Trust your instincts or rely on procedure – when to design with intuition
Main illustration: Amber Vittoria
A few years ago, we had a seemingly simple problem to solve: our customers needed an easy way to set realistic expectations around their response times to conversations in our Messenger.
Did this prove that we should have trusted our intuition all along?
When we first considered the problem that businesses were facing, we instantly thought of implementing something like an “office hours” setting, so businesses could easily alert people to when they were likely to respond.
However, we convinced ourselves that we’d better tackle the problem properly when we had time to fully explore the solution, and so we took our time defining the problem, doing extensive research and exploring lots of possible solutions. The whole process took a while.
In the end, we came up with something very close to our original idea of “office hours.” That tiny, seemingly obvious improvement turned out to be the most upvoted update we shipped in 2017.
Did this prove that we should have trusted our intuition all along? Did we waste our time with those rigorous explorations? Did our customers lose out on a useful feature for all that time?
Well, it might seem like that’s the case, but the balance between trusting your intuition and engaging in a more rigorous, exhaustive design process is a lot more complex than this anecdote suggests.
Why we should trust our intuition
It can be somewhat embarrassing to mention that you made a design decision solely based on intuition.
In order to solve a problem, a designer is usually expected to explore multiple options, diverge and converge on different ideas, evaluate the pros and cons of each and eventually land on a final solution.
Compared to the rigorous process outlined above, relying solely on intuition to make decisions seems rudimentary, irrational and not an approach that will lead to good design.
Our brain learned to optimize its resources by doing a quick synthesis of past experiences
However, in some scenarios, using your intuition can be a valid way to solve design problems. Over the course of evolution, our brain learned to optimize its resources by doing a quick synthesis of past experiences and knowledge in response to your current challenge. Intuition basically helps you make the best decision based on associations, memories, pattern-matching and assumptions, saving your brain unnecessary effort.
Which means that intuition can be developed and improved by feeding more experiences and knowledge into it. We are constantly refining our ability to make snap decisions thanks to something called the “adaptive unconscious,” which Malcolm Gladwell in Blink describes as “a kind of giant computer that quickly and quietly processes a lot of the data we need in order to keep functioning as human beings.”
The theory is backed up by a growing body of academic research that shows that the accuracy of your intuitive decision making grows with the level of your expertise.
So the more things you design, the more research you conduct, the more products you actively “read” as a user, the more life experiences you live through, the better your design intuition becomes. All of which enables you to skip some of those explorations and jump straight from problem to a solution.
If intuition, then, is the result of polishing your skill through practice, allowing your mind to use it as a shortcut to a good decision, why not just apply it to every project?
Why we shouldn’t trust our intuition
Several years ago, when we first started experimenting with a proto-version of our Operator bot, I was instinctively confident it needed to have a personality to connect with people and successfully do its job. Yet as we started researching and testing the prototype with real people, it turned out they intensely disliked it. It was not just the personality we picked for Operator that they disliked, but the whole fact it had one at all. Only when we stripped the bot of any personality did it start doing its job well.
Very often our instincts are flawed, since our brains are full of biases and fallacies
The lesson for me was clear: intuition can’t always be right. In fact, very often our instincts are flawed, since our brains are full of biases and fallacies.
First, humans are extremely overconfident. In his book The Social Animal David Brooks talks about a study that shows that we are wrong much more often than we’d like (up to 99% of cases). And when given more time to reflect on that instinctive position, people usually become even more convinced they are right.
Second, our subconsciousness always searches for patterns and creates coherent stories full of wrong explanations for past events. For example, our memory usually reconstructs past events with an intent to make ourselves look better.
Third, despite however much practice, there are things our brains are still bad at. Daniel Kahneman, in his book Thinking Fast and Slow mentioned one of his studies suggesting that professional statisticians, despite years of working with data, were still unable to develop a good intuition for understanding the correctness of certain stats.
So, regardless of years of experience and polishing your instincts, you can still be wrong. Problems vary greatly in terms of their complexity and nuance – what could be an obvious and industry standard solution to a common problem could turn out to be wrong when applied to your scenario, your product system or your customers.
If there is a risk an instinctive call will turn out wrong, why listen to your intuition at all?
How to decide on when to use your intuition
It’s easy to over-rely on either end of the spectrum, either constantly using your intuition as a shortcut or unnecessarily dismissing your intuition in favor of a one-size-fits-all design process.
Instead, it’s much better to rely on your intuition in some circumstances and be skeptical of it in others.
First, think about the type of project you are dealing with? What’s the potential damage caused by getting it wrong? Is it a tiny Github issue or a cross-functional multi-team project? How many customers will it impact? Will it involve a couple of hours of work for you and a single engineer, or months of labor for a whole team? Are there potential downstream implications with it?
Second, how confident do you feel about your instinctive solution? Basically, if you visualize it to be a part of your product, does it feel like the right approach?
Combined on a single graph, here’s how it would look:
It thus comes down to the cost of being wrong. If you feel confident you are dealing with a low impact project, you should be comfortable making a call based solely on intuition. Yet when faced with a high impact project, your instinctive feeling (even if it’ll turn out right) should not be fully relied on.
Training your intuition
Thinking about that “office hours” solution now, it’s clear we certainly made some mistakes. We might have overthought it and it might not have been the right project to explore so deeply. Yet it took a month of work for several engineers to build and it was experienced by tens of thousands of businesses and millions of their customers. The cost of getting it wrong by solely relying on intuition was way too high.
Your design instincts, formed in the years of practicing your craft, should not be easily dismissed. Your intuition and instincts are not something to be afraid of or embarrassed by. But they are not something to blindly follow either. Only when you are fully aware of trade-offs will you be able to design faster by relying on your intuition and design better by properly exploring and iterating.
