by Ryan Walker •

Sometimes you have to just know when to pull the plug

367 days ago I started a project, GapAttack, based on an idea that I couldn’t get out of my head.

“What should we do next with the business? Let’s collect opinion from staff on what they think the business should focus on. Then we should focus effort on completing work that will benefit the organization most.”

Gaps = a mismatch between current state and desired state (it could be a problem or an opportunity).

Inpsirations for the project

  • “Limiting work in process” which I learned about studying lean software development theory, which of course is derived from lean manufacturing.
  • Crowdsourcing theory, that the employees in a company can help provide important data about the gaps.
  • “Eat that Frog” by Brian Tracey which says to take on hard things first, because they have the potential to deliver the most value and differentiation.

So the system collects ratings on urgency, importance, size, and challenge for each gap, and then computes a prioritized list based on an algorith.

Why am I abandoning it?

  • I don’t even use it myself - without a crowd, it doesn’t work.
  • No one else is using it. When I shared it with some people, I got a lackluster response.
  • So far I haven’t made any money from it.
  • I can free my energy some other more fruitful work.

Mistakes I made

  • Tried to take shortcut - I bought an admin CSS template and the SAAS Railskit from Ben Curtis, and started the project trying to reskin the the default CSS. Big waste of time, because I ended up throwing all that CSS away at some point (it was too loud.)
  • In the beginning, I thought “won’t it be a fun challenge to build the whole thing by myself” - programming, design, site operations, marketing, etc. I wanted to prove to myself that I could make a great product solo. And I probably made a lot of mistakes that a partner would have caught.
  • Didn’t follow BDD - I didn’t really read the specs that came along with SAAS RailsKit, because I really wanted to focus on the core idea to make sure it was marketable. But the system had a lot of well-written specs, and they weren’t “mine” so I started doing a lot of the work without tests. Eventually I tried to run the specs and there were a ton of failures (even though the app worked the way I wanted it to) so I had a big Rspec mess. Never again will I build a project without driving the features through test-first methodology.

Today I opened up the code, and was so disgusted by it all, that I decided to toss it.

I considered starting over again as a way to learn Rails 3 better, but I’ve sort of lost passion for the idea. I expect to come back to it some day, because I know the idea can work. But for now, I’m shutting the project down indefinitely.

And I feel great about it.

Filed under Ruby on Rails Work Management