Deferring Decisions
I’ve spent the last few days working on a presentation I’ll be giving at a conference in a few days. I’ll be speaking about “Resonance,” a design process we use for the development of everything from blenders to businesses. I’m glad to be able to present it, in part because it’s required me to boil the process down to its essence and make a real story out of it.
The gist is this: wait.
We have many, many clients who approach product design this way: here are some brand attributes; design us a few different products that solve problem X; we’ll test them with consumers to see which one communicates the brand attributes best and go with that one. And the test is often a straight preference: here are two pictures of products – which would you buy?
For a long time, we struggled with why this sort of testing bugged us so much. Here’s my answer: it’s because designers hate to be wrong. Hate it. If you’ve asked me to design something that looks, say, “rugged,” and I did, and consumers show preference for another that I don’t think looks rugged, I hate that. Designers are especially vulnerable because of the subjective nature of the work. A team of designers might agree that a particular form communicates a particular attribute, but that doesn’t mean a consumer will agree.
The secret, then, is not to design a product until after we know that the consumer will agree with us. We do that by focusing on the communication of the message, not the design of actual features. We test dummy products with consumers–designs we know we will throw away–because we want to know what they say and how they say it. Once we know, we can design the product (or the service, or the business–we can use similar techniques throughout) for real, using the tools that have been shown to be effective.
The process can feel weird. We put a lot of time into the “fake” designs, and such a process is very new to our clients. But in the end it gives everyone much more confidence in the designs, and consumers agree with their wallets.
Question for you: how much of your design process is guesswork?
Follow Me
I’m intrigued- can you talk more about this process when you get a chance? Perhaps give an example of a “fake” design?
Sure. By “fake” I mean: looks like a convincing product, but with no reality–no engineering behind it, might not be actually manufacturable, looks like it’s made out of materials that we don’t know will work. But it’s important that it look “real” and as integrated as possible; showing material samples separately from unpainted foam models is not as effective as painting the foam model with the right colors. The goal is to get the participants to talk about all of the attributes especially as they relate to each other. Then, very important: throw the examples out. They are for testing communication, not models for the final product. It’s easy to fall into the trap of “well, this is close, so let’s just use it” but better to start over with the new knowledge from the testing, to avoid the “Mr. Potatohead” approach: glomming a bunch of stuff that worked well in testing together. Design from scratch, but be informed.