You have the IDEA and you have the core TEAM, which hopefully includes a great designer (at least as a contractor, but preferably a co-founder or first hire). So now what?
BusinessWeek had a great post on Apple’s design process a few months ago. Here is a snippet from the interview:
10 to 3 to 1
Apple designers come up with 10 entirely different mock ups of any new feature. Not, Lopp said, “seven in order to make three look good”, which seems to be a fairly standard practice elsewhere. They’ll take ten, and give themselves room to design without restriction. Later they whittle that number to three, spend more months on those three and then finally end up with one strong decision.
People wonder why Apple continuously builds such great products. How is it possible? A big part of the answer is that Apple starts product development with pixel perfect mockups. Many would argue that the 10 to 3 to 1 process which lasts months is too cumbersome for your average startup. However, the likelihood that a random design will hit the mark is near-zero. Isn’t spending several years of your team’s time and [and possibly, your investor's money] on a crappy product quite a bit more cumbersome? As Microsoft has learned, it’s incredibly hard to fix bad interaction design after the fact.
You needn’t copy Apple’s approach “pixel for pixel.” But doing several rounds of INTERACTION DESIGN and VISUAL DESIGN mockups is well worth the cost. Just make sure that you hire a great designer who gets interactions. Interaction design does not equal visual design — but you can find individuals who get both.
Visual design is important, but the best overall design is the one that goes with the grain of human nature — that’s interaction design. And very few product managers and engineers get interaction design.
Once you have click-able (HTML mockups) ready, run a quick and dirty usability test. Invite friends over and ask them to do a few key tasks. Videotape their behavior. How many clicks did it take for them to accomplish your desired task? How can you improve your product by cutting clicks to achieve a task in half? How long did it take them to find the link they wanted on a page? How can you improve that? Then iterate at least once more…
Once you have something that you are proud of, you are ready to START writing code.
14 Comments
June 20, 2008 at 12:30 am
Interaction design is my favorite part of the development process. It is both a challenge and an opportunity to create a rewarding user experience.
I do not agree though with hiring a great designer who ‘gets’ interactions. This is a common misconception in my opinion, and you should get a user experience specialist instead - those two vocations are not as compatible as common sense would indicate (though there are exceptions).
June 20, 2008 at 1:13 am
When I was working for big companies and led engineering teams, I used a similar approach though no way near as rigorous as Apple’s. One thing I have learned is that no matter what you think the perfect design is, it will change once you put it in front of the users.
That said, frameworks like Rails MVC keeps the UI, business logic, and database layer separate so it is much easier to make UI changes with minimal impact to the rest of the layers. So for my startup I did some rough sketches on paper and then proceeded to knock out the code and made UI changes as I tested and tweaked it further.
I also agree with Eran about the designer vs. the user experience specialist bit.
June 20, 2008 at 2:49 am
Not sure what you guys mean by user experience specialist? How is that different from ID? Do you mean user research?
I agree that product launch is the beginning of the conversation with your users. But that discussion will be short (that is, not many users will join and those who do will likely bail), unless you get a good start.
To be clear, I’m not arguing for a laborious process. But in a 3 month march to Alpha, why not spend at least 2 weeks getting the interactions right?
June 20, 2008 at 4:49 am
I should have said usability engineer, especially someone specialized in HCI (Human Computer Interaction).
2 weeks interaction design upfront in a 3 months march to alpha is very reasonable. Actually I would argue for agile development process with 2-3 weeks iteration cycle, hence concrete deliverables at the end of each iteration. While the initial 2 weeks design is a good start, it is much better to keep all the key stakeholders (PM, Dev, User, etc) engaged all along the way till the Alpha release.
June 20, 2008 at 4:54 am
Thanks for the clarification Bob. I think we mean the same thing by great designer (me) and usability expert or engineer (you). But I would push to have the designer also handle user research and visual design as a group of 3-5 people cannot afford several people in the job… Many interaction designers have an HCI background; and good designers get user experience with or without a CS-HCI degree. Of course, if you can find an engineer who can do interaction design and code, that’s great.
Will discuss moving to releases every other week in the next post. But that doesn’t prevent you from kicking off the process (prior to writing a line of code) by assembling a complete set of mockups before kicking off product sprints.
June 20, 2008 at 5:01 am
After re-reading your article and following your link to “interaction design”, we are actually talking about the same thing, I thought you meant a web designer.
So we are in agreement
And it is important to have both interaction and visual design.
June 20, 2008 at 5:17 am
Regarding having the same designer with both HCI and visual design skills, absolutely agree, especially more so in such a small team. Heck, I ought to know, for my startup, I am the architect, coder, web designer, usability engineer, graphic artist, systems engineer, operations, business analyst, tech support, tech writer, etc.
Looking forward to your next post on releases and compare notes.
June 20, 2008 at 4:18 pm
I also thing it’s critical to focus on the data structures at the foundation of a product, and to concentrate on developing abstractions that will allow for flexibility. Utilizing fundamental object-oriented design principles early on in the product definition process can give both designers and engineers a solid, well thought through foundation to build upon. Rather than being limiting, I think this approach allows even more creativity, and brings up critical questions about the relationships between objects early on. Additionally, I’ve often seen this process uncover key features that had been previously masked by a “page mocks only” approach.
I’ve found that the challenge with this approach is getting designers interested. At the outset, it can seem very technical and rigid - but many engineers understand this stage naturally, especially architects. That’s one of the reasons I feel that designers with an understanding of engineering concepts are incredibly valuable. The best among them can design amazing experiences that are efficient to build and easy to change as the product evolves.
June 20, 2008 at 5:22 pm
Excellent comment Byron. I agree and think we’re in agreement. See my next post. I said that the mocks should be done before you start writing code, but you should definitely work on data structures, architecture, language choice, etc.
June 20, 2008 at 9:12 pm
Mike - You may have seen this already but there is a decent article in this month’s HBR on Design Thinking: http://harvardbusinessonline.hbsp.harvard.edu/hbsp/hbr/articles/article.jsp?ml_action=get-article&articleID=R0806E&ml_issueid=BR0806
It’s not web-focused but I think mirrors your argument here closely (particularly with prototyping). As the article points out, I think the key is to remember that good design & innovation is a process and not something that materializes fully-formed from nothingness.
June 22, 2008 at 8:00 pm
[...] that it should take 90 days to get from IDEA and TEAM to your first external [closed] Alpha. In Part I of The Product, I suggested leading with design. I do want to be clear that: (1) during the period of time when [...]
June 22, 2008 at 11:27 pm
@Mike,@bob:
What I meant by user experience specialist is someone who is an expert on user interactions and flows. A good graphic designer will have a good understanding of layout but not necessarily flows, and will often choose aesthetics over simplicity of interactions (at least, from my experience).
Of course the user experience guy should have a design background, but he must have user interaction background as well.
Some informative links on usability design:
Apple user experience guide - http://developer.apple.com/documentation/UserExperience/index.html
GNOME Human Interface guidelines - http://library.gnome.org/devel/hig-book/stable/
June 23, 2008 at 3:05 am
@Eran, we are in agreement. What I meant by interaction designer is what you are calling a user experience specialist. I was saying that getting user research, interaction design, visual (what you call graphic) design, ad hopefully a bit of web development (HTML / CSS) in one person is ideal for a startup. If you have to pick a single strength, my pick is interactions. So we agree violently
July 1, 2008 at 7:57 am
Actually, I think you need the designer back in the ideation phase, but then I’m biased; I’m a designer. A good designer will have a very empathetic customer-centric viewpoint which will be a nice balance to the engineering or business leanings that entrepreneurs generally have.
You mention moving design upstream before development is an worhtwhile investment. I argue you should push design (thinking) upstream as far as possible. The product vision is the foundation on which a design is built. It doesn’t matter how great the design is, or how well the code is architected; if the product vision isn’t solid, the product will fail.
Leave a Reply