Imagine receiving a package while going into your home and along the way you pass a recycle bin. You remember ordering the item and know it isn’t fragile. Jamming a key into a gap between cardboard flaps and dragging it to tear through as much of the tape as possible. You then Hulk-Rip™ the rest of the flaps open.
After dropping the remnants of what was previously a box into the recycle bin, you walk into your kitchen to put the item down on your counter. There’s a junk drawer immediately below where you put the item. One of the things in that junk drawer is a razor knife.
Why didn’t you use the razor knife?
I’m having fun with an amusing story that may or may not happen to me all the time, but am serious when asking that question. Have you been in a similar situation? Why did you use the inelegant thing? The blunt thing? The kludge?
I’d argue the situation drove nearly all the behavior in that story. It wasn’t the perfect tool — a razor knife that slides through cardboard like a hot knife through butter. It wasn’t the perfect process — there likely wasn’t one and the desired outcome was happily achieved.
It was a combination of available tools and improvisational processes that made sense in that moment, under those conditions.
Research Trouble
A majority of my career I’ve gotten along wonderfully with researchers who dig deeply into the situation. Who work to granularly understand why others do what they do — discover the influences behind their existing behavior. Being a designer on team with people doing this work is incredible!
On the flip side, I argue with the researchers who skip or treat situational awareness as merely gathering constraints. Who think of the situation as boundaries for their team’s ideal feature and process ideas to fit within. Being a designer on a team doing this work is draining…
The arguments routinely come from seeing a situation as unique to an individual. Therefore, driving researchers to interview prohibitively expensive quantities of people to determine broad themes. Or, over-designing to a subset of users at the expense of the majority audience.
Hogwash.
Notice the centering of general themes and wide audiences. A broad thematic generalization can hold an ideal feature. A majority of people can hold an ideal process.
It’s all chasing ideal features and processes. And it’s all a waste of time.
Research Fun
In that silly story, I’d argue the person-specific situation can be generalized in useful ways other than features or processes. For example, walking by a recycle bin is major behavioral driver for that person. Can we, instead, ask how many other people have recycle bins? Can we theorize moving the recycle bin closer to the razor knife will shift behavior from Hulk-Ripping to razor knife cutting for that audience?
Darn tootin’ we can! And we should.
Did we make the razor knife better? Did we propose an alternative process?
Nope. And. Nope.
We can pull a huge amount of useful guidance out of a few conversations. Not ideal features and processes, but that’s okay. Let’s be real, people don’t want new feature and processes. They want progress. They want to be able to look at today, compare it to yesterday, and see that it’s noticeably better.
Feature Factories
None of the arguments I’ve had with researchers are the fault of the researcher. It’s always the larger operating environment. Unfortunately, symbiotic relationships can form with budding feature factories and certain kinds of research processes.
Does someone have a feature idea? Go ask a lot of people if they want it. If the feature being proposed is in any way hypothetically positive, many people will say yes.
Congrats! That person’s idea is now research validated.
If the idea came from the researcher’s boss, a peer of their boss, or — if they’re lucky — their boss’ boss, the researcher will likely be brought closer to a “seat at the table” where future ideas are considered. It’s a tough incentive to ignore. In many cases, it’s less an incentive and more a demand.
Once this relationship is settled into a general product process it’s nearly impossible to unwind.
Enter “AI”
I’ve kicked the tires on as many tools as I can find that claim to help designers and researchers better collaborate — AI or otherwise. To help scale research. To scale design. Together.
I keep bumping into the same bright, flashing, warning signals.
All these “AI” tools are paving roads leading straight to feature factories. And they’re doing it through this perverse research incentive. Finding a small path of least resistance, breaking it wide open, unleashing a feature factory, and claiming the efficiency gains of the newly effective feature factory.
I’m not convinced the “AI” is doing much of the work. The permission to go full-feature-factory is where the gains are coming from. When the tools enable “I have a cool idea” to become “my idea is validated by research” without understanding the situational context of the user, things will dramatically speed up. The team will appear more productive.
But what is getting better? For whom?
I don’t have an answer right now. But I’m pretty sure it’s not the users of our tools and services.
Ho-hum…
And right when I thought there was a glimpse of a chance of smart companies maturing past feature factories. They get lured right back in with the opportunity of automating and accelerating all the things we shouldn’t be doing in the first place.
≈