Written by Greg Schneider, Contributor, 2026
Picture two people. The first is standing in line on release day. He’s wearing smart glasses that read his notifications aloud, a smartwatch tracking his sleep, and earbuds that translate conversations in real time. Half the features don’t quite work yet, the battery gives out by dinner, and he does not care one bit. He got to touch the future first, and for him, that’s the whole point.
The second is a line worker I’ve known for years. He does his taxes by hand, out of a paper bank book he’s kept updated since before some of his coworkers were born. No app. No password. No subscription. When people tease him about it, he shrugs: it works.
Neither of these people is confused. Neither is broken. They’re standing at opposite ends of a curve that was drawn up more than sixty years ago — and once you see it, you start seeing it everywhere.
The model nobody has managed to improve on
In 1962, a sociologist named Everett Rogers published Diffusion of Innovations, and in it he described how any new technology moves through a population. Not all at once. In stages, in a pattern so reliable it has survived every technology wave since — radio, television, the personal computer, the smartphone, and now AI tools. Rogers divided people into five groups:
Innovators — 2.5%. The release-day line. They adopt because it’s new, full stop. They’re willing to pay early-adopter prices, tolerate bugs, and figure things out from a half-written manual.
Early Adopters — 13.5%. They wait just long enough to make sure the thing isn’t a mirage, then move fast. These are the people others watch: the supervisor whose team quietly copies whatever tools she picks up, the neighbor whose opinion settles arguments.
Early Majority — 34%. Prudent, not slow. They adopt when the price has dropped, the reviews are in, and somebody they trust is already using it. This is where a technology stops being a gadget and starts being normal.
Late Majority — 34%. Skeptics, often for good reason. They adopt when nearly everyone else has, when the old way is getting hard to maintain, or when the new way is simply the only way left to get the job done.
Laggards — 16%. The last to move. They hold on to the old method the longest — because it works, because it costs nothing extra, or because they’ve watched enough “revolutionary” products get discontinued to know that today’s platform is tomorrow’s landfill.
Every seat on the curve is rational
Here’s the part that gets skipped when people retell this model: none of these positions is a character flaw, and none of them is free.
The enthusiast pays for his front-row seat — in money, in bugs, in the smartwatch he’ll replace in fourteen months. He’s effectively paying full price to beta-test, and some of what he buys will be abandoned by its maker within two years. The paper bank book, meanwhile, cannot be hacked, cannot be locked behind a forgotten password, cannot be sunset by a company that lost interest, and will never once ask its owner to accept new terms of service. It also never crashes during tax season. Which of those two sounds irrational depends entirely on what you value: novelty, or certainty.
Most stage behavior is just a different, sensible answer to the same three questions: What does it cost me if this fails? What does it cost me to learn it? And what do I actually gain on an ordinary Tuesday? Change the answers and you change the person’s stage — which leads to the most useful thing about the whole model.
Nobody lives at one point on the curve
The curve describes reactions to a specific technology, not personality types. The line worker with the paper bank book was, I later learned, the first man on his crew to master the plant’s new diagnostic tablet — because it made his actual job easier the same week he got it. The release-day enthusiast, meanwhile, still pays his bills with paper checks, because his bank’s app annoyed him once in 2019 and he never went back.
The same household contains the whole curve. One spouse orders a robot vacuum the week it ships; the other still writes grocery lists on paper and will until the pen runs out. If you want to predict how someone will react to the next new tool, forget their age, their job title, and whether they call themselves “tech-savvy.” Ask instead what the last tool cost them, in money and in hassle, and what this one is really offering them.
Using the curve instead of just admiring it
If you ever have to roll out a new tool — at work, in a community group, in your own family — the curve is a planning document. Each stage needs a different thing from you, and giving the wrong thing to the wrong stage is how rollouts die.
Innovators: give them early access and a feedback channel. They’ll find the bugs before your customers do, and they’ll forgive almost anything except being ignored.
Early adopters: give them evidence and respect. Real numbers, honest limitations, no hype. These are your translators — if they vouch for the tool, the majority follows; if they feel sold to, they’ll say that too.
Early majority: give them proof from people like them. Not the vendor’s demo — a coworker on a similar team, a neighbor with a similar house. One peer success story outweighs a hundred feature lists.
Late majority: give them guarantees and a fallback. Training, a help line that answers, and a clear answer to “what happens if this doesn’t work?” Remove the risk and most of the resistance goes with it.
Laggards: give them time and dignity. Don’t mock the paper bank book — it’s load-bearing. Keep the old way working as long as you honestly can, migrate their information for them rather than handing them a form, and when the change finally becomes unavoidable, say so plainly instead of dressing it up as an exciting opportunity. People who feel respected through a forced change remember it; people who feel ridiculed remember that too.
The curve hasn’t changed since 1962 because people haven’t. Every new technology — AI assistants very much included — will walk the same five stages, in the same order, at the speed trust allows. You can’t skip a stage, and you don’t need to. You just need to know which stage you’re talking to, and give them what that stage actually needs.
Sources: Everett Rogers, Diffusion of Innovations (1962); “The 5 Stages of the Technology Adoption Curve,” Omniplex Learning, February 28, 2023.