7

I've read a lot about personas and have created personas as part of my job, but I am wholly unconvinced about their usefulness. My feeling is that they allow too much room for interpretation, assumption and bias and they tend to lack diversity.

Everything I have researched claims the purpose of them is to evoke empathy and to define goals, needs and pain-points. The latter three are actionable insight, but can a persona actually help someone overcome personal biases overcome in order to feel empathy for a certain person/persona? My gut tells me, no, that people will still be driven by their personal prejudices and assumptions no matter what the persona is. And that personas might, in fact, make more room for their biases in spite of the best intentions.

Are there better ways to achieve what personas were invented to achieve? Or am I misunderstanding their purpose and/or use?

10

Personas are both good and not good.

A few reasons why they are not good - from personal experience:

  • They represent research at a point in time and may become less valid as time goes by, and sadly there is often less drive to repeat the research or update the personas
  • They become a proxy for 'this is our audience' - an absolute - whereas things are always more complicated than that. Things change.
  • They can easily get forgotten - you have to work to make sure they stay alive and visible.

Note that the problems with personas are mostly time related. Personas age.

But there are reasons why they are good - again from personal experience:

  • Whatever happens to personas, if a whole team (or to some extent, a whole company) is involved in the journey of creating the personas, then that experience in creating the personas is at least as important, if not more, than any 'artifact' at the end.
  • Personas are (should be) the manifestation of solid research, and any solid research is good for providing evidence based decisions - provided the research was valid in the first place.
  • Personas help remind us that we're building products for human beings. I've seen and overheard developers speak about personas as something that suddenly helps them understand that they are making products for people. I've had developers quoting personas back at me. This is priceless!

Are there better ways? Maybe, but more to the point there are definitely other ways. Personas are not the only tools. They should be used in conjunction with other research, other evidence, other sources, to keep us as data informed from as many different directions as possible. Personas are one tool among many.

As UX designers, it's our job to keep personas alive, and that process helps us as well as our colleagues. Critically, it's also our job to continually critique the tools that we use and re-evaluate their suitability for purpose. It's our job not to take things for granted. Personas fall into that area, and if they become less relevant, or misleading, or appear biased then either they need updating, to continue being useful - or indeed as I think you suggest, they should stop being used.

However - on balance, I believe that personas, and the journey of creating personas has enough significant value that even if after some period of time they are discarded (like any piece of new tech) they will have usefully served their purpose and provided value in that time.

In fact I suggest that personas are created with an expiry date, because it sure is tempting to let them live on, and on, and on. And the more time goes by the less valid they become.

This is a sad day for me. As I write this, I realize I have some much loved but aging personas in mind, that I think need to be laid to rest. They were useful.

  • I don't agree with you last paragraph, we find them useful and an important part of UX research, as such I've voted for your answer, but with the caveat that I disagree with your final point! – DarrylGodden Jul 4 at 10:49
  • @DarrylGodden Thanks. I'm not really saying that they aren't useful, just that in writing this answer, I realized that our particular team has a set of personas that are four yours old, and have outlived some of their usefulness. I feel the usefulness of personas depends on the age of the personas, and the stage of the product. We either need to revisit or revalidate who our users are, or (my preference) focus more on an alternative tool to Personas. – Roger Attrill Jul 4 at 12:43
  • 1
    Ah, well, I agree with even more - they are generally only useful for the development task in hand, they can stretch beyond that, but people need to monitor against the latest user research as to whether they remain current. – DarrylGodden Jul 4 at 12:47
2

Personas are a tool. Like any other tool, if they're the right tool for the job at hand then they're useful. If they're not being useful, they're probably the wrong tool, or are being used in a way that doesn't fit the job at hand.

I've found them most useful when some of them are emphatically not the people you have on your team or readily available to talk to, or when I need to be able to differentiate between users when validating a design.

Like many other tools, they also need to be "calibrated" to suit the job you're doing. For example, most of the different personas I work with are likely to be the same people in different situations and at different times - and there aren't many of those people. So rather than having Bob the X and Sally the Y, I have personas based around "Frankie while planning for [job X]" or "Frankie while responding to [situation Y]". That might not be appropriate for others - and they might get more benefit from using personas differently or not at all.

Two other things worth considering:

  1. The difference between representative personas and aspirational personas - the first representing the users you have, the second representing the users you want. One stops you forgetting who you're designing (and building) for and the other gives you a goal to work towards. I've found this latter version particularly useful when it comes to diversity - if you occasionally design for users you don't have, it can make it very clear why they're not your users!
  2. Sometimes the finished persona won't be looked at ever again... but that doesn't matter. The research which goes into creating them will have helped establish empathy with the users involved, their needs and wants and what matters to them. Creating them drives research and forces you to listen to users, or at least to put time and effort into collating what you already know about them. The output is not necessarily the goal - the process is often where the value lies.

Edited to add:

It's also worth noting that a perfunctory persona without the research or insight to properly back it up is unlikely to be particularly useful.

1

Personas aren't everything

Having a generalized view of a cohort of users is a nice short hand. But it's just a way of aggregating user research to make communication easier.

Get real

Real value is hidden in atomic, real-world user feedback. I use Personas and Roles as tags on feedback received. When I'm honest about my tags, I often end up with two or more Personas to a user. People are more complex than Personas suggest.

How it helps

Along with many other factors, Personas help me filter down to groups of needs.

Of all the people like { this context }, there is a common desire to achieve { these things }

I can cross-reference Personas with other behavioral and business indicators in my real user feedback to

  1. Identify scalable areas of feature development
  2. Communicate plans to the business in succinct customer-centric terms
1

I have no use for personas, I think what began with good intentions as a way to help software developers think about customer needs has become the go-to for ivory tower UX. As often as I see them used with smug, alliterative names like “Mark the Manager” or “Linda the Librarian” and it seems like they put even more distance between the maker and the customer of the of what is being made. To quote Steve Portigal, it’s just too easy for UX personas to manifest contempt over understanding.

  • 2
    What do you use instead? To help define and align goals, needs, motivations and pain-points? (Which, of course, all boil down to features.) – MRL Jun 17 at 17:43
0

Everything that UX designers will do as part of their job involves communicating with different people. User experience itself is a rather intangible quantity, and much of our work involves trying to relate this to some tangible aspects. For example, who are the 'users'? It is an unknown entity if you can't relate it to something that you are familiar with, and personas are one way of putting something tangible to an abstract concept.

Personas were seen as a critical tool to engage with some of the business and technical stakeholders, who were more removed from the actual users and therefore find it more difficult to see the impact of their decisions (and how it translates to the experience for the end users). It was also seen an important tool for design teams to establish a common understanding on the focus of their design rationale and decision making process.

Furthermore, it can be used as training tool for UX designers in distilling user research information and translating it into actionable design decisions, and for building up a catalogue of information about users over the lifetime of a product or service. The emergence of DesignOps and ResearchOps requires a much more comprehensive and rigorous approach to capturing data about the users, so some of the artefacts (like personas) have been superseded by the availability of direct data captured from apps and correlating relevant data to extract new insights.

I think a tool is ultimately as useful as:

  • the information used to create it
  • the appropriate use of it to solve a problem
  • the continuous improvement or adaptation to suit your needs
  • knowing when not to use it and why

So it is up to you to determine whether it still has relevance or utility to you, but remember that you can use a good tool poorly or a bad tool well and end up with similar results.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.