Editor’s Note: This post was originally published on Seattle 2.0, and imported to GeekWire as part of our acquisition of Seattle 2.0 and its archival content. For more background, see this post.
By David Aronchick
The true lean business demands nothing less than absolute agility. Anything less and you’ll be doomed for failure.
Agility, in the world of software, is the focus on iterating quickly on customer demands, releasing regularly, and getting accurate customer feedback to start the process all over again. Instead of spending countless weeks and months gathering requirements and then putting together the absolute perfect product, the idea is that you release a MVP— minimumviable product —as soon as humanly possible and then change it based on what you see people doing with a live version of your product. There’s a lot to love about this philosophy: it’s creative, it’s easy to test, and, best of all, it throws all the old ways of doing things under the bus!
However, there’s an enormous hidden danger here. When you take on a project like this, all you get, every time you release, is exactly what you need to release and nothing more. One might argue that this is the case with all engineering projects. It’s possible there hasn’t been a project that has launched on time to spec since the first caveman carved a rock into a wheel! Even then they probably slipped up on some feature somewhere, and it’s my guess they probably intended to have a cup holder of some kind—it’s tough chasing a mastodon and holding your mug of beer (mug of beer) at the same time. But it’s more complicated than that, because so often not even you really realize what you need to release.
When we sit down to think about the market we’re trying to address, the best thing to do is to achieve total clarity in the vision. An ultra simple blogging platform. A social network to connect with the friends you already have. A mobile phone that actually works. But, in the strive for perfection, you leave off the little things that makes these products actually work. Posterous is awesome because it’s not just ultra simple, it’s because you can submit ANYTHING from ANYWHERE. Facebook wins because it isn’t just leveraging existing networks, it’s also because they have a look that is clean and simple, opposed to the endless choices (and, sadly, hideous design) of MySpace. iPhones are works of art, being just a great phone (networks not withstanding) would not have been the same kind of game changer.
The problem is that when we’re agile, these little bits of your product that change your product from good to great often fall between the cracks. There’s a real tendency to look at something and say you’re done and move on. Okay, search is all set, on to Twitter integration! But then you’re left with a search product that may be 100% functional, but only appear 75% of the way there, and, as Joel Spolsky likes to say, “People Who Aren’t Programmers Do Not Understand That Things That Appear 5% Done But Are 95% Done Think They Are 5% Done.” (TheIceberg Principle Revealed) So, while you certainly have something awesome, and mostly done, the average person out there will not realize that it’s 95% done and feel a sort of cognitive friction which may be JUST enough to put them over the edge and never come back. And, unfortunately, this is also the kind of thing that never shows up in the customer feedback.
The fact is the little things do matter. Any time people see little bits and pieces that don’t fit together just right, they get the wrong idea(tm). But releasing SOMETHING because you were able to get it out quickly is better than releasing NOTHING because you’re getting to 100% perfection, right? How do you remain agile without blocking on being pixel perfect?
The answer must be you. You have to keep focused, that release after release, you have the commitment to quality to carry your niggling little bugs through to the next release. You have to have the center of gravity to realize what your company represents and, while you want to focus on the big stuff first, never be satisfied with a product that does not meet the goal of what you want your company to be. When corners fit together, when the search boxes respond to tabs/enter/return on all platforms, when you can accept every file format from .wmv to .ogg— that’s who you are. Getting there isn’t easy, and you’ll be making trade-offs between big features and fit-and-finish forever. In the end, as you build a product that actually goes from perceived 5% done to perceived 95% done, the response you get from your customers will be worth it.
