Clients ask us all the time to help them with the right characteristics for personas–which are essentially market segments. We use what we call the A2G framework to help you think through all the characteristics that might be helpful:

  • Attitudes: What are the visitor’s psychographics? Sometimes called attitudes, these characteristics include Personality, Values, Interests, Lifestyles, Purpose, Feelings, and more.
  • Behaviors: What did the visitor do? These actions include search keywords, the timing of the visit, other sites they have visited in the past, behavior within the session, purchase history, lifetime value, and more.
  • Completion: Did the visitor complete a task? This includes conversions, but also the completion of other tasks, such as paying a bill, registering for an event, and many others.
  • Demographics: Who is the visitor? These characteristics include age, gender, income level, and lots more.
  • Entry: How did the visitor find your site or app? Did they come from Google, social media, an email, and ad, or a link from another site?
  • Firmographics: What organization are they from? The organizations’s characteristics include the industry, company size, and more.
  • Geographics: Where are they from? This can be a country, or much more specific, even down to a block within a town.

We focus mainly on these characteristics to target marketing messages, but they are also highly valuable for segmenting your analytics. Each time you run an A/B test, or look at your analytics, you can look at the results in aggregate, but sometimes it can be very helpful to break things down by these segmentation characteristics. Maybe your test “failed,” but it might be helpful to know that it failed miserably among men but actually worked for women. If you can’t personalize your approach to show something different to men than women, this information might still help you to decide what to try next.

If you aren’t segmenting your numbers, you are missing a key means of insight into your test results and other analytics.