I’ve been working on the Agile-persona issue. The way I’ve introduced personas into Agile is by segmenting/dividing/separating the personas by experience (i.e. the buy experience, the help experience, the download experience, etc.). This works well when we segment development by themes.

Cheers!

How do you make sure the agile pesona stays light-weight and can “be easily updated throughout the project?” Do you update the goals & motivations, add a new picture, swap something out?

I use personas to help with storytelling and to remind myself that I have different user groups and not to start blending interviews. Any specific advice for me?

I liked the post because it investigates agile and UX. You are right about not wasting time in developing detailed documents in a fast paced development environment. Who has got time for it?

However, I am skeptical about using real person as a persona. One downside you already pointed, which is getting biased by one person. The other downside is the name itself. One person on the team may like this “real” persona and other may not. These personal opinion about the real persona may end up in heated arguments and twisted design. Even clients has internal politics, which we regularly face as designers. I would rather stay away from this “real” person. Did you use it in your project? Was it a success?

The possibility of meeting the end users during the design depends on the project. For example, the project that I lately worked in agile mode barred us to meet end users because of client’s legal issues. All we had to depend on was Client’s SMEs, Market Research, and Best Practices. It was really difficult to work in such a situation. For that project we had to rely solely on fictitious persona based on the marketing data and SMEs input. For every sprint we would quickly walk this persona through the scenarios (functionality).

As you mentioned, our persona was very light weight with couple of bullet points and it worked well for us.

But let’s be clear, in an Agile model, it would be difficult for developers who are in contact with real users on a regular basis to do that. If all they are doing is simply collecting pictures and key attributes of the actual people they are in contact with, without necessarily thinking of or approaching them formally as Personas, that is nonetheless likely to still be a valuable document to have as a reference for when actual users are not around.

Looking back at my post, I realized it seems to make a point I did not intend to make: that a single real persona can replace a well-crafted Persona. Just like Personas are based on a small but carefully selected sampling of users, so too should Agile teams not be working with just one user, and I don’t think Agile teams are doing that. But after you have done some initial research with that sampling of users, traditional Personas tend to culminate there and sort of get frozen in time. In the Agile model, we have something you simply cannot get from piece of paper: A Living persona, i.e. a Persona seen over a period of time. And by that I do not mean just one person. Yes, you may work with one real user for a few days and then with another real user for a period. One of those real people, in my experience, tends to become an archetypal user, such as Barb who is an archetypal administrative asst. say. When we talk about “Barb” as a team, we both have a clear image of a real person with real needs, but we also understand that Barb is one of many administrative assistants. Additionally, when working software is released, every couple weeks or whatever, it gets used by and received feedback from a much broader group, which normalizes any bias issues.

… don’t get me started on designers who create “Personas” out of thin air based on imaginary users rather from actual research data. At that point, you may be creating a nice reminder that you’re designing for real people, but the persona itself is basically useless or even misleading

Fair enough, but what if the developers know that personas have value, but they don’t have access to UX professionals or researchers, whether because of management inertia/ignorance or budgetary problems?

Are you saying that their amateur efforts are worse than useless, that they are in danger of deluding themselves that they are doing something worthwhile?

Another thing I notice a lot is the word persona used to describe this kind of abstract, high level customer profile based on a job title, often transcending verticals, pay scales, etc…this seems to allow organization of content or themes for marketing to some degrees but in no way ensures experiences are optimized. Generally I have seen that the experiences are not optimized downstream in the waterfall when this forced abstraction is not exposed to real users or the process is not iterative.

And I think your nerdy plural usage is simply correct.