A user once told me they loved how configurable my product was. I checked the analytics. They had never opened the settings page.
What they actually meant
They didn't love that it was configurable. They loved that the defaults were right. Configurability was the safety net that let them trust the defaults — knowing they could change something made not-changing-it feel like a choice instead of a constraint.
Good defaults are invisible. Bad defaults force the user to discover, evaluate, and override. Every option you ask the user to make a decision about is a tax on their attention.
The defaults checklist
When I'm shipping a new feature, I run through this:
- What's the most common case?
- What's the least dangerous case?
- What's the case that lets a new user succeed in 30 seconds?
Then I make those the same thing.
Where this fails
Sometimes there is no common case. A reading app for kids and adults can't have one default font size. A scheduling tool used in twelve time zones can't have one default time zone.
When that happens, I ask once, with strong defaults, and I never ask again. Onboarding is where you earn the right to never bother the user with these questions again.
A great product asks one question. A good product asks five. A bad product asks twenty-five. A truly bad one starts asking again on the second use.
The boring secret of every product I've loved is that someone, somewhere, sat with a furrowed brow and decided — for me, on my behalf, before I knew I had a preference — what the right answer was. And they were right.