9-box grid and talent review FAQ
Answers from the 9Box team · Last updated 22 August 2026
Twelve questions people actually ask when they are about to run a 9-box talent review, answered properly. If you are starting from scratch, read what the 9-box grid is first, then how to run the session.
What is a 9-box grid used for?
It is used to compare a group of employees on two things at once — how they are performing now, and how much larger a role they could grow into — so that several managers can agree one comparable view of the population. The output is not the picture. It is the decisions that come out of the discussion: who is a retention risk, which senior roles have no credible successor, who needs a stretch assignment, and who needs a conversation that has been put off. In practice most organisations use it for three jobs — succession planning, allocating development budget, and identifying the people they cannot afford to lose.
How do you measure potential?
You cannot measure it the way you measure performance, and pretending otherwise is the main reason 9-box reviews go wrong. What you can do is agree a small number of observable signals in advance and rate against those. The ones that hold up best are learning speed, behaviour when the problem is genuinely unfamiliar, whether the person's influence already reaches beyond their own remit, and whether they actually want a larger job — that last one is skipped constantly and it invalidates everything above it. Write the definition down before anyone is rated, give an example of what high potential looks like in your organisation, and name what it is commonly confused with: confidence, visibility, articulacy and availability are not potential.
What is box 9 on a 9-box grid?
Box 9 is normally the top-right box — high performance and high potential — usually called Star or High Potential, holding the people who are delivering now and are ready for materially bigger work. Be careful with the numbering, though. There is no standard for it, and organisations number the grid in different directions, so box 9 in one company means the opposite of box 9 in another. If someone hands you a grid with numbers on it, ask which corner is which before you read anything into it. 9Box names every box rather than numbering it, for exactly that reason.
What is the difference between a 9-box grid and a performance review?
A performance review looks at one person against the expectations of their own job over a period that has just ended, and it ends in a conversation with that person. A 9-box review looks at a whole group at once, adds a second axis the performance review does not have — potential — and ends in decisions about development, succession and risk that are usually not quoted back to the individual verbatim. They are complementary rather than alternatives. Most organisations run the 9-box review just after the performance cycle, using the performance ratings as the input to one axis and a separately defined judgement for the other.
How many people should be on one 9-box grid?
Few enough that everyone on the grid can sensibly be compared with everyone else, which in practice means one function, one level band, or one leadership team's direct reports. Somewhere between about fifteen and eighty people is comfortable: below that the distribution tells you very little, and above it the calibration meeting runs out of time before it reaches the middle of the grid. If you need to cover a larger population, run several grids and roll the conclusions up. Putting a graduate cohort and a senior leadership team on the same three-by-three produces a meaningless picture, because high potential means something different in each.
Who should be in the talent review meeting?
The managers of the population, all at the same level, because they own the placements and do the talking. Their common leader, chairing, because someone has to break ties and it has to be someone whose decision the room accepts. An HR partner, facilitating — holding the definitions steady, asking for the evidence, and surfacing the patterns across the grid that nobody in the room wants to name. And someone capturing actions as they are agreed. Keep it under ten people, never invite anyone to a session where their own manager is discussing them, and resist adding a wider audience for visibility, because attendance changes what people are willing to say and candour is the whole point.
How long does a 9-box calibration session take?
Budget roughly two to three minutes per person, plus twenty minutes at the start for the definitions and the ground rules and twenty at the end for actions. For a grid of thirty people that is about ninety minutes. The two failure modes are both about pacing: spending forty minutes on the first three names and then racing the rest, or booking an hour for eighty people and calling the result calibrated. If the population will not fit in the time, split it and hold two sessions rather than shortening the discussion.
What do you do after the talent review meeting?
Three things, in this order. Turn every placement into an action with a named owner and a date — a development plan, a stretch assignment, a role change, or a conversation that has been avoided — because a grid with no actions attached is a meeting that produced a picture. Extract the three lists that actually move: retention risks, senior roles with no credible successor, and the overdue difficult conversations. Then decide deliberately what employees are told, apply that decision consistently, and store the grid with the same care as any other performance record.
Can you use a 9-box grid for succession planning?
Yes, and it is one of the things the grid does best — but as an input, not as the plan. The grid tells you where your capable and willing people are. A succession plan additionally needs the roles you are planning for, a named successor or two per role, and an honest view of how ready each of them is and when. The mistake to avoid is reading readiness off the grid: high potential means someone could grow into a larger role, not that they could do a specific job next quarter. Keep readiness as a separate judgement, recorded against the role rather than the person, and be explicit about the gap between the two.
Should you use a 9-box grid Excel template or a dedicated tool?
A spreadsheet template works right up to the moment the meeting starts and someone wants to move a person, at which point you are cutting and pasting text boxes in front of the room. A purpose-built grid gives you the nine boxes already named, drag-and-drop placement that saves as you drop, a note attached to each person, and a PNG or PDF export for the pack afterwards. That is what 9Box is, and it is a paid tool rather than a free one — $49 a month or $490 a year — although billing is not switched on yet, so an account created today has the whole thing while we finish building it. If you would rather stay in a spreadsheet, the thing to fix first is not the formatting: it is the box names. Agree them in writing before anyone is rated, so that every manager means the same thing by the same word.
Is a 9-box grid personal data, and how should you store it?
Yes. A completed 9-box grid is a list of identifiable people next to written judgements about their performance and their prospects, so it is personal data and it carries the same obligations as any other performance record — a lawful basis, a retention limit, access restricted to the people who genuinely need it, and the possibility that an employee asks to see what you hold about them. Write the grid on the assumption that the person on it will eventually read it. Keep it out of general-access drives and shared inboxes, and delete old cycles on the same schedule as the rest of your talent records. This is general guidance rather than legal advice, so have your data protection lead confirm your retention period.
Should you use the 9-box grid to decide promotions or redundancies?
Not on its own, and for redundancy selection you should be very careful indeed. The grid is a snapshot of a group at one moment, produced by a discussion, and its potential axis is a judgement with no agreed definition — none of which makes it a defensible selection criterion by itself. For promotion it is a reasonable input to a shortlist, alongside evidence of readiness for the specific role. For redundancy, use documented, objective and consistently applied criteria and take proper advice, because a grid position that was never shared with the employee and was never intended as a rating is a poor place to start.
Answered elsewhere
Stop rebuilding the grid in a spreadsheet
Nine pre-named boxes, drag-and-drop placement, a note on every person, and a PNG or PDF for the meeting. Private to your account. $49 a month or $490 a year — and while we finish building billing, signing up gives you the whole thing.
Build your first grid