How to Use Performance Reviews for Promotion Decisions

Why Does a Strong Review Not Mean Someone Is Ready?

Performance reviews should inform promotion decisions without deciding them. A review answers whether someone is performing well in their current role. Promotion asks whether they can succeed in a different one, which needs separate evidence: work already done at the next level, not excellence at this one.

The two questions diverge most sharply where the next level is a different kind of work rather than more of the same. A senior engineer rated outstanding has proved they write excellent code and ship complex projects. The staff role asks them to set technical direction, influence teams they do not manage, and choose which problems not to solve. Their review contains almost no evidence about any of that, because none of it was their job.

Where levels are genuinely incremental, the review carries more weight. A support agent handling more volume at higher quality is doing more of the same work better, and promotion to senior agent is reasonably predicted by the review. The error scales with how different the next level is.

So the practical rule is not “ignore reviews”. It is that a strong review makes someone eligible for consideration and tells you almost nothing about readiness, and conflating the two is the most common structural flaw in promotion processes.

What Does a Review Tell You, and What Does It Not?

Laid out side by side, the mismatch is obvious enough that it is worth asking why reviews get used this way at all. The answer is usually that the review is the only document that exists.

What Each Process Answers

Two rows deserve attention. Compares against: a review compares someone to their role’s expectations, while promotion compares them to people already operating at the target level, which is a different and harder bar. And who should decide: a manager can assess performance in a role they supervise, but assessing readiness for a level they may not occupy themselves is a judgment that needs people who do.

Where Should Promotion Evidence Come From Instead?

Five sources, in descending order of predictive value. The review is on the list, near the bottom.

1. Work already performed at the next level. The strongest evidence by a wide margin. Someone who has already led a cross-team initiative, mentored without being asked, or made a call that was theirs to make has demonstrated the thing rather than been assessed on the possibility of it.
2. judgment under ambiguity. How they behaved when the right answer was unclear and nobody told them what to do. Most senior levels are defined by this, and it is almost never captured in a review template.
3. Testimony from people at the target level. Peers one level up can assess readiness in a way a manager at a different level cannot. Their view is also the one that determines whether the promotion is accepted socially.
4. Scope handled voluntarily. Did they take on more surface area without being assigned it, and did the quality hold? Volunteering is a signal; sustaining is the evidence.
5. The performance review. Confirms they are not being promoted out of a role they are failing at, and supplies the record of delivery. Necessary, not sufficient.

The first four need deliberate collection, because nothing in an annual cycle gathers them by default. A manager who wants someone promoted next year should be creating opportunities for next-level work this year, which is what a career development plan is actually for.

How Do You Spot Someone Ready Before They Have the Title?

The most reliable test is simple and uncomfortable: are they already doing the job? Promotion at its best recognizes a change that has already happened rather than authorizing one that might.

Five observable signals, all of which precede the title:

• People route around the org chart to reach them. Colleagues bring them problems that are not formally theirs, because the answers are good. This is the clearest single signal and it cannot be faked by ambition.
• They decline work with a reason. Saying no to something low-value, and explaining the trade-off, is a senior behavior. People operating below their level say yes to everything.
• Their questions change. They stop asking what to do and start asking why this rather than that. The shift from execution questions to prioritization questions is usually visible months before a promotion case.
• They make other people better without being asked. Not formal mentoring, which can be assigned, but the review comments and design feedback that nobody required.
• They surface problems early and unprompted. Including problems that reflect badly on their own work. Willingness to deliver bad news on time is a scope-handling signal that gets better as the level rises.

The counterpart is a signal that someone is not ready despite strong performance: their excellence depends on the problem being well-defined. Give them a clear brief and they are outstanding; give them ambiguity and they wait. That is not a criticism of a strong performer at their current level, and it is a reliable predictor of struggle one level up.

Why Do Strong Performers Fail After Promotion?

When a well-regarded person struggles after promotion, the cause is usually one of four things, and only one of them is a selection error.

• The promotion rewarded output, not capability. They were the most productive individual contributor, so they were made a manager, where productivity is no longer the job. This is the classic case and it is a selection error: the evidence used predicted the wrong thing.
• The new role was never defined. They were given a title and left to infer the job. Many “failed promotions” are people performing a role nobody specified, judged against expectations nobody wrote down.
• Nothing was taken off their plate. They now do the new job and the old one, which is a workload problem presented as a capability problem. Extremely common in the first six months.
• Support stopped at the announcement. The promotion was treated as a reward rather than a transition. A level change is a new job and warrants the onboarding any new job would get, which almost nobody provides internally.

Only the first is a decision failure. The other three are implementation failures that get retrospectively blamed on the decision, which is worth knowing because they are considerably easier to fix. Before concluding a promotion was wrong, check whether the role was defined, whether capacity was made, and whether anyone supported the transition.

What Does a Defensible Promotion Case Look Like?

One page, assembled through the year rather than written the week before the panel. Six components:

• The target level, stated as criteria. Not “senior” but the three or four specific things the level requires in your organization. Without this the panel assesses against private definitions.
• Evidence against each criterion, with dates. Two or three concrete instances per criterion. “Led the payments migration, Q2, including the decision to defer the reporting scope” is evidence. “Consistently demonstrates leadership” is an assertion.
• What they did that was not their job. The next-level work, named specifically. If this section is empty, the case is not ready regardless of how strong the review is.
• Testimony from someone at the target level. Two or three sentences, attributed. Panels weight this heavily and correctly.
• The performance record. Ratings and goal outcomes, included to confirm sound delivery rather than to carry the case.
• The honest gap. What they have not yet demonstrated, and the plan for it. A case with no stated gap reads as advocacy and gets discounted; a case that names the gap and addresses it reads as judgment.

That last item is the one managers resist and panels value most. Including a gap does not weaken a case. It establishes that the person making the case can assess fairly, which is exactly what the panel is trying to work out about the manager as well as the candidate.

Should Promotion and Review Cycles Run Together?

Most organizations run them together because it is administratively convenient. It costs more than it saves.

Running them together means the review conversation is dominated by the promotion question. Employees arrive wanting to know about advancement, so developmental feedback is heard as either justification or consolation. Managers soften honest assessments to avoid damaging a case, or inflate ratings to support one. The review stops being useful in the same cycle it becomes consequential.

Separating them costs calendar time and buys two honest conversations. A common pattern that works: reviews in December, promotion decisions in February or March with their own panel and their own evidence pack. The review informs the promotion case without being the promotion case.

If they must run together, two mitigations help. Hold the development conversation in a separate meeting from the rating and promotion outcome. And require the promotion case to include next-level evidence the review does not contain, which forces the two to stay distinct even when the calendar does not. The underlying design question belongs with the rest of the framework decisions, since decoupling promotion from review is exactly the kind of choice that has to be made once and held.

For managers delivering the outcome either way, the conversation itself is covered in performance management for managers, including why blaming calibration is the worst available option.

Where Does the Data for This Come From?

A promotion case assembled the week before a panel will be built from recent memory, which is precisely the bias it needs to avoid. The evidence has to accumulate through the year. Profit.co’s performance management keeps goal outcomes, feedback and development records on one employee record.

See how Profit.co turns a year of evidence into a promotion case. Book a demo.