3 min read · Starter draft
Designing for Someone Who Doesn't Have Time to Learn Your Interface.
Nobody arrives wanting to use your product. They arrive wanting something done.
Designers tend to imagine a user who is curious about the product. Someone who explores, reads the labels, and builds a mental model over a few sessions.
That person exists, and they are rare. The usual visitor is doing something else. They are between two meetings, or holding a child, or trying to finish the task that your product is one step of. They have given you about as much attention as a road sign gets.
Design for the glance
A glance can take in one thing. So the question for every screen is what that one thing is, and whether the layout agrees.
If a screen has three things of equal visual weight, I have not finished deciding. Hierarchy is the designer doing the prioritising so the reader does not have to.
Familiar is a feature
Novel interaction patterns cost the user something every time. Sometimes the cost is worth it. Usually a standard control, placed where people expect it, is the more respectful choice.
I keep a simple rule: be unusual in one place per product, at most, and make sure that place is where the product's actual value is. Everywhere else, be boring on purpose.
Write like the person is in a hurry
They are. A few things that follow from that:
- Buttons say what will happen, in a verb
- Errors say what to do next, before they say what went wrong
- Nothing important lives only in a tooltip
Then watch someone use it
The only reliable test is five minutes of watching a person who has never seen the screen. Wherever they pause, the interface asked them a question it should have answered itself.
I have yet to run one of these sessions without being surprised, which tells me how well I can predict it from my desk.