In “Actually, You Might Be Your User,” Jared Spool looks at the pros and cons of self design. That’s designing without relying on things like user interviews, contextual inquiries, surveys, card sorting, and usability testing.
He examines how 37signals gets by without using these things.
They design from their own perspective. If they want to add a feature, they look within themselves to figure it out. No usability testing or contextual inquiries needed.
Jared then looks at the pros and cons of working this way.
Another big disadvantage of Self Design is it only works if the designers use the product a lot. 37signals built Basecamp because they needed to manage their own projects. They use it every day. Their product, Campfire, is their main communication method, since they have team members all over the world. Apple’s team uses their phones every day, all day long.
When there is something frustrating that happens in the daily use of the design, it surfaces pretty quickly. The designers themselves experience that frustration and, because they control the design, focus on eliminating it.
Made me think back to the old adage “write what you know.” If you’re an Eskimo, it’s going to be tough for you to write a book about growing up in the Italian countryside. You’d have to spend months (or years) learning Italian. You’d have to learn the idioms. You’d have to visit your chosen locales repeatedly. You’d have to meet lots of locals and ask them questions. Even if you do a great job at all this research, no one will be surprised if the end result still winds up kludgy and full of mistakes.
Or you could just write a book about growing up as an Eskimo. Then you bypass all those discovery layers and just get to the doing. You already have the knowledge you need.
The same thing happens when you design what you know. You get to bypass the “learning the language” phase and get right to the “building something” phase.
I know it’s not always possible, but, when it is, pick something to work on that you’re around all the time. Something that bugs you. Something that you’ve been paying attention to for years. Solve a problem that you yourself experience. Design what you know.
Craig Bovis
on 26 Jul 10I would say this goes hand in hand with “Do what you love”.
joe larson
on 26 Jul 10There are other disadvantages in self design in terms of software dev (though I don’t think they come close to outweighing the advantages).
Programmer’s are not very representative of the population at large in terms of how they think. So a programmer’s self-designed todo list, personal finance tool, etc., may make perfect sense to someone who has a technical mind, but not so much to the general populace.
Next, it oftentimes people have a hard time systematizing and analyzing something they know organically. They know too many details, have seen too many corner cases, and see too many cross connections to distill things down to a manageable design.
Also, when a user complains about something missing or broken, the self-designed designer reflexively thinks it is the user who is the odd duck, since the designer has never run across the need or problem themselves.
With some extra effort, an understanding that not everybody thinks like you, and a diverse team (including non-programmers obviously), these can definitely be overcome. But they are some pitfalls to watch out far.
Peter P
on 26 Jul 10This all rings a little hollow for me. By making technical support duty and customer blog monitoring mandatory for the development folks there, you really are assessing and verifying your design decisions. I guess you can keep on patting yourselves on the back for “Self Design” but I don’t think that’s really what you’re doing. You’re collecting customer input, but you’re calling the activity something else.
Jeff Croft
on 26 Jul 10Who said anything about programers? You’re right…programers alone wouldn’t make a very good group of self-designers. But the companies mentioned (37signals, Apple) certainly employ a lot more than programmers.
ML
on 26 Jul 10@Peter P: You’re collecting customer input, but you’re calling the activity something else.
That’s true once our products are out in the wild. But when we first design a product, we do go the “self design” route.
Peter P
on 26 Jul 10I really don’t get it, ML. I don’t believe it’s possible that you can (or would want to) separate what you’ve learned from your customers when designing something “new”. Furthermore, I don’t know why it’s so important to you guys to be known as “self designers”.
ML
on 26 Jul 10@Peter P: Actually, we don’t really care what you call it. (Never heard the term self design before this piece.) Just explaining to you that when we create a new product, it’s based on how we think should it work. After it’s released into the wild is when we start soliciting feedback. But yeah, we certainly are factoring in what we’ve learned from customers in the past when designing something new.
Steven
on 26 Jul 10Great post. Something we’ve been considering for a while.
We recently had someone ask us to develop one of our apps into a new category that we do not have any direct experience. (E-commerce)
The potential is huge, but so is the chance that we are not experienced enough to build the best solution.
Any thoughts from the outside looking in?
Jonathan
on 26 Jul 10Whether or not “self-design” is any good will be determined by whether anyone wants to use your stuff! Well, either that or it depends on whether you can persuade people to think your stuff is great (AKA the Microsoft Way).
But it seems a rather silly article by Spool because 37S are demonstrably still alive and kicking. In any case 37S’s technique is impossible to critique because it’s self-fulfilling. All manner of stupid ideas can be magically justified as being just fine because, well, the designers had those idea so they are automatically OK because they are their own target audience. Better than being slaves to idiotic hoards of users – right? Ha!
Personally, I found Basecamp (admittedly the only product of theirs I’ve used) frustrating and difficult to use on the occasions I’ve been made to use it. The last time was about a year ago when my OpenID login suddenly stopped working, and other bugs.
Jemmy
on 26 Jul 10“Write what you know”, as intended for writers (of fiction or nonfiction), is often derided by writing instructors, because, ironically, you often don’t really know what you think you know (or you just know it for you). Your assumption that you “know” something lead to other, often false assumptions, and might not foster critical analysis. On the other hand, writing what you do not know can result in a gathering of knowledge, a request for opinions, an open mind, and a forming of opinion based on empirical data (at least, that’s the ideal), which sounds like good UX to me.
As a writer and UX developer, I dread the day I am told to go “build what I know”. There is no “knowing,” only a steady movement towards learning and thus, knowing more. The only way I know of to do that is to start a project believing I don’t know anything.
Guy at HockeyBias dot com
on 26 Jul 10In general, Dave Winer is an advocate of this and so am I…
justin
on 27 Jul 10This is the worst idea I’ve ever heard of in my life. Anyone who has used SuperMemo for more than a few days can confirm (a memorization software tool under continuous development since the 80’s). The genius author Dr. Wozniak uses it himself daily, but nobody else in the world has a complicated enough mind to understand anything but a subset of what it can do
Sridhar
on 27 Jul 10But instead of having Beta customers, you do all the activities as mentioned within the group right ? ” user interviews, contextual inquiries, surveys, card sorting, and usability testing.”
The essence of Beta customers, is to help you get their PoV or domain expertise to your product. The world might be a dull place if mechanical engineers start designing CAD apps :)
Gerald Irish
on 27 Jul 1037 signals can get away with this approach because of the business you’re in, software as a service on the Internet. If there’s a major bug or users are having a very hard time with some features you can make changes and do another release. The cost of doing so wouldn’t be much larger than more development hours.
But if you were selling a hard product like a consumer electronics device or a car you couldn’t afford to wait until after you’ve delivered it to customers to do end user testing. The cost of re-tooling and doing another version is high, and the cost of replacing poorly designed product that has already been sold is usually VERY high.
Companies can get away without doing usability/user testing if they’ve got really good designers, especially in the software industry. But that doesn’t necessarily mean it’s a good practice, it just means you can get away with it.
Arabella
on 27 Jul 10I respect your process – and your success bears out that it works. And I also get the “write what you know” analogy. My only question is…what happens if the cool/smart kids only design for other cool/smart kids because they only know about being cool/smart kids?
I understand your business model and why you do what you do. You aren’t out to solve the world’s problems and there is no reason why you should.
But, somewhere in all this talk there should be acknowledgement that there are people out there who need good design as much as – probably MORE than – business people and other designers. You know..the folks who don’t have a lot of education and need healthcare info for example. The process of designing for these groups is just as important and is possibly more difficult because of the cognitive gaps between people like you (great designers and coders) and them (regular folks).
Or is this going to be the next stage of the digital divide?
Alexander
on 27 Jul 10This eskimo love learning about the Italian countryside.
As a career user researcher, I’m fired up by learning about how other people live and relate to their world (and the technology and products within it).
Mert Torun
on 27 Jul 10Self-design is what people have been doing for themselves since some hominids started carving axes from flint. When people have an itch, they scratch it, and who better than themselves to find the exact itchy spot. Same is true of programmers, when they have a problem of their own, they go out and solve it using software they make, and chances are that it will be also good for other programmers.
Solving a problem that you don’t have yourself is much harder however. When you try to solve somebody else’s problem, the focus tends to shift from really solving the problem to finding a passable (but not necessarily useful) solution and being done with it because “your problem” isn’t the stated problem, your problem is finding a solution to the stated problem.
It is a meta-problem that you end up solving. A derivative of the real problem. It is not the real problem.
Avoiding this perspective shift calls for some serious insight or quality training in one of the design fields. Actually, this may be the single point that defines all kinds of good designers; that they are able to understand, internalize, define and solve other people’s problems.
Going back to programmers doing self-design, yes it works nicely, because it fits the “scratch your own itch” bill. However there are many more problems out in the world that could be solved with good software, yet they have to make do with mediocre or outright bad software, because they happen to lie outside the programmers’ domain of experience. They get the meta-problem treatment.
I’d say the world would have a lot more good software if only good programmers would team up with good designers more often. The results are outstanding when they do.
zhanxiaoli
on 30 Jul 10Finally, try not to buy ugg correct a wrong had now. emphasizes moderation in Chinese philosophy, winter boots Ugg for Women, because the doctrine of the ends are very harmful. In the process of medical reform, the allocation of government resources and the allocation of market resources is the combination, but this combination should be by cheap thomas sabothe guiding ideology contrary to economic reform, the government should be raising and allocating directed by medical resources. medical reform, not just the reform of hospital, public health system, health insurance, health system reforms, you can use the market approach and to what extent the allocation of resources depends on the total capacity of government regulations.thomas sabo schweiz
This discussion is closed.