How to perform a talent review in 2026 [in 5 simple steps]

A talent review is a structured meeting where leaders assess every employee on two questions: how well they perform today, and how much they could grow. Done right, it replaces scattered manager opinions with one calibrated view of your people, so you can promote the correct person, name a backup for every critical role, and fix skill gaps before they turn into a resignation or a failed project.

It matters more now because the job itself keeps moving. The World Economic Forum estimates that 39% of a worker’s core skills will change by 2030, so a rating that was true 18 months ago is often stale. This guide walks through what a talent review is, the 5-step process, the frameworks worth using, and how to run one for existing employees without it turning into a box-ticking exercise.

TL;DR

• A talent review scores employees on performance and potential, then turns that into promotion, succession, and development decisions your leadership team agrees on.
• Run it in 5 steps: set objectives, gather evidence, calibrate in the meeting, build development and succession plans, then follow up.
• The 9-box grid is the common map, but the value comes from the calibration conversation, not the grid itself.
• A talent review looks forward at potential; a performance review looks back at results. You need both, and they should stay separate meetings.
• Base the ratings on evidence (skills data, project outcomes, reviewer input), not one manager’s gut. That is where most reviews go wrong.

What is a talent review?

A talent review is a recurring process where managers and HR sit together and rate each employee on current performance and future potential, then align on what to do next: who to promote, who to develop, and who is ready to step into a bigger role. It is a group calibration, not a one-on-one. The point is to get leaders comparing people against the same bar instead of grading on their own private scale.

Think of it as the difference between five managers each guessing who their strongest people are, and those same five managers in one room, defending their ratings with evidence and challenging each other’s blind spots. The second version is far harder to game, and it surfaces the quiet high performer that one manager kept forgetting to mention.

Why run a talent review?

Because most companies do not actually know who their next leaders are. In DDI’s Global Leadership Forecast 2023, only 12% of organizations felt confident in their leadership bench, and the share of leaders who rated their company’s leadership quality as high fell to 40%. A talent review is how you find and grow that bench on purpose instead of hoping it appears.

The payoff shows up in four places:

• Succession that is ready before you need it. When a director resigns, you already know the two people who could cover the role and what each still needs to learn.
• Retention of the people you cannot lose. Replacing an employee costs one-half to two times their annual salary, per Gallup. Spotting a flight-risk high performer early is cheaper than backfilling them.
• Development aimed at the right people. Instead of spreading a training budget evenly, you invest it where potential is high and a small nudge unlocks a lot.
• Fairer promotion calls. Calibration across managers reduces the “loudest advocate wins” bias that quietly shapes who moves up.

There is a cost angle too. External hiring is not cheap: SHRM benchmarking puts the average cost per hire at $5,475. Every role you fill from a talent pipeline you built internally is a role you did not have to source, screen, and onboard from scratch.

What is the talent review process?

Here is the talent review process most teams can run in a single quarter, then repeat. The steps are simple. The discipline is in doing them with real evidence and honest conversation, not rushing to fill in a grid.

Step 1: Set the objective and scope

Decide why you are doing this before you invite anyone. Are you building a succession bench, planning next year’s promotions, or finding skill gaps ahead of a reorg? Pick one primary goal. Then set scope: which teams, which roles, and how the output gets used. A review that tries to do everything for everyone produces a grid nobody acts on.

Step 2: Gather performance and potential evidence

Give every manager the same inputs before the meeting: recent goal results, a skills read, project outcomes, and any reviewer or peer feedback. Performance is the easier axis because there is a track record. Potential is where people guess, so define it in advance. Potential usually means the ability to grow, to handle ambiguity, and to succeed at the next level up, not just being good at the current job.

This is also the step where a skills-based view beats a resume-based one. What someone did five years ago says less about their potential than what they can demonstrably do right now.

Step 3: Calibrate in the talent review meeting

This is the heart of it. Each manager presents their ratings and, more importantly, the reasons, with examples. Other leaders question the placement and offer their own view. An HR facilitator keeps the standard consistent so a “high performer” means the same thing on every team. Expect to move people as the conversation exposes cases where one manager was too generous or too harsh.

Pro tip: Cap each calibration meeting at 60 to 90 minutes and no more than 20 to 25 people per session. Past that, the discussion gets shallow and managers start rubber-stamping to finish on time. Split a large org into role-family sessions and roll the results up afterward.

Step 4: Build development and succession plans

A rating with no follow-up action is wasted effort. For each person, agree on one concrete next step: a stretch project, a mentor, a role to prepare for, or a specific skill to close. For critical roles, name at least one ready-now and one ready-later successor. Write down who owns each action and by when, or the plan evaporates the moment the meeting ends.

Step 5: Follow up and measure

Revisit the actions in 90 days, not next year. Did the stretch assignment happen? Did the flight risk stay? Track a few outcomes over time: internal promotion rate, regretted attrition among your top group, and how many critical roles now have a named successor. A talent review earns its place only if last quarter’s decisions actually moved.

Which talent review framework should you use?

Most talent reviews use one of three maps. The 9-box grid is the best known: a 3×3 grid that plots performance on one axis and potential on the other, sorting people into 9 cells from “risk” in the low-low corner to “star” in the high-high corner. It gives leaders a shared visual and a quick way to spot where the bench is thin.

Here is the honest caveat: the grid is a prop, not the point. A box does not develop anyone. The value is the conversation the grid forces and the actions that follow. This is where an evidence-based scorecard helps. Combine skills assessment results, reviewer ratings, and structured feedback into one consistent view, and let a person make the final call rather than a tool. AI can structure the evidence; your leaders still decide. That keeps a talent review defensible instead of reducing someone to a single number.

Talent review vs talent evaluation: what’s the difference?

These terms get mixed up, and the confusion causes real damage when a growth conversation gets tangled with a pay decision. A talent evaluation is the act of measuring one person against role criteria. A talent review is the group calibration that compares those evaluations across the org. A performance review looks backward at what someone delivered, usually tied to pay. Keep them separate: the moment potential ratings feel like they affect this year’s bonus, managers stop being honest.

Talent assessment process for existing employees

Running a talent assessment for existing employees is not the same as screening a candidate, but the evidence problem is identical: you want a fair, skills-based signal, not a hunch.

The talent assessment process here is lighter touch. Use a short skills check or a structured self-and-manager rating against the competencies the next role needs, then feed that into the review as data rather than opinion. It works best when people understand it supports their growth, not a hidden ranking.

Map the competencies first, then pick the evidence for each, the same way you would build a skills mapping framework. Skills-based decisions pay off: McKinsey found that hiring for skills is five times more predictive of performance than hiring for education, and the same logic applies to promoting from within.

Key takeaways

• A talent review is a group decision, not a form. The calibration conversation is what makes ratings honest, because managers have to defend their calls to peers. Skip the meeting and you just have a spreadsheet of private opinions.
• Separate potential from performance, and both from pay. A talent review looks forward; a performance review looks back. Mixing potential ratings into bonus decisions kills the honesty the whole exercise depends on.
• Evidence beats gut, every time. Base ratings on skills data, project outcomes, and reviewer input. A skills-based read is far more predictive than tenure or a resume, and it is what makes a promotion call defensible later.
• The 9-box grid is a starting point, not the work. Plotting people is easy; the value is the development and succession action that follows each placement. A grid with no follow-up is theater.
• Name successors before you need them. With only 12% of organizations confident in their leadership bench, an empty pipeline is the norm, not the exception. Every critical role should have a ready-now and a ready-later name attached.
• Close the loop in 90 days. Revisit each action within a quarter and track internal promotion rate and regretted attrition. A review that nobody follows up on trains your managers to treat the next one as busywork.

Frequently asked questions