Why shouldn’t I kill personas?

In search of the answers why I should love personas.

Over the last few years I’ve become more and more sceptical on the value of personas. I know they’ve always been a popular part in the user centred design methodology, which kind of means that they are holy. But I also believe that from the moment they were introduced they were also misused or based upon the wrong data. For me this was the moment to call in the help of a few experienced UX friends. Why should I still use personas?

For the discussion I e-mailed Steve Baty, Adrian Chan, Will Evans and Dennis Koks. This resulted in a four-day long mail discussion. Below is my translation and interpretation of that discussion, which proved to be of great value for me. Note: right before I finished this article there was an interesting discussion on the IxDA discussion list.

The way personas should be used

Before we dive into the discussion it’s important to look at personas from an encyclopedic perspective. What are they and what are they supposed to do? According to Alan Cooper, the grand creator himself, personas are and were originally intended as a heuristic. They are for client communication and orientation. As Adrian Chan states it “they encapsulate and personalize use cases.” To be able to do this you need to have quantitative and qualitative research data, which can be analyzed thoroughly and translated in worthy archetypes.

How are they often used?

Unfortunately the textbook description of personas is too often a myth. Everywhere I look they are NOT based upon thorough research and mainly a tool to please the client. Will Evans says “that a great many advocates of personas simply don’t get to do them – or do them right – because no money is allocated for design ethnography – and even if there is money dedicated to research, practitioners don’t know how to take that research and turn it into personas which are actionable in the actual design process.” I think his remark hits the spot. Maybe big companies such as Microsoft, Philips and Nike have huge budgets to do proper research, smaller companies and design companies often don’t have this luxurious position. But this doesn’t take away the need for proper data. We need to look for possible solutions here, which isn’t coming up with personas based upon assumptions and ‘experience’.

Adrian Chan, in my opinion a pioneer on the field of social interaction design, states “I’m all for a move among social interaction designers to replace cardboard user peronas and instead use psychologically-grounded personality types.” This could prove an interesting improvement. According to Adrian these personality types should not become archetypes, since these are usually not interaction oriented and not specific to social media. It’s an interesting thing to try and understand the personality types, expecially when they are based upon actual psychological research.

So at this point the scepsis was still in my head. Fortunately the discussion gave me three useful reasons why I should continue with personas. Let’s check them out:

1. Great for design decisions

One of the biggest advantages of having personas, based upon good research, is that you can use them for design decisions. As Will Evans puts it “Someone is going to make decisions – often a vocal stakeholder who thinks they know a lot more than they actually do – the VP of Marketing that wants to prioritize a certain feature or piece of functionality because “My daughter is our target audience – I know my daughter – so we should do X” and if a designer has no data, no research, and no personas derived from that – no matter how bad the idea – there is no way to argue against it. User research and resulting personas helps in making prioritization decisions about initiatives and features based on actual users as opposed to the whims of:
1. The designer with their own prejudices
2. The powerful stakeholder
3. The developer
4. Anyone who is not the intended audience”

2. Great as a debrief / briefing

When working for a client it’s really important to dive into their world and prove you really understand it. Personas can help you with this. They form a good summary of the target audience and are a great debrief for your customer. After presenting them I always start a discussion, trying to see how their world matches with the personas. Being truly interested in their audience, but also in how the client perceives it is worth so much. It brings you on the same level as the client, making them feel you really care about their message/product/service.

Another great thing, besides debriefing, is actual briefing. When fresh team members join the design team personas form a perfect guide into the project. They are a nice summary of the data, which gets the fresh people up to your knowledge level faster. This is the real goal of personas: being a means to translate and transfer research and knowledge to others. As Steve puts it “The persona provides us a way to transfer some of the value we derived from the research & analysis process to other members of the team. It isn’t a perfectly efficient transfer – some of the value is lost. That’s what we’re doing when we communicate *through* the artifact: transferring that knowledge/insight/learning out to people who couldn’t be involved in the process of producing it.”

Remember to approach personas from this perspective and it will help you make them more communicative. They should tell a story for themselves, dipping you into their mental and physical world.

The end result and the data is far LESS important than the immersion in the experience of the target audience – Will Evans

3. It’s about the process

I always learned to save the best for last. So this third reason was the one that made me a believer again. It was a quote from Steve Baty that made me see the light: “Personas are an outcome; not the process.” Everybody is staring at the outcome all the time, using the personas as if they are treasures… but that’s not it. It’s the process itself that’s priceless. Personas are only a tool that force us (designers) to do actual research. It’s a means to make us submerge in the world of our target audience and truly, deeply understand them. It helps Dennis Koks “to create a better understanding, and it helps [him] interpret the outcomes better during the rest of the design process.“ Steve Baty states that “only some of the value of personas is encapsulated in the end result; much of it comes from the exploration of the data itself.”

Personas are an outcome; not the process – Steve Baty

In the IxDA discussion Dave Malouf said that “if you are selling Personas then it seems that that is your first mistake. Sell research and don’t even tell people how you are going to model it. Maybe the research itself will tell you the appropriate way to model your analysis.”

So immersion is the key. In my mind it has even become one of the few user centred design approaches each designer must undergo. Will even wants to go so far “as a designer the whole goal is a violent sense of empathy for the user in their own context – the data in this case and the persona are completely useless if there is no bond with the end user. Design insights that were not directly experienced by the designer are worthless.”

“This raises an interesting question: do we need to carry out our own research? Is it enough to analyze the data? If we get a series of ‘design insights’, how bad is that?”  – Steve Baty

As a closure (in other words: I’m starting to repeat myself)

A thing about personas which, in my opinion, is still overrated… Is the value throughout the project. It is often believed that after personas are created they will live on and should come back in all discussions. Even when I’m typing this I want to believe it. I want to put their portraits on my desk and always use their names in discussions, but it almost never works out like this. For me they are important because they make me do research. When I create personas I have to force myself to trully understand the user and context. It makes me want to collect all the data and the translation into personas is more important than the end result. And after that the presentation of the personas to the client also helps to show and test if you really understand their customers. It’s a great starting point for trust with the client and forms a good discussion. After this I think they are less important for the people who actually created them, since they already submerged… they can use them as a reference, reminding themselves sometimes. But they are a great tool for new members of the design team.

I wish to thank Steve Baty, Adrian Chan, Will Evans and Dennis Koks for a great discussion. And Dave Malouf for stealing his quote. In my opinion this discussion is what Johnny is all about.

Images by moonsoleil & randy son of robert

Jeroen van Geel

Jeroen van Geel is founder/chief kahuna of Johnny Holland and the interaction director at Fabrique [brands, design & interaction], a Dutch multidisciplinary design agency. You can follow him on Twitter via @jeroenvangeel.

10 comments on this article

  1. Great insights!
    I personally also lost part of my faith in Persona’s.
    When I’m submerged into the research and a design process I find it’s hard to see the true value of persona’s, but it is there!

    It reminds me of this paragraph of Getting Real which talks about ‘Dead Documents’:

    Making things like persona’s sometimes make me feel like making a dead document, but as you can read above, the true value is the process.


  2. Phillip on

    It’s good to see articles like this that do not have at their center a desire simply to sway opinion. Thanks for your willingness to challenge yourself and others and arrive at a place that perhaps you didn’t expect to.

    I am glad this is what Johnny’s up to.


  3. Pingback: peterpixel writings » Blog Archive » Arguments for the Usage of Personas

  4. @rowan Great link! At Fabrique we’re also always trying to evolve our proces, killing documents that are dead from the start. But it is always a difficult decision at the start of a project, since all projects are unique.

    @Phillip Thanks, we try to do our best. We’re always thinking on how we can make more of our articles a starting point of discussion and curiosity for our readers. I think the most important point is that for the writers themselves each article must be a search and adventure, stimulating their senses.

  5. Pingback: Coffee is for Closers. Ought-Nine.

  6. Pingback: Personal Card Set › Wireframes Magazine

  7. Pingback: Design Ethnography & Mood Maps

  8. Pingback: Johnny Holland - It’s all about interaction » Blog Archive » Design Ethnography & Mood Maps

  9. Pingback: Design Ethnography & Mood Maps | UxSoup

  10. Pingback: Personas are like statues « Jakub’s Thoughts