The risk with your best performer isn't the plateau. It's that you keep grading them like nothing changed.

You're sitting across from the person who has carried your ops team for six years. Her numbers are flat this year — not bad, just flat, the same rough shape they've had for three cycles now — and her pay has crept up every one of those years, because that's what happens to people who don't leave. You're doing the math the way you always do it: output against cost. The math doesn't look as good as it used to. Quietly, you're starting to wonder if she's coasting.

A study in the Journal of Management says you may be asking the wrong question. Michael D. Ulrich of Utah State University and Matthew L. Call of Texas A&M's Mays Business School set out to answer something narrower and more useful than "are stars worth it": what happens to the value a star employee creates as they age, once you separate what they produce themselves from what they cause other people to produce. They tested their predictions inside one financial services organization — the university's own coverage of the paper doesn't say how many employees were in that sample, and a single-company field study is exactly that: one company, not an industry-wide claim. Their finding held there. Whether it holds in a forty-person shop is something you'd want to watch for yourself, not assume.

What they found, in that one organization: star employees command pay that outpaces their individual output more, the longer they stay — a function of the leverage and visibility that builds around anyone who's been unmistakably good for a long time. Judged only on their own numbers, a long-tenured star can start to look overpaid. But the researchers also tracked something most performance reviews don't: the effect a star has on the people around them. That effect — the positive spillover onto peers' performance — gets stronger with age, not weaker.

"What we found was that older stars have a more positive impact on their firms and their colleagues than younger stars do."Dr. Matthew Call, Mays Business School, Texas A&M University — quoted in Texas A&M Stories, Aug. 12, 2026

Here's why that lands differently for you than for a VP reading it inside a Fortune 500 slide deck. A large company can afford to be wrong about one senior performer — the org chart absorbs it. You can't. In a company like yours, the person who's been there six years isn't just a producer. She's very likely the reason three other people know what they're doing, the one new hires ask instead of you, the informal answer to "how do we actually do this" that has never once been written down. If you're only measuring her against her own old numbers, you're running a review that misses the job she's actually doing now.

The part the research doesn't say, but that I'll say plainly: this is not a license to assume every veteran who's slowed down is secretly your most valuable employee. Some people plateau because they've checked out, not because they've quietly shifted into a role nobody assigned them. The study doesn't tell you which one you're looking at — it tells you that the scoreboard you're probably using can't tell the difference either. If the only number you track is hers, you have no way to know whether the flatline is decline or a job that changed without a new title.

The move this week isn't a raise or a talk about her numbers. It's an inventory. Separate from her own output, write down who on your team is measurably better at their job because of time spent with her in the last ninety days — a process she fixed, a hire she kept from quitting in month two, a client relationship she rescued that wasn't hers to rescue. If you can name two or three of those without straining, you're not looking at a plateau. You're looking at someone whose job quietly changed while her title didn't, and the conversation you owe her is about that, not about her chart.

If you can't name any, that's useful too. It means either the shift the researchers describe hasn't happened here, or it has and nobody's been watching for it. Either way, you don't know until you look somewhere other than her own numbers.

That's the same blind spot the Leadership Profile is built to surface: how you read people and pressure, as distinct from whether someone's numbers are up or down.