Solve any problem — Myddelton

The most useful thing I learned last month was “a technique to solve any problem”. Seriously.

It’s called the KJ Method and it came to me via Leisa Reichelt, but it dates to the late 1960s and a man named Jiro Kawakita. It’s also called an “affinity diagram”, but I used this term at work and was told off for using technical terminology when plain English would do. Fair enough.

The KJ method

I’ve done two workshops with Leisa recently, Hands-on UX and Strategic UX. Featuring strongly in both was a method that she convinced me could be used to solve (almost) any problem. The KJ Method.

Chances are you’ve already used the KJ Method or one of its many subtle variations. It’s a group activity that works like this:

Brainstorm lots of ideas for your problem (individually)
Sort these ideas into groups and label them (collectively)
Rank them in any way that makes sense
Make a decision based on what you’ve done
Clearly articulate what you’re going to do next.

If that doesn’t sound familiar, imagine writing ideas on post-its, grouping them, naming them and then prioritising. That’s the KJ Method as most of us know it.

Creating a business case

But does it really work for any problem? I’m ashamed to say that I tested it on a friend. When she had a work crisis. Yep, UX designers need empathy…

She runs a project and her budget is being cut by 50%, so when she had 24 hours to write a business case, I made her brainstorm ideas for cuts. She hated the ambiguity. “I don’t have time. Where is this going? Why are we doing it?”.

But as the walls filled up with post-its, we discussed her ideas and started grouping them. Relationships between ideas formed – if one thing is cut by 50%, you need less of this other thing. We talked about how individual cuts would affect her project, which helped to prioritise the cuts into 3 scenarios. This led to a final decision, and suddenly all that was left was to write up the chosen scenario into a business case.

She turned to me. “But how did you know to do that? How did you know that it would work?” The truth is that I had no idea it would work. But after being told twice in a week that this technique could solve any problem, I reached for it in a time of crisis. And it did work…

Simple and complex

Of course, this isn’t really going to solve any problem. But it works for more than you might expect. After all, I thought it was only for category-and-label issues, but I was wrong.

The power of the KJ Method is in its similarity to the concept of emergence, where “complex patterns arise from many relatively simple interactions”. In the work that we do you need simple interactions to make it easy for people to engage with problems. But you also need complex patterns to emerge, because most problems are anything but simple. The KJ Method combines these two needs into a single, repeatable system.

Next time you’re stuck, try it out.

I’d love to hear what you think. What other problems have you solved with this? Has it ever failed you? Let me know on @myddelton. And check out Jared Spool’s introduction to the KJ method for a fuller description.