Early in my career I worked at a SaaS startup building a tool for sales reps. Early on we had a simple insight: the five minutes right after a meeting with a prospect are golden. Memory is fresh, details are vivid, and most reps do nothing with that window.

So we built a conversational app that caught them right then, in their "salesflow." A smart alert, a few quick questions, and your Salesforce record was updated before you got back to your car. It worked beautifully.

Then we pivoted into pharmaceutical sales.

The same conversational interface. The same logic. Except now a multiple choice question could surface dozens of medically specific options, each one regulated, each one carrying compliance weight. We didn't have six choices. We had a hundred.

Nobody needed a UX researcher to explain what happened next. It was obviously, viscerally broken. Tech saw it. Product saw it. Sales, customer success, executives. Everyone in the room looked at 100 options stacked inside a chat bubble and understood immediately that we had a problem.

The solution wasn't to abandon the conversational approach. And it wasn't to just build a traditional form. We blended them. Conversation where conversation worked. Structured UI where structure was required. The result was better than either approach alone, and it became one of our strongest verticals.

What broke us open wasn't a failure of execution. It was an encounter with human diversity we hadn't designed for. Pharmaceutical sales reps live in a different world than typical sales reps. Different vocabulary, different compliance constraints, different mental models for what "capturing a meeting" even means. Our original product assumed a user. The pivot taught us there's no such thing.

I've been thinking about that lesson ever since. It took me a while to find the right language for it.

Turns out, Gene Roddenberry got there first. In 1968.

IDIC isn't a utopian idea. It's a description of your user base

Star Trek introduced a concept called IDIC. Infinite Diversity in Infinite Combinations. Roddenberry put it at the center of the show because he believed diversity wasn't a challenge to be managed. It was the source of strength. The Federation worked because it brought different species, different perspectives, and different ways of thinking into contact with each other and found that the combination produced something none of them could have built alone.

Many product teams treat diversity as an edge case problem. A localization checkbox. An accessibility audit. A persona for the "non-technical user."

That's exactly backwards.

Your users are not a modal person with minor variations. They are genuinely, substantively different from each other in ways that matter to how your product works. Different industries. Different levels of technical fluency. Different physical and cognitive needs. Different cultural contexts that shape what feels intuitive and what feels foreign. Different amounts of time. Different stakes.

When you design for the distribution instead of the assumed center, you don't water down your product. You make it honest. You stop pretending your assumptions about "the user" are universal and start building something that actually works for people as they are.

The pharma pivot and UX didn't fail because the product was bad. It failed because it was optimized for one kind of person. The fix wasn't a compromise. It was a more honest product. One that admitted: people are different, contexts are different, and good design holds space for that.

The good of the many

Star Trek gave us another principle that's harder to practice than it sounds. The good of the many outweighs the good of the few (or the one).

In product terms: you are not designing for your loudest user, your most powerful stakeholder, or the person whose feedback you hear most often. You are designing for everyone who will live inside what you build.

This is easy to say and surprisingly hard to hold onto. Power users are vocal. Executives have opinions. Sales can tell you exactly what the enterprise customer wants. And all of that is real signal worth listening to. But none of it is the whole picture.

The good of the many is a discipline. It asks you to keep asking who else is in the room. Who hasn't complained yet because they've just adapted? Who doesn't have access to the feedback channel? Who is the product silently failing?

Most of the time, the answer is: more people than you think.

Why this matters right now

We're building products at a scale and speed that Roddenberry couldn't have imagined in the 60s. AI is being embedded into tools that millions of people use to do their jobs every day. The decisions product teams make about whose needs get centered will play out across enormous numbers of human lives.

That's not an argument for slowing down. It's an argument for taking the ethics seriously.

Star Trek wasn't popular for 60 years because it had good special effects. It was popular because it kept insisting that the future could be better than the present, and that getting there required treating every person as worth designing for.

IDIC. The good of the many.

These aren't soft ideas. They're load-bearing principles for anyone who builds things other people have to use.

Roddenberry had it right. We just stopped listening.