users within each group/segment show similarities in behavior. While manual segmentation is often available based on one or more dimensions/attributes, data at the segment level where the segments are automatically generated based on behavioral differences is far more insightful to understand variations across groups.
  3. Such tools, by default, capture and present analytics data that is agnostic to the industry or vertical that your business belongs to. In other words, whether a business belongs to the retail or healthcare industry, the set of dimensions and metrics presented by the such tools is the same and no industry specific insights are directly available.

An alternative to presenting the aggregated behavioral data using metrics and associated visualizations, as typically seen in analytics dashboards, is humanized views in terms of user/buyer personas. With personas, business/marketing owners no longer need to manually create mental models of digital users and buyers from dimensions and metrics.

Personas

By definition, user/buyer personas are fictional representations or composite views of audience segments based on various factors. They include inputs from customer demographics, behaviors, motivations, goals, data of existing customers, data from competitor’s customers, research, etc.

In an era where privacy of users is at the forefront, as is evident through ad blocker trends and increasing privacy related laws across the world, focusing on group behavior using fictional representations such as personas, rather than on actual individuals, enables business owners to meaningfully leverage data from their digital user segments to understand their jobs/tasks to be done, while honoring individual privacy.

Personas are used by organizations to market, sell and serve customers better. Humanized digital views using personas help businesses understand their digital users and their interactions better, deliver exceptional customer experiences and power growth.

Benefits of using personas

Personas help businesses with better customer relationships, marketing strategies, usability and consistency.

  1. Know your customers (Assimilate audience)
    • Gain clarity on audience, segments & their evolution over time
    • Leverage group and user level views for audience insights
    • Identify and understand niche audiences
  2. Drive top-line growth with better audience targeting and content strategies (Guide audience)
    • Get targeting ideas to expand business by contrasting desirable segments with other segments
    • Lower cost of acquisition with focused campaigns
    • Guide content strategies to build stronger relationships
  3. Refine engagement/conversions (Grow customer base)
    • Provide delightful customer experiences, online & offline, with unique insights
    • Help people make better decisions and deliver better customer success
    • Align customer experience strategy and digital execution
  4. Analyze competitors (Competitor intelligence)
    • Generate personas for competitors and discover trends as they happen
    • Reveal best strategies from the market via competitor intelligence
    • Uncover content, advertising and partnership opportunities/strategies
  5. Bring empathy to your data driven marketing efforts (Make better decisions)
    • Keep people at the centre of decisions & activities
    • Move from data dimensions and metrics to people
  6. Refine usability
    • Validate assumptions made during design on target persona and their likely interactions
    • Improve product/website usability and deliver engaging experiences

Some of the functional roles (with use-cases in parenthesis) that data-driven personas can be used by, include:

  1. Digital marketers/agencies (Automation)
  2. Content marketers (Content strategy)
  3. Designers (Design/UX)
  4. Product Managers/developers (User stories)
  5. Sales/e-commerce (Buyer persona)
  6. Recruiters (Candidate persona)
  7. Customer service (Customer support persona)

Specifically, in marketing, personas can be used to improve a variety of use-cases, such as:

  1. Targeting
  2. Recommendations
  3. Personalization/1:1 engagement
  4. Prediction/forecasting

How to create personas

Personas are currently created primarily by qualitative methods such as user research that involves interviewing or surveying users, prospects and/or customers. While such methods provide depth of insights such as motivations and challenges/pain points, they are neither easily scalable to millions of data points nor amenable to frequent, near real-time updates. As a result, persona creation tools today (such as Hubspot’s Make My Persona, Userforge, Xtensio, etc) are primarily limited to templates or presentation/visualization/collaboration tools that rely on inputs from the user surveys/interviews.

A viable alternative is to use quantitative methods that enable frequent updates and data inputs at scale as complementary means to creating user/buyer personas. In an era of rapidly changing user behavior, “live” personas that are updated frequently are needed to understand shifts in consumer behavior, their evolving needs over time and detect anomalies/changes as they happen.

Such quantitative data-driven approaches enable rapid generation with frequent updates of personas and use of data at scale. The resulting humanized views can help answer simple questions such as:

  1. How many types of users (user segments) does my website/app have?
  2. How would you describe who they are?
  3. What are the differences between users across segments?

Further, as machines are used to crunch data to create personas, industry specific insights can also be derived using deep libraries of domain specific intent.

Create personas automatically from digital data

Delve AI offers the world’s first software platform to automatically generate buyer personas for a given website/mobile application, business or industry from digital data. Personas can often be generated in minutes.

Compared to the conventional means of understanding users using surveys/interviews and/or analytics, Delve AI leverages advancements in artificial intelligence and machine learning to deliver insights that are deeper, easier, and more human.

Sources of digital data used as input to generate personas include:

  1. Web/mobile analytics tools capturing first-party traffic data such as Google Analytics, and
  2. Third-party tools that provide competitor intelligence and/or client panel data such as Amazon Alexa Internet, Ahrefs, SimilarWeb, SEMrush, etc. Such data about competitors can be used to create competitor personas.

Data from such sources are typically provided to the platform via ongoing programmatic access (using API/feed integrations, for example), as dimensions/metrics and may include historical/projected data.

Further, input data may optionally be filtered by one of many attributes to create narrower segments, including:

  1. Country/region/city/locality/postal code
  2. Channel/source/medium
  3. Pages/screens/content
  4. Device type/make/model
  5. Date ranges

A single persona can be generated for the entire audience (i.e. without segmentation). A preferred approach though is to create personas segment-wise with:

  1. manual segmentation using one or more dimensions, or
  2. automatic segmentation e.g. using behavioral segmentation. In Figure 1, we show a sample of a segment specific persona (summary view) generated by Delve AI platform.
sample of a segment specific persona
Figure 1: Sample of a segment specific persona

Figure 2 shows the detailed view of a sample persona generated. Attributes of persona generated, as shown, can either be inferred or be directly abstracted based on data. Attributes inferred and displayed may include:

  • Name
  • Profile avatar/picture/photo
  • Demographics
    • Age
    • Gender
    • Marketing generation (e.g. millennial)
  • Location
    • Country/region/city/locality
    • Urbanicity (e.g. semi-urban)
    • Territory (e.g. located in same city as the business)
  • Type: Business (B2B) and/or consumer (B2C)
  • Quote/Job to be done
  • Work
    • Company(Employee count)/Industry
    • Job function/Job title
    • Income
  • Household
    • Marital status
    • Family/pets
    • Home ownership status
    • Automotive ownership status
  • Communication preferences (e.g. phone, email, chat, social, in-person)
  • Brand affinity
  • Preferences
    • News
    • Television/Radio
    • Sports
    • Music
    • Travel
    • Entertainment
    • Food
    • Movies
  • Products and/or services likely to be purchased
  • Places likely to visit
  • Values
  • Hobbies
  • Tools used
  • Likely interactions (Acquisition, Repeat)
    • Device
    • Connection
    • Channel
    • Time/Day
  • Resources used in decision making
sample persona generated: profile
sample persona generated: interest
sample persona generated: interaction
Figure 2: Detailed view of a sample persona generated

Attributes analyzed and displayed when generating personas may include industry specific insights based on views/searches or other interactions. Figure 3 shows examples of inferred attributes such as apparel type and color for Apparel & Fashion industry.

sample set of industry specific insights
Figure 3: Sample set of industry specific insights for Apparel & Fashion industry

Our solution:

our solution enrich learn segment and humanize
Figure 4: Our Solution - Enrich, Learn, Segment and Humanize

Figure 4 provides an outline for a solution that can be used to generate personas automatically from digital data.

As shown, depending on the sources of data used to generate personas, such a platform regularly:

  • pulls the analytics data, usually in aggregated and anonymized manner,
  • enriches data for deeper context,
  • unearths behavioral insights with machine learning,
  • automatically groups users based on their behaviour,
  • abstracts personas for each of the segments, and
  • notifies business owners/marketing managers when changes occur.

Enrich: Augment data for deeper user context, with external/generated data sources/models, such as:

  • How: Query analysis, Internet Service Provider, Connection speed, Device features, Display size
  • What: Content analysis, Action/event analysis, Goals, Transactions
  • Where: Urbanicity, Territory, Climate zone
  • When: Weekend/weekday, Part of day, Holiday/occasion, Weather, Season
  • Who: Organization, Industry, Language, Translation
  • Industry specific insights

Learn: Inferred insights using machine learning may include:

  • Intent: Know, Do, Website, Visit in person
  • Decision phase: Research, Intent to convert (online/offline), Conversion

Segment: Behavioral attributes used for automated segmentation may include:

  • Engagement: Context, Intent, Actions
  • Outcomes: Conversions, Decision phase

Humanize: Abstractions for human assimilation and follow-up may include:

  • Personas
  • User flows
  • Funnels
  • Sample user/organizational journeys

visitor group classification
Figure 5: Visitor group classification

Another variant may include a step for visitor group identification before generating personas, based on profile, intent and behavior. As shown in Figure 5, users may be automatically classified into one or more of the following groups:

  • Business prospects
  • Job seekers/recruiters
  • Investors
  • Partners/competitors
  • Press
  • Service providers
  • Blog readers
  • Government

Understanding digital users: Comparison between traditional analytics-based solutions and quantitative analysis-based persona generation solutions

Traditional analytics-based solutionsPersona generation solutions
Data transformationRaw dataAugmented/enriched data
ProcessAggregate and tabulateLearn and abstract
RepresentationAttributes/dimensions + metricsPeople
VisualizationDashboards using numbersPersonas/journeys
GoalShort term trends/movementLonger term strategic insights

Conclusion

Finding buyer personas is certainly not easy. Currently, they are created primarily using qualitative methods based on user research. This can involve interviewing or surveying users, prospects and/or customers. While such methods provide depth of insights such as motivations and challenges/pain points, they are neither scalable to millions of data points nor amenable to frequent updates. As a result, persona creation tools today are primarily limited to templates or visualization tools that rely on inputs from the user surveys/interviews. Accordingly, improvements to the automatic creation of buyer personas are desired.

In this article, we have shown how personas can be generated automatically for your and your competitor’s business. With a platform like Delve AI, no code is needed and results are often available in minutes. You can use these personas to serve your customers better and grow your business.

Related articles

use personas for competitor analysis

How to use personas for competitor analysis

Performing regular competitor analysis is all about seeing where your company stands, where it could be and to identify opportunities to piggyback on the strategies of competitors. Learn how to conduct competitive research to build competitor personas.
What is an ideal customer profile?

What is an ideal customer profile?

Ideal Customer Profile (ICP) describes someone who would benefit a lot from what you offer and provides you with significant value in exchange. Having an ICP helps you with meaningful conversations and sharper campaigns. Learn about ICP and how to keep it updated.
View all blog articles ->

Subscribe for blog updates