Delete personas from your portfolios
One thing I will never understand is why so many UX and product design courses have people new to the field create personas
Welcome!
In November of 2018 I participated in a 24 hour distributed hackathon. I built Degreeless.Design, a collection of resources, books, and articles that helped me along my path to learning design. It was everything I had learned in school on one page.
As of yesterday, a new Degreeless.Design website with even more content is live. This is also the inaugural newsletter kicking it all off. Showing behind the curtain a little bit here… I’m building a product design course. One that deep dives into the topics other bootcamps and courses gloss over. More on that soon…
One thing I will never understand is why so many UX and product design courses have people new to the field create personas. At best, personas are almost always unnecessary, and at worst, they are literally harmful to the product
The most common format you’ll see for a persona in a UX case study is often a proto-persona; one not based on research. These are the worst kind, but I’m not a fan of any.
For every step along the design process, you should be asking yourself what problem you’re trying to solve, and if the current actions you are taking are the best ways to solve those problems.
Personas are trying to answer who a product is for and generally “build empathy” for that person. Why is that a problem?
Biases and assumptions are baked in: Personas are stereotypes operationalized. They’re based on our pictures of people in our minds. All of the insights you pretend to glean from imagining conversations with your imaginary persona are based on your model of that stereotype of a person in your head.
Personas advocate for simple answers to complex questions: Yes, even if they’re backed by some ethnographic data. People are just more complex than that. The common use case for a persona is to pretend to ask it questions or to ask your coworkers, “Would Samantha want this feature?”. Personas do not truly answer those questions.
Superfluous information: I don’t care what kind of car User McUserface drives. Personas are often full of bogus made-up information that doesn’t do much of anything in helping me make design decisions.
Inefficient: Rather than spin cycles designing some persona template with drop shadows, skill graphs, and picking the perfect profile picture from Unsplash, my time would be much better spent defining the user roles, user flows, object models, and journeys users take through our product.
The alternative?
You may be saying, “but we must design for a person,” and you’d be right. You need to have a target in mind. Take the useful demographic information you might have used to create the persona and instead write it out. Don’t make a profile picture, ignore the bio. Just write it.
“Millennial moms with full-time jobs.” Awesome. “Product Managers at B2B SaaS companies with fewer than 200 employees.” Perfect.
Now you can…
Use that information to recruit participants for research.
Use that information to target ads or segment with A/B tests.
Define user roles and actors in the system based on these traits.
What’s a user role?
User roles are based on role theory and activity theory. Essentially, you take the actors in a system, the potential roles they can assume while acting in a system, and map them over time.
For example, let’s say you’re designing a task management app like Trello. The actors you have may be product managers, designers, engineers, and stakeholders. The roles the actors can assume in the system may be backlog groomer, ticket writer, and ticket mover. Each actor can pass through all those roles at different times depending on their context. Mapping these out against a user journey can help you decide where you could make meaningful interventions, bolster the experience, or flag things to watch out for.
TL;DR: Take personas out of your portfolio. They aren’t useful and, in my eyes, make your portfolio look much more junior simply by being there. Then replace them in your process with more effective alternatives.
Good read! I've always felt the same, personas have too much influence from the people who create them, not to mention the fact that there's redundant information in it.
I've recently started to use JTBD framework instead of personas and it has been a super helpful experience. Albeit it's difficult to understand first but once I got the hang of it, it felt much better than personas and also helped me not lose sight of the core problem.
Would love to know your take on JTBD!