The 9-box grid is a talent review tool that plots each employee on two axes, current performance and future potential, each split into low, medium and high, giving nine boxes. Teams use it to discuss development, succession and retention. It's only as fair as the evidence behind the potential rating.
This guide explains what the 9-box is, what each box usually means, and how to run a 9-box talent review with evidence and calibration. It includes a free Excel template. It also covers where the grid goes wrong — especially the bias risk in potential ratings — and what to use instead if you have outgrown it.
What is the 9-box grid?
The 9-box grid is a talent review tool that plots each employee on two axes, current performance and future potential, each split into low, medium and high, giving nine boxes. Teams use it to discuss development, succession and retention. It's only as fair as the evidence behind the potential rating.
Performance is usually a judgement of what someone has already done against the expectations of their role. Potential is a forecast of how far they might go. In a typical grid, the horizontal axis is performance (low, medium, high) and the vertical axis is potential, with high potential at the top. The result is nine combinations, from low performance and low potential through to high performance and high potential.
Companies often give the boxes informal names — “star”, “core player”, “rough diamond” and so on — but those labels are not standard. They vary by consultancy and by employer, so this page uses the two axes as the primary labels. You will also hear “9-box model” and “talent grid”. In this playbook they mean the same tool: a 3-by-3 plot of performance against potential, used in a talent review.
HR and talent leads mainly use the grid in a talent review or succession discussion: who to stretch, who to retain, who needs a clearer plan, and where the bench is thin. A 9-box is a conversation aid. It is not a performance-management system on its own, and it should not replace a written rating scale, a completed staff appraisal or scorecard, or a proper calibration. For how those pieces fit together, see the performance management framework.
The nine boxes explained
Read the grid from the top right: high performance and high potential. The typical actions in the table are playbook suggestions for a talent-review conversation, not research findings and not a prescribed process.
| Pot.Perf. | Low | Medium | High |
|---|---|---|---|
| High | Low performance, high potentialDiagnose first. Time-boxed plan. | Medium performance, high potentialStretch, with evidenced potential. | High performance, high potentialRetain and stretch. Not an automatic promotion. |
| Medium | Low performance, medium potentialA clear performance plan, with support. | Medium performance, medium potentialThe core of most teams. Keep developing. | High performance, medium potentialDeepen expertise. Do not force a management path. |
| Low | Low performance, low potentialA formal plan and a fair process. | Medium performance, low potentialBe honest about the role. Do not write them off. | High performance, low potentialRetain the specialist. Recognise the work. |
| Box | What it usually means | Typical talent-review action |
|---|---|---|
| Low performance, high potential | Below the current bar, but there is a written case that the person could do a bigger role — for example they are new in post, blocked, or still ramping. | Diagnose before you label. Set a time-boxed plan with support, and a date to review the performance record again. |
| Medium performance, high potential | Doing solid work, with evidenced signs of readiness for a bigger role. | Coaching and stretch work. Check that the “high potential” rating has a dated example, not just confidence or visibility. |
| High performance, high potential | Consistently strong against the current bar, and already showing the skills of a bigger role. | Retain, stretch, and discuss career path. This box is not an automatic promotion. |
| Low performance, medium potential | Below the current bar, without a strong evidenced case for a bigger role yet. | A clear performance plan, with support, and a written review point. |
| Medium performance, medium potential | The core of most teams: meeting the bar, with room to grow in the current path. | Steady development and recognition. Do not neglect this box in the conversation. |
| High performance, medium potential | Strong in the current role, without a strong evidenced case for a different or bigger path right now. | Deepen expertise, reward contribution, and do not force a management path. |
| Low performance, low potential | Well below the bar, and no evidenced case for a bigger role. | A formal plan and a fair process if performance does not improve. Follow your own policy and seek advice where needed. |
| Medium performance, low potential | Doing enough in the current role, with limited evidenced scope to go further in the way you have defined potential. | Be honest about the role, keep the person engaged, and do not write them off. Review the definition of potential, not only the person. |
| High performance, low potential | Excellent in the current role; the potential rating is low. Often a trusted specialist. | Retain and recognise. Do not punish someone for not wanting, or not yet showing, a different career path. |
Free 9-box grid template
The free 9-box grid template is an in-house Excel workbook with:
- a blank grid you can type names into
- a plotting log for evidence, draft and calibrated boxes, and the agreed action
- a one-page guide to the nine boxes
- a clearly labelled fictional worked example, which is not a real team and not statistics
Write down what performance and potential mean in your company before anyone is plotted. Collect evidence first, plot drafts in the log, then calibrate and record any change of box with a reason. The five rules below are also on the first sheet. You can work from the grid and table on this page if you would rather not download the file.
The template is free to copy and adapt. It is a starting point, not a required process, and it is not legal advice.
How to run a 9-box talent review
A 9-box talent review only works if the room is discussing evidence, not impressions. Use this running order, then the calibration meeting guide for the checks themselves. Keep the talent review separate from the pay review: the meeting can inform development, succession and retention plans, but the box should not be what sets a bonus or a salary. A calibrated performance rating and a merit matrix are a clearer basis for the pay review.
- Publish the criteria first. Write down what “performance” and “potential” mean, and share that with managers before they rate anyone. Potential should be observable signs of readiness for a bigger role, not a feeling about how far someone might go. The talent bar by seniority and rating scales are a solid starting point.
- Collect evidence before the room. Every person should arrive with a completed scorecard or appraisal, dated examples, and a draft box. If the evidence is not written down, the box is an opinion. The free tools include a review bias checker you can run on last cycle’s ratings before you meet.
- Plot drafts on the grid, then walk the exceptions. Do not invent boxes from memory in the meeting. Start with people whose potential rating contradicts their performance record, new joiners who may still be ramping, and any list of “high potential” that has no examples attached.
- Calibrate across managers. Compare how each manager has used the boxes. Ask for the evidence behind high and low potential calls, and record every change of box with a reason. See what performance calibration is and how to run a performance calibration meeting. You can practise the shape of the conversation first with the calibration simulator, which puts a fictional team on a rating scale and a 9-box.
- Agree an action for every person, and write it down. A box without an action is just a label. Stretch, retain, develop, plan, or recognise — and say who will follow up, and when.
- Review the grid again next cycle. A box is a snapshot. If last year’s label is still travelling into this year’s discussion, with no fresh evidence, it has become an identity.
For a wider view of where bias enters reviews, see performance review bias.
Where the 9-box goes wrong
The grid is quick, visual and familiar. It is also one of the easiest ways to bring bias back into a process you have worked hard to make fair — usually through the potential axis.
- Potential is an opinion, not evidence. Performance can be tied to goals and deliverables. “Potential” rarely has written criteria, so it defaults to the manager's impression: confidence, visibility, similarity to the people already at the top. The axis with the least evidence ends up deciding who gets developed or promoted. Our post on potential ratings and promotion bias sets out the problem in more depth, including the research below.
- It multiplies rater judgement. A performance rating already carries a lot of the rater's own habits of perception. Scullen, Mount and Goff, in the Journal of Applied Psychology (2000, volume 85, issue 6, DOI: 10.1037/0021-9010.85.6.956, as checked 13 November 2026), quantified how much of the variance in job-performance ratings comes from the rater's idiosyncratic tendencies rather than from the person's work. A 9-box asks the same rater for two subjective judgements and then combines them.
- Boxes become labels. Once someone is a “low-potential” box, it follows them, often without them knowing. The grid is a snapshot, but it gets treated as an identity.
- It rewards the most persuasive manager. In a talent review, people move boxes based on who argues best, not on what is written down. That is calibration theatre, not calibration (see calibration theatre).
- It isn't comparable across teams. A “high performer” in one department and in another were judged against different, unwritten bars. You can't use the grid to compare, promote or pay fairly across the company.
- It mixes up potential and readiness. What you usually need to decide is whether someone is ready for the next level. That is answerable with evidence: do they already show the next level's skills?
There is now direct research on what happens when potential ratings are allowed to outweigh the performance record. Alan Benson, Danielle Li and Kelly Shue, in “Potential” and the Gender Promotion Gap, American Economic Review (February 2026, volume 116, issue 2, pages 375–417, DOI: 10.1257/aer.20220831, as checked 13 November 2026), studied 29,809 management-track employees at a large retail chain. Women received higher performance ratings and lower potential ratings than men. Differences in potential ratings accounted for approximately half of the gender promotion gap. Those lower potential ratings were not accurate forecasts: women subsequently outperformed male colleagues, both on average and on the margin of promotion. The authors highlight two mechanisms: stereotyping, and strategic retention. This is one employer, in retail. It does not prove that every 9-box is biased. It does show why an unevidenced potential rating is a risky input to promotion. A fuller summary, including what to change in calibration, is in potential ratings and promotion bias.
If you keep the 9-box: five rules
If the grid is still useful as a conversation starter in your talent review, use it under these rules. They are playbook practice, grounded in the problems above, not a claim that a 9-box becomes unbiased once you adopt them.
- Define potential. Publish what it means before anyone is plotted. Use observable behaviours expected at the next level, not a feeling. The talent bar by seniority and competency matrix templates help you write that down.
- Require evidence. A high or low potential rating needs a dated example, just as an extreme performance rating does. If the evidence is not written down, the box should not survive the meeting.
- Calibrate. Compare draft boxes across managers. Spend time where potential contradicts performance, or where one manager’s “high potential” list looks very different from another’s. See performance calibration and how to run a performance calibration meeting.
- Review often. Revisit the grid in the next cycle. Do not let last year’s label travel quietly into this year’s promotion or pay discussion.
- Keep it separate from pay. Do not use the box to set bonuses or salary. Use evidenced, calibrated performance ratings for pay decisions. Use distributions of boxes as a prompt for discussion, not as a quota. Research on forced ranking is a warning here: Bond, in Cut to the Curve, Management Science (volume 72, issue 4, DOI: 10.1287/mnsc.2023.01204, as checked 13 November 2026), found that employees downgraded from the top ranks because of a strict cutoff were at least 34% more likely to leave voluntarily, even when the organisation tried to manage the fairness concern and paid larger bonuses than their top-ranked peers. For the line between a forced cutoff and quota-free calibration, see stack ranking vs calibration.
Replacing a 9-box talent review, or tightening the one you have? Talk it through on a 15-minute review, or get a picture of what your current cycle costs.
What to use instead
Many scale-ups replace the potential axis with a published talent bar by seniority and skills-based scorecards, then calibrate ratings with evidence. That keeps talent conversations grounded in what people have done, rather than a judgement of “potential”.
Three separate lenses
Score deliverables (speed, quality, complexity), role skills and culture separately, as in the framework. Someone strong on one and weak on another gets a specific development plan, not a box.
A talent bar by level
Compare each person with the written bar for their level, not with colleagues. See the talent bar by seniority and the competency matrix templates.
Readiness, not potential
Promotion eligibility comes from consistently meeting the bar and showing the next level's skills, with evidence, rather than from a manager's sense of potential.
Evidence-based calibration
Grades are calculated from scorecards before the meeting; calibration checks the outliers and the evidence. See what performance calibration is and how to run a calibration meeting, or practise first with the calibration simulator.
If you are still designing the cycle around the grid, start with the free tools and the framework, rather than with a blank 9-box. For the process itself — critical roles, ready-now and ready-later names, and a simple succession template that does not use a 9-box — see succession planning for People teams.
Sources, as checked 13 November 2026: Benson, Li and Shue (2026), “Potential” and the Gender Promotion Gap, American Economic Review 116 (2): 375–417, DOI: 10.1257/aer.20220831; Scullen, Mount and Goff (2000), Understanding the latent structure of job performance ratings, Journal of Applied Psychology 85 (6): 956–970, DOI: 10.1037/0021-9010.85.6.956; Bond (2026), Cut to the Curve, Management Science 72 (4): 3487–3507, DOI: 10.1287/mnsc.2023.01204. A longer write-up of the AER study is on potential ratings and promotion bias.
Related guides
Stack ranking vs calibration: why forced ranking fails and what to use · Performance management framework: scorecards, talent bar and calibration · employee performance rating scales explained · performance calibration · how to run a performance calibration meeting · staff appraisal template (UK) · potential ratings and promotion bias · performance review bias · talent bar by seniority · competency matrix templates · succession planning and talent reviews for People teams · merit matrix: how to link ratings to pay · free tools · People Performance Playbook home
FAQ
What is the 9-box grid?
The 9-box grid is a talent review tool that plots each employee on two axes, current performance and future potential, each split into low, medium and high, giving nine boxes. Teams use it to discuss development, succession and retention. It's only as fair as the evidence behind the potential rating.
How do you use the 9-box grid in a talent review?
Publish what performance and potential mean, collect evidence and draft boxes before the meeting, walk the exceptions, calibrate across managers, agree a written action for every person, and review the grid again next cycle. Keep the talent review separate from the pay review. The free template on this page has a blank grid and a plotting log for that work.
What are the problems with the 9-box grid?
The potential axis often has no written criteria, so it becomes a manager's impression. Boxes become labels, the room can reward the most persuasive manager, and ratings are hard to compare across teams. Research in the American Economic Review (February 2026, as checked 13 November 2026) found that lower potential ratings for women, despite higher performance ratings, accounted for approximately half of a gender promotion gap at one large employer.
What can you use instead of the 9-box?
Many scale-ups replace the potential axis with a published talent bar by seniority and skills-based scorecards, then calibrate ratings with evidence. That keeps talent conversations grounded in what people have done, rather than a judgement of “potential”.
Is the 9-box grid always wrong?
Not always. It can be a useful conversation starter in a leadership team that has written criteria and evidence. The problems start when its boxes drive promotion, pay or exit decisions, or when the potential axis has no criteria.
Should we use the 9-box for pay or promotion?
No. Keep it separate from pay. Use evidenced, calibrated performance ratings — and, for promotion, evidence of readiness for the next level — rather than a box.
Follow us on Google
Next step
Want to see what your review cycle costs? Get your number and a 1-page benchmark report by email, or talk it through on a 15-minute review.