Personas Didn’t Work at My Company
Personas are a popular design tool that have been shown to improve the quality of designs. We dreamt big for them at our company, imagining everyone from sales to engineering using them. It’s been two months since then and it has become clear that they have not been as successful as we’d hoped. The troubles can be broadly grouped into trying to appease too many people, and not selling them well enough in the company. Despite all of that, creating them was definitely a great process to go through to learn about our users.
“Everybody wants something different”
I was working closely with an ethnographic researcher on the personas. She had started building some personas long ago, but had abandoned them due to a lack of interest. One of the higher-ups decided that they should be a priority, so we got to work on them again. We started talking to different teams about what they wanted from personas, and we got a ton of different answers. There were a lot of stake holders and they wanted very specific things.
- There was concern about personas being relevant enough to our teams. Some coworkers wanted to make sure that the personas were centred around how they used our product. As such we focused more on making them relevant to the product than making them a direct reflection of our customers. We ended up having one for each product, which essentially turned them into a synonym for the words “the user”.
- Former users of the software who now worked at the company had strong ideas about the user’s goals and daily work flow. This influenced how we interpreted some of our research around customer habits. We validated what our stakeholders suggested, rather than discovering.
- Some stakeholders had seen personas at previous companies and expected various features or formats. These became a laundry list of required elements. No one request was too much on its own, but having so many meant doing a lot of work to make everyone happy.
- One of the largest issues was that the company simply changed its target market, causing our personas to be outdated and less relevant.
“Why should I use this?”
I’m still proud of what we created. They were solid personas that conveyed a lot of information about our users. That being said, we could have done a better job of communicating them and their benefits to the teams that we worked with. Specifically the developers and quality assurance folks were confused about how to use personas, and the product teams had various complaints.
A large problem was that I underestimated the challenge of educating people about personas. I sat down independently with each QA and developer on my team to explain what personas were and how to use them. It was tough to explain exactly what a persona was to someone who had never heard of them. Making matters worse, I often didn’t have a great answer for how team members could use them. This lead to a lack of excitement around using the personas.
The product managers had a few more problems. Some product owners didn’t think personas were useful, questioning how a fake human could help them create better software. Others felt that they already understood their users well enough to be able to make these decisions on their own. Finally, some people just had different ideas around software design, wanting to design software for how the user should work.
What are the solutions?
I don’t know what the solution is, but I do know what I’m going to try differently next time.
Consult, but have someone in charge
It’s important to talk to a lot of stakeholders. It helps people feel involved and will make selling the completed personas easier. However it can lead to a laundry list of features. Remember that you don’t have to include every request that you get.
Make them an accurate reflection of the user
There will be many different people with many different ideas about who your users are. However, if you’ve done your job right, you have solid research to back up your personas. Only deviate from your research for a good reason.
Make sure they reflect the correct market.
We researched our users diligently and created fairly accurate personas (even if they were biased towards our products). Then the company reoriented towards different product segments. Make sure your personas match your target market.
People don’t get excited about new processes without some leadership. We all know you have to sell your designs, but it turns out you also have to sell your new business processes. Get people excited by explaining how the personas benefit them.
Give people practice using them
Personas on their own are a great communication vehicle. They are even more useful, however, when they’re used and referred to regularly. Group activities can help give people experience in how to use personas. Activities could be things like writing some user stories from the persona’s perspective, or talking about which feature you worked on that would have the biggest benefit on each persona’s life.
While our personas are only being used within the design team, a lot of the work that we did to build them has lived on and been quite successful. Our research discovered a lot about our users and how they work. This information has been communicated to the company in many other, non-persona, ways. I definitely still think personas are useful, but next time we’ll have a better idea of what we need to do and what to expect for them.