Building & presenting the personas
The main pitfall here for me would have been to build personas the way I used to do during marketing projects: group by age, income class, family situation, maybe profession or gender,… . This would have lead to marketing personas. Not that there is anything wrong with marketing personas, but it's just not as useful for design decisions as design personas. Moreover, 20 qualitative interviews is probably not the best way to build marketing personas either. Quantitative research would be more valuable in this case.
Building the Personas
So I focussed on the key information I was after: people's motivations, behaviours and attitude towards certain important project related topics. I printed out the notes I made with every person I interviewed and started coding them based on which behavioural and motivational profile they had.
- In a first phase, I used a colored marker to highlight key sentences or words that clearly defined the interviewees position on a specific topic (ex. Motivation, historical evolution, priorities, … ). I used 1 color for each topic.
- In a second phase, Is started grouping the pages ( = interviewees) based on how similar their answers were for a given topic. I did this for each topic and always drew a symbol in the top right corner of the page that would symbolise the category (ex. “$” if money is an important motivation). This made it easier to navigate and regroup in the third phase.
- During the third phase, I had to decide which topics were most defining, group the pages and then form sub-groups based on the answers (or symbols) for the secondary topics. Some odd cases didn’t strictly belong to one specific group, so I reread the interview answers and decided in which group he/she would fit best.
I ended up with 6 distinct clusters. Out of these I made my initial personas: I started from the defining topics (motivations, attitudes, behaviour, …) and aggregated all the other info (job, age, quotes, … ) into a coherent and representative persona.
Involving the team
Only at this point did I start involving other people in the process to create empathy with the users and personas. Most literature and experts advice to involve people as much as possible (designers, developers, sales, … ). They can observe or even conduct interviews, put together the personas in a workshop, … . I definitely share this opinion. For my situation however, I made a trade-off: I needed to get things done fast and there was no previous experience with personas projects in the company.
Furthermore, in all honesty, I had no idea how to get all this information from the interviews across in such a short time that would make it practical for a workshop. So when I had my 6 initial personas, I sat down with the project leader that would manage the further design and development of the service for which I was making these personas. We eliminated a persona that simply wouldn’t want to use our service (no use of continuing the exercise). We also deprioritized a persona (so we presented it, but didn’t completely work out the details of the profile) simply because we realised that if we could make a solution good enough for any of the other personas, it would be good enough for this persona. And we fused 2 personas because even after my grouping exercise, they still seemed quite similar. So we ended up with 3 personas and a 4th deprioritized persona. I deliberately wanted to keep the number of personas low to facilitate the knowledge transfer during presentations and meetings. We made an initial draft exercise of feature prioritisation per persona , to be able to show management what these personas could be used for . Lateron we used the personas to come up with new features we hadn’t thought of aswell as for a thorough feature prioritisation exercise.
Good or Bad practice
I used alliterations when naming my personas (ex. Pragmatic Patrick). There are pros and cons for doing this. A pro would be that it makes it easier to remember them (not unimportant when you want the whole team to use them). A con on the other hand, would be that it makes it harder to create empathy: the alliteration creates some humour around the persona and makes it feel less serious and reliable. In my opinion, the pros outweighed the cons.
I did use stock pictures to represent my personas, which is often discouraged! I did it anyway simply because I hadn't thought of taking pictures of my interviewees. I did invest some time in choosing the least stock-picture-looking photos though. Although it worked out fine, I'll definitely go for authentic pictures next time. It's really hard to create empathy for someone who seems to come straight out of a tooth paste commercial.
NN/g Conference London
Near the end of my project, Movify sent me to the NN/g conference in London for 6 days. I followed a lot of interesting UX research courses there, one of which was Kim Flaherty's (@kimmyaf) presentation on Personas. Although most of my work was already done, I still got some valuable advice on how to present these personas, how to use them to create business value and how I could further improve my work for projects to come.
A small sample of what we learned:
- Different ways of making personas & different types of personas . You got proto-personas and Assumptive Personas who are both light weight persona methods that don’t involve actually interviewing clients. Nevertheless, they still have the advantage of aligning the team and promoting user centricity.
- The most important building blocks of a Persona: the quote, a picture, biographical information to create empathy (should be more than just demographics, but can’t be to fluffy), behaviours and frustrations.
- Other building blocks: usage, domain experience, brand loyalty, devices and platforms, … . With these blocks you need to be aware of what is relevant for your context and what would only be extra clutter.
- Include a real quote that isn’t too long and really catches the persona’s spirit. Which worked very well to get the information across in my final presentation.
- Use personas to brainstorm & prioritize features, to analyse competitors solutions, to apply in story mapping and involve in the backlog.
Picture with Jakob Nielsen. Impressive how he keeps up this genuine smile although he had to pose for the same picture with 50 other participants as well.
Presenting the Personas
By this time I had done the real work of making the personas and was absolutely confident about them, but the hardest part still had to follow: convincing management that these personas where the right way to go and getting them implemented (where it usually goes wrong with most of persona projects). Luckily for me, making convincing presentations was finally something I DID have experience with.
I started by going over the case for personas again very briefly, just to get everyone back in the story. Next, I quickly presented a slide with my methodology and some metrics about the sample, just to get some potential questions out of the way before starting with the real deal. Thereafter, I elaborately went through the 3 primary personas, depicting them as people that could actually exist, people whose pains and motivations they could easily imagine. In doing so, I followed the advise of one of my mentors by backing up each persona with stories and anecdotes about the users behind this persona. Calling them by name and citing their quotes and behaviours drastically increased the empathy around the meeting room table and got people along. Finally, after tackling some questions, I finished my presentation with what's on every manager’s mind: the next steps! I even split them up in 3 levels: What we should do next, what we could do on top of that, and what we should do with future personas projects.
I’ve presented the personas at multiple meetings and people went along with the story, some even more enthousisticaly then I had expected. Management got convinced of the usefulness of personas and just this week, I’ve been given the go-ahead to do more personas exercises for other high priority projects.
I hope this post was of use to you, either instructive or recreational. Feel free to share and be sure to let me know your thoughts (positive or negative) and questions. You can contact me on Linkedin or Twitter.