Learnability

Products were judged by feature count for a long time. More things you could do meant a better product, and anything hard to use was something the user would learn. Inside a company that premise partly holds, because training time gets scheduled.

For a product people sign up for themselves it collapses. Five hundred people a month signed up for Pagedeck and fewer than half created a single document in their first week. Features were not the problem. The screen right after signup offered twelve things to do, and a first-time visitor spent their time deciding which to press and then closed the tab.

A product you can use and a product you can pick up quickly are different measurements. This axis exists to measure the second one.

Only what happens within about the first thirty minutes counts. Speed once someone is fluent belongs to efficiency, so it is not counted here.

Measuring it is simple: give five first-time users one task with no guidance and watch. At Pagedeck the task is creating a document and sharing it with a colleague. Three things get recorded.

1. How many finished unaided. Two out of five is a low score.
2. How long those who finished took. A task a fluent user does in 40 seconds taking a newcomer six minutes means they got lost somewhere.
3. Where they stalled. Note the stretches of scrolling up and down and the buttons pressed by mistake.

Raising the score is not about cutting features. It is about cutting what is visible first.

1. Narrow the first screen to one thing to do. The other eleven stay in the product and appear from the second screen onward.
2. Name things in the words users use. A term that only circulates inside the team means a newcomer has to learn that word before anything else.
3. Use the empty screen as guidance. An empty list is the first screen a new user meets, so one line there about the next thing to do replaces a separate tour.

"Learnability is the problem. Signups are up but under half create a first document."

People who arrive are getting stuck on the first task.

"Let's give five first-timers the task with no guidance and decide from that."

This axis gets measured by observation, not by guessing.

"Twelve buttons on the first screen is too many. Let's keep one."

Reducing the first set of choices to raise this axis.

At Pagedeck, Noa runs this check once a quarter. When five genuine first-timers are hard to find, she mixes in colleagues who don't use the product. Mia watches it from the numbers side, since the share of signups that create a document in their first week works as a proxy for this axis.