Personas and the Advantage of Designing for Yourself

Obviously, being a freelancer myself this brings up a quandary. How can I do great work when I’m only being brought in for a portion of the lifetime of the web application? How can any consultant or designer?

We just closed a client relationship after completing phase 1 of a project for this very reason. There were a few snags that sapped enthusiasm for moving forward, but at the heart of the matter was a failure on the client’s part to understand that we had to have to more of a role and interest than just the current milestone, deadline and budget. We were almost dissuaded from caring about anything post-launch.

Thanks for the informative post, Joshua. I’m a highschool student who has done quite a bit of web work (design, development, etc.) and I’ve been reading your blog, 37Signals, Cooper’s books, etc. I haven’t been sure what to think about personas so I’m glad to see this clear-headed post. Though I’ve never used personas extensively, I could imagine that they would be useful in a project where I would not be using the product (i.e. a medical application) but I think I agree with you that, whenever possible, it is best to design for yourself.

Wow, Joshua, this was a great post. Lots to consider.

I think that what you’re saying regarding designing for yourself, while absolutely true, is a bit misleading. If I am designing an interface that I would use I don’t need a persona developed because I am the persona. I know myself, so why bother writing something down about my likes/dislikes, interests, usage patterns, etc.? So, the persona exists–it just exists in my head rather than on paper.

Yes? No? Maybe?

Hi Scott…good question. I think the answer is no. You are actually not the persona. If you’re designing for yourself, you know all of the relevant research in your head, so that part is right. But you don’t need to create a persona (which is a design artifact…a document) in order to help you design.

So yes, you have all the info you need in your head. No, you don’t have a persona up there.

Hi Joshua
As always, a fantastic post. Your writing and thinking is some of the best on the web, in my opinion.
You seem to feel that the contextual nature of personas is a poor argument for why they aren’t shared. But I for one probably haven’t shared personas for precisely that reason (plus that it didn’t occur to me).
Your persona document for Kelly, for example, contains about 50% content that could be used whenever the user is a marketing assistant, but the other 50% is specific to the particular website you are designing.
Maybe people think this specificity makes personas less useful (but not necessarily impossible to understand) when shared beyond the project?

Nice article Josh.
If only I had the luxury of working on websites I would use.
I work on a major business directory website and people like my mother would use it, if I designed it for me I’m not sure she would be able to find what she wants.
I agree with your view that Personas are a communication tool more than a tool to design by and I would also agree. And I think that this is reason enough to spend the time and energy on them. I consider the majority of my design work to be communication and personas assist in this communication.

OK, so here’s my question - in an era where we can reach out to a great many customers, potential customers, and existing community members, why do we need to invent personas? Why not pick a real person, have them meet the team, and then ask them questions when things popup.?

Gee, Jake. Good question!

There are several issues here. First, Cooper was writing before the social features you mention really took off. He was focused on desktop apps more than web, so that may be part of it.

Second, it might be a scale issue. Cooper was working on things like Visual Basic, which is huge. It might have been that there were so many people using it that he had to wrap up that research into something more tidy. However, I’m still vexed by the fact that people call personas by a real name and put a picture on them.

In my personal work, I don’t use personas. I do just what you suggest. I do interviews with real users, and when I talk about them with the design team I use those interviewees names. It’s not rocket science at all…simply saying “Yeah, Suzie complained of that…she said this…” is powerful. At this point it is important to track trends, but I’m not sure they need to be in the form of personas.

Personas is a useful tool, beyond any designers discussion. There are a lot of communication resources, where this one shouldn´t be left apart. It plays with the “identification” or “would like to be” phenomenon that everybody has.

So far I’ve also been doing what Jake suggests. In doing research on personas I ran across Jason Fried’s post. He suggests the same: “So if you can’t design something for yourself, design something for someone you know. Get that person or people involved in your project early on. Basing your decisions on a matrix of personality traits isn’t what I’d recommend if you really want to build a great product.”

I haven’t used personas, so I’m not really in the position to argue strongly for or against them. I think Terry Bleizeffer’s response to Jason’s post is interesting. While Terry isn’t a big fan of personas, she does clarify some crucial points. To me the most interesting is: “For most organizations, it’s just not feasible for every person in the organization to talk to real people and have the skills to successfully interpret what they actually mean and not just what they actually say. For the people who do talk to people and do have those skills, we need a way of communicating the results of those discussions to the rest of the team. Personas are one way to do that.”

The guys at 37Signals have an ideal situation. They are small. I think more (especially web related) companies should be like them because it gives them the flexibility to do stuff like… design without personas. But for big organizations with lots of stakeholders in a project, the designers need a way to communicate with the rest of the stakeholders why a certain design decision makes sense. The problem is that the engineers say “well, people really want feature x” and the designers argue that “according to our research, people want feature y” but there is no common ground to communicate WHO “people” are. Different “people” want different things and a line has to be drawn about which “people” the product is really for. Not every stakeholder (decision maker) can be in touch with real people , so personas are a way for those who are in touch with real people (and who have the skill required to interpret their research) to communicate to the other stakeholders who these people are, what they want, and why specific decisions should or should not be made. Again, I have no experience with large organizations or personas so I can’t say if the above observations are true in practice.

My latest project is great because the target market is a person who is living with me: my mom. I hope the projects I take on in the future will be similar.

Superb post! I myself have not used personas, although I was involved on the outer edges of a project that brought in a consultant who did use personas. I think it helped the consultant get an idea of what our needs were as far as website design. By communicating them back to us, the consultant could measure whether he understood our audience.

I just realized after reading your post that I have used the persona concept without actually writing up a persona. At least, I never write a complete persona (age and all that other demographic info). I jsut ask questions like, “Why do people visit the site? What specifically do they need to accomplish here? Where do they get stuck?” Some of that might be answered by mining the analytics for pages most visited and searches performed most often.

In my job, the boss would rather see quantifiable research — like how many people visit which parts of the website — and not a fictitious person.

Josh, I didn’t mean to imply anything - I still hear people talking about personas all to often these days. You’re right, it may have made sense in days past, but not so much any more.

In my experience, personas have been an exceptionally helpful shorthand for distilling the intractable mounds of research we often have on a brand’s target market or target users.

Whether you call them “personas” or not, at some point, you must make a judgment of whether a live interviewee fits the bill, and that is based upon a gut-level abstraction of your interpretation of that research or how literally they match an agreed-upon persona document.

Now, people are multi-dimensional fickle creatures that will have you tilting at windmills for weeks on end. And that is especially true if I’m designing for myself.

A persona is a simple way to place parameters on the target and avoid scope-creep because some friend-of-a-guy made a disparaging comment about your search field. A persona can be a reliable comrade, designed with the best available research.

And if you’ve done your work correctly, personas can be very human: totally flawed.

Personas are neither good nor bad, useful nor a waste of effort. They are another tool to use when, and if, you need it. If you don’t have the resources to conduct a wide spectrum user study or if you don’t have personal experience in the target group, personas may be useful.
Take the area of people with disabilities, for example.
If you are a designer with a disability,(raise your hand, if you’ve got one to spare), you will be able to readily relate to people with a similar disability. You may want to accommodate other characteristics of people with disabilities but don’t have the sphere of reference to make that personal knowledge determination. Here is where some type of personas may be helpful.
Most existing personas have some level of subjective interpretation and should not be used as the be-all and end-all of your design considerations.
Not all of your user base will interact with your design as you would expect. Acknowledging differences across your user base and being able to anticipate divergent user interactions will only increase the marketability of your product.

My own personal experiences using personas in an online social networking company are definitley mixed. Everyone agreed they were valuable but I’m not sure that everyone agreed on *why* they were valuable. I get the same feeling from reading your post. Even the experts don’t really seem to be in concurrence! I think what everyone agrees on is that personas are too touchy feely for the bean counters and too abstract for engineers. One thing I’ve noticed working in a number of technology companies is that business people and engineers seem to be trained to operate in a vacuum, outside of any customer/user influence. They are, after all, the experts. Creating a ‘fake’ user is an order more preposterous than bringing in real life users to tell them how to develop their product or what features to build. If anything I’ve found personas to be a great way for marketing and creative to build up their wall of B.S. Business people have the numbers and engineering has the holy grail, sorry, code.. now marketing can have their army of made up people.

OK, everyone - Alan Cooper only invented the term persona. We called the User Profiles for many years before he switched from Visual Basic to UCD/UX and adapated old tools with new names. In UX, everything is new again, all the time. But User profiles were around in the 80’s, and we used them to describe representative users in sufficient detail to support design rationale arguments to developers and product managers. We never used them as major design artifacts though, they are communications tools.

As far as the axioms of designing for yourself, it depends. It seems people in UX are often not trained in Human Factors, or understand the psychology of tacit knowledge. You cannot do knowledge elicitation on yourself, and you cannot measure your own responses to interaction. If you are considering product design, it helps to have separation and empathic understanding. If you are a designer, you are NOT an expert in your user’s work practice, but you can become a kind of participant observer if you are a good researcher. I design for doctors sometimes - I’m not a doctor, but have learned a lot about their work practice and everyday drivers and constraints. So I advocate research-design cycles so that designers can learn over time.

They’re also known as User Stories in Extreme Programming (XP).

http://en.wikipedia.org/wiki/User_story

With all due respect to Alan Cooper, who popularized and refined the use of personnas, might I suggest reading Henry Dreyfuss’s classic 1955 book, Designing for People?

It is really worth tracking down a first edition (on abebooks.com, for example, because of the phenomenal design), but it is still in print.

There you will find a non-dogmatic, pragmatic use of approach described.

There is no formula. Each problem demands its own approach. That is why I dislike blind dogma. And for me, going back more that 5 years in the history of design and UX is a good place to start. Seeing the true roots of the concepts today is enlightening, to say the least.

@Bill Buxton: Thank you!

Yes, there is no formula. All this talk of “Method X is right, Method Y is wrong!” has always bugged me, even though I myself have engaged in it sometimes. The real answer is to listen to design process success stories, going back fifty or even a hundred years, and to apply lessons learned to your own unique needs and processes.

(This, by the way, is why I think design/UX conferences should consist of nothing but case studies and why, also, I’ve asked recently for persona pundits to show us some real user personas. I’d rather see a real process in action than see a generalized — and inevitably idealized — “methodology” advocacy.)

I’m not surprised to read this comment from the author of “Sketching User Experiences”, a book that advocates a dogma-free, open-minded approach to the design process. You should add Bill’s book to your Abe’s order of “Designing for People”.

Love the thoughts in this post about the effect of “passion”, Josh. Agreed that nothing beats a situtation in which you have a sense of ownership and a passion for solving the problem at hand.

Note however that solving someone else’s problem can be tremendously satisfying too; in fact, passion can flare bright & hot when you know you can vastly improve improve someone’s life or their job through your work. Some of my favorite projects have involved understanding completely unfamiliar domains (e.g., protein research & radiology to name two) and producing designs that have a real consequence. That sort of work… well, it simply ROCKS!

Personas — or specifically the persona process — can really fire up that passion and provide a much-needed focus when you’re not designing for yourself. To the question of “Why create these fictional characters, why not just design for the real customers you’ve met in person?”, I suggest it’s a matter of focus. The hardest part about any strategy is deciding what you’re NOT going to do, and the process of building personas helps with those decisions. Essentially, personas are segmentation for designers: they remind us who we’re focusing on and who we’re not.

FYI here’s another example of a persona, from one of my past projects. The full story of how this work affected design & marketing appears in a chapter in User-Centered Design Stories.

This is a great conversation that addresses both sides of the issue. Nice to see some debate over UX processes going on.

To me personas are a great way to quickly gain focus and communicate if you are working in or with a large organization. However, I find that all too often they are used as a crutch and can prevent the designers from getting out and spending time with the users and sometimes they can get in the way of taking risks and innovating. We may be creatures of habit but thankfully habits can change.

I think if they are used correctly personas are a great tool but not the only one. Thanks for a good discussion!

I always enjoy using personas ass well. They really get you going but the difficulty is knowing when to let go of them. Nice article Josh!