Designers, here’s why pixel perfect design is bad

Hold on, don’t take my head off just yet, hear me out. I love a beautiful interface as much as the next person. It’s the approach that gets me. The notion that you can and should aspire to create the perfect design.

I see digital products as living things. They are like us evolve, learn from our mistakes, change our mind, explore the world, grow up, grow old. The same way no person is perfect, neither are our products, created by us.
Let’s say a feature or a prodct seems amazing right now, your best work so far. In a year from now it can get lost in a sea of similar products, redundant by the company’s pivot or irrelevant thanks to a disruptive technology of a competitor that led to a completely new way of doing business. “The only thing that is constant is change”. Heraclitus said that 2500 years ago, and it’s still true.

Perfection is a concept of the eternity, while reality is always changing.

Usability testing session for an IoT product
Usability testing session for an IoT product

Now let’s talk methodologies. How do you know it’s perfect?
Have you tested it with customers of the target audience? Did you prepare a thorough list of assumptions you wish to validate? Did you make sure you eliminate bias? Did you use both qualitative and quantitive methods to test your design?
If you answered yes to all of the questions above, nice work! But even so, I am willing to bet your design is not “perfect”. There are always unexpected behaviour patterns and unkowns. If you assume you missed things, you will be more inclined to finding those things and improve. This is what Build-Measure-Learn is all about. Do small and fast MVP experiments, analyse and return to the sketching board for another round. On the other hand, when something is perfect, there’s no point in questioning it, right?

Each company works differently, and maybe user testing or tracking performance analytics isn’t your responsibility. But you can still get involved, ask those questions, make a habit of being informed. Your superiors are more likely to see it in a positive light when you take initiative.

Dan Nessler’s revamped Double Diamond overlaid with a “build-measure-learn” loop
Dan Nessler’s revamped Double Diamond overlaid with a “build-measure-learn” loop

Another problem with perfection, it takes time. “We have all the time in the world and no backlog to worry about” said no company ever. Agile took over as best practice for development, meaning big features will be cut to stand-alone bite-size features, usually exposed to customers one at a time. You don’t have time to go into a major project and reach perfection, since the development team is waiting for something ready next week. Maybe only part of the project will be developed, moving on to the next project for now. Don’t fight developers and product managers, join them. Change your mindset and learn to design agile yourself. Done is better than perfect.

One last point if you’re not convinced yet. Design is creation, you put your heart and soul into it. You might identify with it, feeling it’s part of you. When a feature fails you might feel you failed. When your design is “perfect”, there’s a higher chance you will feel you’re being criticised when your design is being criticised.
Let me suggest a solution — instead of one perfect design, do 3 really good ones that you’re happy with. Now put them to the test, compare them and get feedback. It will open a healthy discussion. “I really like what you did here in option 1, but this thing in option 2 works better.” Force yourself to come up with actual different solutions. It will improve your design and bring new ideas that you wouldn’t have thought of otherwise.

Product designer @Augury |