We tried that—it didn't work.

The Improvement Kata

"We tried that, it didn't work" is a common response when I suggest improvement. I approach that in a couple of ways.

First, there's the reality check. I ask, "Could you describe exactly what you did?" This is just information gathering, so I just listen. No suggestions. Restrain yourself! 😄 More often than not, they were not actually doing the thing that "didn't work" at all, but I need to verify that. Never say "you didn't do it right." This is your starting state.

Then I ask, "And what problems were you trying to solve with that?" The only questions I ask here are clarifying questions. Step one of any improvement is: Identify the pain. You don't know what it actually is if you don't ask.

Now I'm in a position to actually suggest something. I'd add that I *never* use the names of the "failed" whatever when I make those suggestions. "Well, you have to do it right,” is not helpful feedback. Instead, I use a process:

(1) Suggest one or two small experiments—and it's essential to frame these as experiments—that will move them in the direction of alleviating the pain. Make it clear that, if it doesn't work, you'll try something else. I'll say "In the past I've had good luck solving <some tiny aspect of the larger problem> by <doing this>." It's best if you can cite a situation where what you're suggesting actually did work. They've already been burned by theory.

It can't hurt to put both the problem you're solving, the ultimate goal, and the experiments on a virtual or physical board so that you can keep track. Put the experiments on a sticky so you can move them to "worked" or "didn't work" sections of the board. It's good to annotate the sticky with a "why" in both cases.

(2) Repeat, but with a new experiment that moves you even closer to the ultimate goal. Adjust the board to show your progress. We’d like that “it works” section to grow, but it’s useful learning if it doesn’t.

What I just described is Mike Rother's "Improvement Kata" from his book "Toyota Kata." He's a manufacturing, not a software, guy, but the book is a great read. I've successfully used the technique many times.

Incremental improvement focused on solving specific problems is always better than a big all-at-once "transformation," in my experience.