What does the thirtieth study tell you that the first does not?

Nobody watches one sighted participant miss a button and writes "sighted users miss buttons" in the summary. Yet everybody watches one blind participant struggle with a table and writes "blind users need."

One person becomes a population as soon as the population is small and unfamiliar to us, and that should embarrass us. I wouldn't call the result a finding. It's a stereotype, even with a recording attached.

The opposite failure is harder to notice and looks more rigorous. A real finding gets set aside because it came from one person, in a study of one, with nothing to weigh it against.

It's a bit like directions. Ask one person and you get a route. Ask thirty and you find out the town has two stations, and that most people assumed you meant the other one.

For a long stretch of my career, accessibility research was the first kind of asking.

I wouldn't talk anyone out of that first session, though. Something real happens the first time a team includes a disabled participant, because the room stops arguing from imagination. For most teams it's the single biggest change available.

It buys less than we pretend, is all. One session can't tell us whether we watched a pattern or a person, so it gives us a feeling when we needed a direction.

Thirty sessions showed that participants disagree

At the Rhonda Weiss Center, under a multiyear Department of Education grant, I led design for a system that publishes IDEA data. That data is what states report about how children with disabilities are served.

We ran more than thirty usability studies with disabled participants, including blind and neurodivergent users, around a stakeholder group of about eighty people.

I didn't expect them to disagree with each other, and I now think that disagreement is the entire point. They didn't disagree about whether the thing worked. They disagreed about what a good answer looked like.

Participants using screen readers didn't all want the same treatment of the same table. People navigating by keyboard didn't want what people using magnification wanted.

Participants who process dense text quickly asked for the opposite defaults from participants who don't, and each of them was right about themselves.

So past a certain number of sessions, what do blind users need stops being a question with an answer. What we get back is a range, and that range is the finding.

A range is harder to hear than a defect list, because we can finish a defect list. If thirty people all hit a control with no name, we name it and the work is done.

If thirty people want three different reading orders, there's nothing to fix.

What we have is a decision about which default ships, what becomes adjustable and what gets offered as a second route to the same meaning.

That costs more, but it's ordinary design work and we already know how to do it everywhere else. Nobody ships one font size and considers the argument settled.

Only in accessibility have we been permitted to treat a category of people as having one set of needs, mostly because we had met one of them.

Plurality is the word I'd put on it. We include disabled people in research to stop designing for a composite who doesn't exist. We aren't there to acquire a spokesperson.

Posts like this usually skip the cost, so here it is. Thirty sessions is mostly logistics. Recruitment has to reach past the same few people who always say yes.

Scheduling has to fit the assistive technology the participant actually uses, which may not be the one we happen to own. People have to be paid, because unpaid consultation selects for those with spare time. And sessions overrun because the setup did.

A multiyear federal grant paid for all that. A product team on a quarterly cycle has a harder case to make, and I'm not going to pretend otherwise.

The sessions themselves were almost never the expensive part, though. The expense was in everything arranged around them, which is mostly a budgeting and scheduling problem.

Teams rarely stop at one study because they learned enough. They stop because nobody owned the logistics of the second one.

Three sessions are enough to find a disagreement

• Run three sessions, not one. Three isn't statistically anything, but it's enough to surface one disagreement, and one disagreement changes how a team talks for the rest of the project.
• Recruit for difference inside the group. Two screen reader users with different tools and different tenure will teach you more than two who match.
• Pay everyone, at the rate you pay any other participant. This is the line most likely to be cut, and it decides who can afford to show up.
• Stop writing findings as "blind users." Write "two of the three participants using screen readers." It's longer and it's true.
• Book the second study before the first one runs. In January it costs nothing.
• Keep the disagreements in the report. A summary that resolves every conflict into a single recommendation has thrown away the most expensive thing you bought.

Putting one disabled person in the room, once, is better than nothing. But it's a version of the mistake I started with, because one voice still stands in for everybody, and we're the ones who selected it.

So look at the last research plan you wrote and count the sessions that included disabled participants.

If the number is one, I wouldn't start by asking what you learned from it. I'd ask what the second session would have told you.