Create, deploy, and maintain analytic applications that engage users and drive revenue. See a Logi demo

Tips + Tricks

Discovering Patterns: Working with Persona Data

By Mary Mahling | October 5, 2015
Share on LinkedIn Tweet about this on Twitter Share on Facebook

By now, you have done a lot of old fashioned detective work in order to get data for your personas. You have come up with provisional personas to help guide you through the process, and now you have data gathered from interviews with a cross-section of your users. Your data is most likely in the form of a stack of interview notes, hopefully in a searchable electronic format like Evernote. The challenge here is to get an overwhelming amount of data in a form that you can use to make design decisions.

A stack of interviews can be difficult to parse, especially if you’re used to building charts and graphs regularly. As always, your data is trying to tell you a story, but when it is all text it can be very hard to find your story. This is another place where your spectrums, described in the previous article, can help guide you.

If you remember from the last post, you assembled a worksheet with two spectrums based on two of the biggest questions you want to find out. If you recall from the potato chip example, those questions were:

  • Whether or not people prefer flavored chips over plain
  • If people are eating chips in one location or if they eat while they are on the go

After conducting 25 hypothetical interviews, you may find that your spectrum diagram may look something like this:

persona graph

If you have a diagram like the one above, it’s much easier to find patterns visually. You will almost always have one or two outliers, but in this example you see a trend: on the go consumers prefer plain while stationary consumers tend to opt for flavored more often.

The data will also tell you how many personas to create. There is no rule for how many personas is the correct amount – all of this depends on your data. Sometimes two personas is enough, sometimes you need six or more. Your spectrum can give you a suggestion if you have users clustered together, but it all depends on how many unique trends you see. Your data can also identify how many primary personas and secondary personas you need, if you need to get that granular.

Now that you have a view of your trends, you can now dig in your interview data to find the motivations behind preferences. You may find a pattern among the at-home consumers – they may opt for flavored (and messier) chips like BBQ more often since they have a sink nearby to wash their hands. You could find a pattern among the office consumers that shows that they go for plain chips because they’re eating chips with a sandwich and soda. You may also find a pattern among the on-the-go consumers in that they are more likely to opt for plain chips since they will be less likely to get their hands dirty when they eat.

Considering the user experience of a bag of chips, you now also know that your consumers are making their choices based on different motivations: the preference of a neutral flavor to accompany other food, and the preference to not make a mess of themselves when there isn’t a place to clean up. They are choosing the same behavior, but their motivations are completely different depending on their conditions.

As you sift through your data, as in your interviews you want to keep an open mind as you look for behavior patterns. If you are designing a business to business product for example, are executives using products differently than everyone else? Do users get interrupted often while they are doing their jobs? How much time do they realistically have to learn how to use something new? These contexts emerge once you consider a user’s experience in the context of how they use something and during an interview process (versus another method like a survey).

As long as you dig deep into the motivations for your users’ behavior and be open minded to trends you may not expect, you will discover all you need to see as you gather data on your users. Your data will always tell you a story, but having a good visual gives you a map to help you find the story you need to tell. Having a compelling story is the key to your personas’ success, although you need other elements as well. I will detail other elements you need in the next article.

Read the other posts in this series below:

Read Part One Here

Read Part Two Here

Read Part Four Here

Want to learn more about how to create the write user interface for your business intelligence and analytics applications? Check out our UI for BI webinar on-demand.

 

About the Author

Mary Mahling is a User Experience Designer for Logi Analytics, where she is responsible for user research and designing new products. She previously was on the User Experience team that was behind the award-winning responsive website for Medicare.gov. She has a Bachelors in Humanities from Harvard University Extension.

Subscribe to the latest articles, videos, and webinars from Logi.