Do you build your Persona just by interviews? that may be a trouble!
If you build Persona just by interviews, It’s very dangerous and may mislead your decision making. You need quantitative data to support you!
The trap and deceptive words of UXer
I usually hear some UXer say: Since I showed the video recordings of interviews and usability tests to my boss. He/She never negated our decision again.
That make me think: Is it correct?
If a UXer just shows only the quality result of users to the boss but without quantity support, that may be a terrible blunder!
Sometimes, we just interview with few users. And then we build the Persona of product to make decisions. But, how much percent of Persona A? how much percent of Persona B? without data to support, we actually can not estimate which Persona is important and which is not. And then, the decision gets wrong.
What will be the next? Product manager then ranges the user story to developers. developers spend two weeks to complete the user story. And then we may found that, OMG, the Persona just occupied 1% of the market. The development cost the team a lot of time and money on the wrong dirrection.
We need to integrate data with UX(Pesona). But not only focus on empathy or convince our boss by UX skill(video).
The normal way to build your Persona
First, we can use the current information to sort out several kinds of Persona. And then we can interview these users, (at least 3 users of each kind). If the information we acquired is almost as the same as existing data, we can go ahead to make decisions by these Personas. And by the way, if it’s hard to sort out Persona, you can use the method “Persona Spectrum” to help you.
Why Persona may not work if we build it just by interviews?
The problem is that we don’t know which kind of Persona is the most important. We need to know the weights of percentage and how much they’d like to pay. And then we can make decisions of product.
By just interview with few users, it is not enough to satisfy the quantitative statistical standards. And, when the needs of the Personas are conflict, It’s hard to choose a user journey to follow (You know, user journey usually comes from Persona)
For example, we assume that if Google Map has the following three Persona:
Walker : The Persona don’t have a sense of direction. Use the navigation feature even the range that can be reached by walking
Bus rider : Take Bus to get to several places of a city. They usually use the planning feature to get the best shortest route.
Driver : They usually move a long distance. They especially need the information of traffic jam and resting stop.
So, if you are the product manager of Google Map, which needs of which Persona should you focus on? We know, There is a big difference between these three kinds of Persona:
Walker : They need more information about overpass and underpass on Google map and can get the navigation.
Bus rider : They need the accurate time of when a bus will get arrived. And It had better directly shown the bus on the map
Driver : They need more information about resting stop. like: how much to get there, rest room? restaurant? gas?
So, I guess you can find that, the three Persona want to add different information.
If we just serve for the one Persona and then we will give up the other two. Even we interview with several Persona. We still can’t make a correct decision which Persona we should focus on.
Quantify Persona can show you the correct decision
Let’s follow the same Google map case. Assumption that you get the number of users. No matter we collect the data through a questionnaire or a traffic tracking:
Now, we get the weights of each Persona. If you are the prodcut manager of Google Map. You can now make the decision easily. We can serve the most part of user(driver) or explore the new market(walker). No matter what’s the result, The quntity information do support Persona to making decisions.
So, how can we gather the data of each Persona? You can refer to my series article of Data x UX! I will continually share my experience of that.
Sometimes we don’t know the interview is just an extreme case
Let me use the normal distribution explain:
We can see, even if we already interview with 3 users of each type Persona. Sometimes some Persona is just an extreme case on the normal distribution. We can not know a type of Persona is the most case or extreme case without quantitive data.
Persona can help us to sort out few user types. But which is the most case? which is the most imprtant one? We can not know that if we don’t get the support of quantitive data!
The Conclusion
In the article, I want to share the story that a UXer cannot make decisions just by interview but also need data. We need both quantitative data and qualitative research. We can get a few types of Persona by interview, use data to quantify the result and then get the weights to help us to make decisions.
Extend reading
About the concrete way to quantify the weights of Personas, you can check that in the series of article. It’s all written by my working experience:
And, final, if the article is useful to you. please show me some applaud, thanks :)
