Field test · Type. · in my own classroom

Type.

I built a typing course for my students. Then I watched 15 of them use it.

My roleTeacher, designer, and tester. I directed the build with Claude Code.
Testers15 of my adult ESL students, ages 22 to 65
Method8-task usability test, in person, on classroom desktops
WhenNovember 2026, five weeks into their first computer class
Why it exists

A scam ad popped up in the middle of class.

We were practicing on a free typing site when a scam ad showed up on a student's screen. Most of my students had learned to turn on a computer only a few weeks before, so they could not tell the ad apart from the lesson. That week I made Type.: no ads, no sign-up, and nothing to click that is not the lesson.

I did not run formal research first. I built it around what I saw in class every day, then tested it once my students had been using it. Everything it has came from that room: hands colored by finger so both hands get used, the next key lit up so students know where to press, and an accuracy score so they know when to redo a lesson or move on.

The test

8 tasks, 15 students, one stopwatch.

Each student sat at a classroom desktop and tried 8 tasks while I watched and timed them. I gave each task in plain words and did not show them where to click.

  1. Open lesson 6. Can they find a lesson on the screen?
  2. Open lesson 72. Can they scroll to one that is off screen?
  3. Find the lesson called "The Comma." Can they find a lesson by its name?
  4. Turn on key sounds. Does a one-click toggle make sense?
  5. Open lesson 48 and type two lines. Can they actually do a lesson?
  6. Turn off hand colors.
  7. Turn off hand hints.
  8. Turn hand hints back on.

I sorted students by the levels I already use in class: 9 upper beginners who work on their own, 5 middle beginners who are fast but still building confidence, and 1 lower beginner who needs extra help.

Results

91% of tasks done, most in under a minute and a half.

91%of tasks completed (109 of 120)
8 of 15students finished every task
1:17median time for all 8 tasks

Students out of 15 who completed each task.

Time tracked skill level closely. Upper beginners averaged 1 minute 9 seconds, middle beginners 1 minute 49 seconds, and my lower beginner took 2 minutes 48 seconds.

What I learned

Lesson numbers worked. Lesson names were hiding.

The toggles just worked.

All 15 students turned key sounds and hand hints on and off without help.

Scrolling was not really a problem.

5 students skipped the scrollbar, but 4 of them still reached lesson 72 by tapping the arrows or jumping from lesson to lesson. Only one student could not get there.

Lesson names were the real problem.

A third of the class could not find "The Comma." Lesson topics only live inside a dropdown, and the main screen shows numbers. Students who did find the dropdown told me seeing topics helps them pick which keys to practice.

The surprise: most students do not care about names.

They like climbing lesson by lesson because it shows steady progress. So the fix has to keep the ladder and add the names, not replace one with the other.

What students liked

In their words, more or less.

What I am changing

One list, with the number and the name together.

Now
123456→
Lesson topics ▾

Numbers across the top. Topics hidden in a dropdown.

Next
31 Question marks32 The Comma33 Quotation marks34 Numbers 1 to 5

One list down the side, in order, with every topic visible.

The order stays the same so students can still climb the ladder. Once the list is in, I will remove the row of numbered buttons at the top so the screen does not get cluttered.

Topic names in the sketch are placeholders, except "The Comma," which is lesson 32.

Honest notes

What this test cannot tell me.

Next round: add the lesson list, then retest with students from another class. The goal is simple: everyone finds "The Comma."

Back tomy toolsSee everything I built →