Personas and outrageous software – an interview with Alan Cooper

Download ‘printer-friendly’ PDF version (42 KB)

Listen (MP3, file size: 11.3 MB)

Cartoon of man looking confused by his mobile phone

Gerry Gaffney:

My guest today is sometimes referred to as the father of visual basic. He’s founder and chairman of the board at Cooper Consulting. He’s written two highly influential entertaining and quite provocative books: ‘About Face 2.0: The Essentials of User Interface Design’ and ‘The Inmates Are Running The Asylum’. Alan Cooper, welcome to the User Experience Podcast.

Alan Cooper:

Thank you very much Gerry, I’m glad to be here.

Gerry:

You’ve said that you were outraged by inferior and unusable software products. Can you tell me why?

Alan:

Well, let me first qualify it by saying that every product has software in it these days, and in fact every business conducts all aspects of its business by software so I’m outraged by everything. There’s no reason for software empowered systems and products to be hard to use the way they are, and we struggle with them all the time. In San Francisco, for example, there are hundreds of office buildings with thousands of offices and every single one of those offices – every cubicle in every one of those offices – has a telephone on top of the desk that is state of the art and virtually unusable, that is hated by everybody, and that is just ridiculous. That’s why I’m outraged, because bad hard-to-use software is so prevalent and it’s so easy to fix.

Gerry:

Now, you say that it’s easy to fix but sometimes the client will say to me “well we’ve got this application it’s inherently complex; it’s doing something that’s inherently complex, and we can’t actually make it simpler, and our users are just going to have to learn how to use it”.

Alan:

Right, they’re wrong. That’s just like saying “… we have a feudal autocracy and monarchy and we just have to keep the lower classes oppressed because they’re just, you know, not intelligent capable people and we have to oppress them. They like it.” I mean, people have said this kind of oppressive obnoxious untrue stuff for thousands of years and a lot of people believe it, but that doesn’t make it true.

Gerry:

Okay, I did want to get that little bit of background I guess about your philosophy in life before we launched into what I really wanted to talk to you today about. You’re known for several things but one of them is the concept of the persona, can you tell me a little bit about how your original idea for using personas came about?

Alan:

Well, sure. I like to refer to personas as the bright light under which we do surgery. I don’t want people to misunderstand and think that personas are in and of themselves design. I know that there a lot of people out there who create personas or think they’re creating personas and then they just do this business as usual and not much happens except they can now say they did personas.

If you’re going to do user centred design you’ve got to understand the user. Okay well there are two logical approaches to understanding the user… Well, there are three logical approaches. One is to say, “well I’m a user, so I’ll just do what I want”. Well that doesn’t work – that just gives you what works for engineers, because that’s called self-referential design and it doesn’t work.

The other one is to say “well, we’ll go out and we’ll ask all of our potential users what they want and we’ll just put it all in the product” and I call this the sum of all desired features and you see a lot of that in bloatware like you get from Microsoft and that doesn’t work either. It creates, well it can create, competent software but it always creates unloved software.

The third way is to say “well, we’ll just go out and we’ll get one person and give them what they want” and a variant to that is “we’ll just try to do something and we’ll change it based on when people yell at us”.

And none of those approaches work very well. What you really need to do is you need to go out and pay attention to the users, study them, understand them, and in particular understand what motivates them. But then what you have to do is you have to be able to model that in such a way so that you can use it for actual interaction design and that’s where the persona comes in. It’s a hypothetical archetype, it’s a distillation of what you’ve learned based on your field research of actual users, it’s a way to express what those users’ actual goals are, what they’re trying to achieve, and then personas become a very powerful malleable tool for being able to look through the eyes of your users and get out of the eyes of the engineers and the various stakeholders and constituents.

Gerry:

So are personas something then that emerged from your user research work or did you sit down and say “well I’m doing this user research and I’m getting good data back but I’m not really communicating it to the developers and the engineers as well as I might?”

Alan:

Well, you see I’m not a researcher. I’m not a scientist. I’m a software inventor and I spent many years inventing software and then when I got into the design business I asked myself a fundamental question, which is “how did I invent that software, did I do it because I was just a clever guy having clever ideas or was I actually using a system, and if I was using a system, what is it, what was it and can it be systematised and understood and taught, transferred to other people?” and I realised that in fact when I invented a design of software and then built it what I did was I imagined the user and I role-played who that user was and what they were trying to accomplish and personas are a formalisation of that what Christopher Alexander calls un-self conscious process.

Gerry:

So you’re externalising what’s going on in your head?

Alan:

Yes, and very much to my surprise and pleasure discovered that in fact it was a methodology that I was using and it is teachable and it’s remarkably powerful, anybody can use it.

Gerry:

Can I ask you how many personas are typically needed on a project?

Alan:

Very few. The thing about personas is they’re not so much things you make as they are things you discover and so if you go out and discover that there’s seven personas it means you do not have appropriate focus. You see, even though you have a broad constituency of users you should have a very narrow selection of goals because that’s what by definition a product serves – a narrow spectrum of goals – even if it serves a broad spectrum of demographics. So we find typically that a given product will have a an informative suite of about five or six personas of which we will focus our design on one or two. And if you find there are more then that, it means that you’re you don’t have an appropriate focus for your product, which is a very useful thing to learn.

Gerry:

Indeed. We do see people who’ve got like whole suites of personas and, you know, anti-personas and so on; sometimes it just seems to become a huge undertaking.

Alan:

Well, that’s not doing personas.

Gerry:

OK…

Alan:

I mean, they may call it that, but that’s not what it is. As I say, personas are not something we make, they’re something we discover. You know, Microsoft may create 200 personas for Microsoft Word but that’s because they don’t understand the process.

Gerry:

Now we see quite a few of what I think of as data-free personas, where people say “well we don’t have any time to do user research but we kind of like this Cooper guy and his personas idea so we’ll just sit down and have a brainstorming session and invent two or three personas”. What do you think of that?

Alan:

Well I think that’s a lot like having a Diet Coke and a Snickers bar for lunch. [Laughter.] You can tell yourself you’re being conservative by having that Diet Coke but you’re not. And the thing about personas is they need to be based on reality and they cannot be based on self-referential thinking. Now it is possible to build personas without a lot of research but what happens is the confidence in it you can have in the validity of those personas shrinks dramatically. The good news is that really good, valid, accurate, statistically useful personas can be determined in a remarkably short period of time, because they’re developed through a qualitative not a quantitative process. You think in terms of a research project taking you six months or a year to learn something; we can typically form a very robust set of personas in a month or less.

Gerry:

It’s often difficult to communicate to organisations that personas are in any way different from some sort of market segmentation representation. Can you articulate what the difference is? You’ve already referred to personas being goal-directed; is there anything else?

Alan:

Well, yeah, they have very different purposes. Not that marketing segmentation isn’t useful to interaction designers and personas useful to marketing people but they serve two dramatically different purposes. Marketing segmentation is a tool for understanding how you’re going to sell and communicate the sales process, okay, and interaction design is a product design process. They’re very different things. What interaction design does is we go directly to that motivation and not for why somebody would want to buy, but why somebody would be satisfied using, because that’s a deeper thing, it’s longer lasting. You know that old saying that quality is remembered long after price is forgotten? The marketing people deal with those pricing issues and communicating availability and sales but you can communicate very clearly to people something that they don’t want. I mean Yahoo! is in real trouble right now and they’re not in trouble because their marketing is failing, they’re in trouble because they don’t really understand what their product is, they’re not really delivering one that gives their customers value.

Gerry:

I believe that you generally use scenarios that you associate with your personas; within your persona development methodology you attach scenarios to your personas. Is that correct and if so how does that work?

Alan:

Well, design is an iterative process, but you don’t want to iterate in software because software is big and slow and expensive and iterating in software is like iterating in concrete. You get committed to something you don’t want to do, whereas in design you can keep playing different possible trial solutions over and over. So with a user persona you know who your user is, you have an archetype that represents the user and you know their goals, you know what they’re trying to achieve, not what tasks they’re doing, but why they’re motivated to perform those tasks.

Then what you do as a designer is you propose software solutions, behavioural solutions. If the person wants to make an appointment on time then this user needs – this user persona needs – something that will alert them to the appointment. Then what you can do is take your proposed solution and you can run a representative scenario, run that solution through the representative scenario by role-playing the user in the scenario. If the user achieves their goals then you know you’re on the right track with the design and if the user does a bunch of tasks that don’t achieve the goal then the scenario has just shown that your design is not yet good enough. Personas and scenarios aren’t design, they’re tools that you use to design. The design itself is the real solution, that’s what you do using your design methodology.

Gerry:

Is there a risk that by developing a set of personas, design teams or organisations will feel that that’s it, user research is a done deal and it never needs to be revisited?

Alan:

Yes.

Gerry:

Do you see that in practice?

Alan:

Yes.

Gerry:

[Laughs.] Do you want to expand on that or does yes sufficiently answer the question? It may well do.

Alan:

I think people in the software construction business in general are in denial about the responsibilities they have and they find lots and lots of different ways to not do what they need to do, and that’s why I’m outraged, and that’s why we have so many bad hard to use bad experience products out there, because the people who build software empowered products kid themselves. They kind of say “oh it’s easy” and don’t pay much attention to it.

Gerry:

That’s interesting. I was at a meeting the other night of the Interaction Design Association here in Hong Kong and one of the people who’s just come out of uni said “I thought usability was finished; I thought everyone knew how to do this and I’m really disappointed to hear you guys talk about how to sell the concept of user centred design into organisations”. Why hasn’t the industry evolved or why haven’t these bad practices been weeded out?

Alan:

Well, now, that is a big question. A lot of it has to do with the fact that software is a post-industrial craft and organisations. Companies today are managed using industrial-age management tools, and so what you tend to get is kind of an armed truce between business people and technologists, and the technologists find themselves being misunderstood by management and management in turn finds themselves being misinterpreted by technologists and so what the management tends to do is just kind of build a fence around the technologist. They don’t really want to understand it, or they can’t understand it and so what they do is they just limit technology. They just say “okay this is in general this is what you should do we don’t want to know the details but we want it all done in three months and under so many dollars”, and they don’t really get into the specifics, and they need to because it’s a very strategic thing these days. And by limiting the resources without really understanding what resources are necessary it becomes a kind of zero sum game.

The programmers kind of form their own alienated subculture and there’s very little communication, and the software construction subculture is a subculture of scarcity, they’re not given enough time or resources or support or facilitation to do their job and so they see anybody who comes in and gives them advice… they see that as threatening their scarce resources and so software builders, programmers and engineers tend to see interaction designers as competing for scarce resources. They see them as an adversary rather then as their biggest helper which is what they really are. And this kind of polarised work environment is perpetuated by managers who continue to think in terms of industrial age management criteria which really don’t apply in a post industrial age. That’s a big topic, that’s a big, big topic.

Gerry:

That’s also a very bleak summation I guess of the situation out there. [Laughter.]

Alan:

Well, the bad news is it’s very bleak the good news is the very bleakness of it will trigger the change. The thing is that in a world of exciting new technology around every corner, you can really ignore the human problem and just keep wowing people with the new technology. But I don’t need a more powerful computer. I don’t need a more powerful cell phone. What I need is a computer that doesn’t make me feel bad, and a cell phone that doesn’t make me feel stupid. And what that means is the technologist, the high tech companies can no longer rely on new whiz bang technology. I just got a new telephone and I specifically selected one with fewer features and less power and a smaller keyboard because I don’t want to be made to feel stupid by my telephone [laughs]. And I think this is what’s going to happen, is that high technology companies are going to discover that they really do in fact have to pay attention to their users and what motivates them instead of just wowing them with new technology.

Gerry:

Do you have any advice, Alan, for time-poor web developers who haven’t used personas before but want to explore their use without affecting their deadlines?

Alan:

I think that, number one, I absolutely know for a fact that planning saves time in execution. This is true in virtually every discipline. It’s been proven over and over again and it’s only some kind of a real strange reality distortion myth in the world of software that planning somehow is a waste of time. And creating and using personas is a planning methodology and it will save you time. Now it may not save you time in the design phase but it will certainly save you time in the programming phase and anybody who doesn’t see that is a fool, and I would just not work for them… or better yet what I would do is I would just ignore them and one of two things would happen – you get promoted or you get fired. If you get promoted you’re doing great if you get fired you know, well…

Gerry:

[Laughs.] You’re better off.

Alan:

Yeah, you are better off, really, because you don’t want to be working with those guys. And if you’re good at what you do, God knows it’s a sellers market these days in the world of interaction design. So if you get fired by a company because they don’t value your dedication to planning and personas in interaction design, you, my friend, will get job offers by the dozen in the industry. Call me.

Gerry:

Call you, OK.

… Alan Cooper, thank you so much for joining me today on the User Experience Podcast.

Alan:

Oh, Gerry, thank you for your great questions.

Published: December 2006.

A note on the transcripts

We make verbatim transcripts of the User Experience podcast. We then edit the transcripts to remove speech-specific elements that interfere with meaning in print (primarily space-fillers such as “you know…”, “um…”).