HR articles

9-Box Grid: What Each Box Means, a Template and How to Run a Talent Review

2026-10-10 17:57
SHORT ANSWER

The 9-box grid is a 3×3 matrix that places each employee by two ratings: performance in the current cycle and potential to take a bigger role in two to three years, each low, medium or high. Every box gets its own action, from a stretch assignment and a successor slot in the top right to an improvement plan in the bottom left.

Course on this topic: Talent Management Course

I have worked on talent pools and succession in a bank, a TV channel and a game developer, and everywhere the same question came up: we cannot develop everyone the same way, so who gets what? The 9-box grid is the simplest honest answer I know. It is not a ranking of good and bad people. It is a map of different actions for different people.
Below: the two axes, all nine boxes with what to do in each, a template you can copy into a spreadsheet, and the way I run a talent review with it. If you only need the answer for one person, go straight to the interactive 9-box grid.

What is the 9-box grid?

The 9-box grid is a 3×3 matrix. You may also hear “nine-box matrix” or “performance–potential matrix”; it is the same tool. Performance goes along the bottom, potential goes up. Each axis has three levels: low, medium and high. So every person gets two ratings, not one, and lands in one of nine boxes.
I think of it as a coach's depth chart: who plays now, who is ready next season, and who needs more practice. In practice it is the main tool of a talent review meeting, where managers look at their people together, and the usual source of names for the talent pool and the succession plan.
One warning before the boxes. Companies name and number the boxes in different ways, so do not argue about the labels. Think in terms of the two axes and you will read any version of the grid.

The two axes: performance and potential

Performance looks back. It is about results in this cycle: how well the person meets or beats their current goals. It comes from the calibrated performance rating, not from a new opinion formed in the talent review meeting. That is why calibration of ratings goes first and the grid comes second.
Potential looks forward. It is the ability to take a bigger role in two to three years. I read the three levels this way:
  • Low — the person is not ready for a higher role. They can be excellent where they are.
  • Medium — ready for something more serious and complex than today.
  • High — can move two or three steps beyond the current position.
Potential is the harder axis, so judge it from evidence, not impressions. The signs I use in my courses: the person learns fast and changes approach, sees their own growth areas and works on them, thinks beyond their own department, handles ambiguity, leads others, and actually wants a bigger role. “A manager took over a new site with no process and built one in three months” is evidence. “He looks like a leader” is an impression. In the talent review every potential rating comes with an example.
And do not confuse potential with polished skill. Someone can look brilliant simply because they have done the same work for years and it runs on autopilot. That is mastery of the current role. It tells you nothing about the next one. A top scorer in the junior team is not automatically ready for the senior league.

What the nine boxes mean and what to do with each

These are the box names from the worksheet I give in my screening course, with the actions I teach in the HR Director, succession planning and certification-prep lessons. Performance is listed first, potential second.
Top row — high potential
  • Star (high performance, high potential). Your future leaders. Action: a stretch assignment and a successor slot. Develop them through real work — a task a bit bigger than the role, a few months in another function, a mentor who has already walked this road.
  • Growth Employee (medium performance, high potential). Develop and stretch. Find what holds the results back and work on that first.
  • Enigma (low performance, high potential). Untapped talent. Something blocks the person: the role, the manager, or they are simply new. Action: coach and check the role fit before you conclude anything.
Middle row — medium potential
  • High Performer (high performance, medium potential). A strong contributor. Stretch and develop inside the current area.
  • Core Player (medium performance, medium potential). A reliable team player with stable results who can develop. Most people sit in the middle boxes, and that is normal.
  • Dilemma (low performance, medium potential). An inconsistent performer. Observe, find the cause, give support or try a rotation.
Bottom row — low potential
  • Trusted Professional (high performance, low potential). Low potential is not an insult. These people are the backbone of the company: strong where they are, and often not interested in a bigger role. Keep, reward, recognise and offer the expert track. Do not push them into management and do not ignore them.
  • Effective (medium performance, low potential). Does the job and does not grow. Observe and keep the standard clear.
  • Underperformer (low performance, low potential). Perhaps the person is in the wrong place. Action: an improvement plan or an exit.
Who goes into the talent pool? A large company with many employees may take only the top-right box. A smaller company that needs more candidates takes the four boxes of the top-right corner: Star, High Performer, Growth Employee and Core Player.
FREE TOOL

Interactive 9-box grid

Pick performance and potential, or click a box. You get the box name and what to do with the person.

Potential ↑
Performance →
Star
High performance · high potential. Strong results now and able to take a bigger role.
What to do: give a stretch assignment and name the person as a successor for a key role.

The full version — the talent pool programme, successor readiness and development plans — is in the Talent Management Course.

FROM MY PRACTICE · STB

When I worked at STB, a TV channel in the StarLightMedia group, we were growing very fast. We did not have enough external candidates, or they lost to our internal people. And television gives people room for trial and error.

So many times we gave people a chance to try themselves in a higher role. And often the people who seemed to have no potential at all could do it. Today, when I look back at many of the leaders on that channel, I remember them as rank-and-file employees, and about many of them you could not have said they would achieve anything like this.

A low potential rating is a guess, not a fact. We underestimate people systematically, so review the rating and give people a chance to prove it wrong.

— Mike Pritula

9-box grid template you can copy

You do not need special software. A simple spreadsheet works for one team. Draw the 3×3 grid on one sheet, and keep one row per person on another with these columns:
  1. Name and current role
  2. Performance — low, medium or high, from the calibrated rating
  3. Potential — low, medium or high
  4. Evidence line — the facts behind the placement
  5. Box — one of the nine names above
  6. Readiness — ready now, ready in 1–2 years, ready in 3+ years
  7. Development gap and action — an assignment, a relationship (mentor or coach), a course
  8. Owner and next review date
The evidence line is the column people skip and the one that matters most. If you cannot write with facts why this person is in this box, the placement is not ready yet. For the next step after the grid, see my succession planning template.

How to run a 9-box talent review

  1. Calibrate performance ratings first. One manager rates everyone high, another gives everyone a three, a third is strict with everyone. Put them in one room and it becomes visible very fast. The calibrated rating is the one that goes into the grid.
  2. Collect data from more than one source. Performance review results, interviews, feedback. One source always gives a one-sided picture.
  3. Ask each manager for a draft placement with an evidence line. An example for every potential rating, no exceptions.
  4. Run a panel calibration. HR and managers from different departments discuss people together, so each manager hears how the others rate, and their shared senior leader can say: I think you rated this one too high.
  5. Use a facilitator. An HR business partner or an external consultant keeps the discussion on facts and reduces bias.
  6. Record the conclusions and build development plans from them. The grid is a tool for decisions, not for reports.
  7. Review it regularly. In my succession planning course the rhythm is a talent review every six months and a full audit of the plan once a year.
Some inputs must never go into a box placement: age, retirement eligibility, leave taken, an accommodation used. In the United States these sit behind the ADEA, the FMLA and the ADA, but the principle is good practice in any country. And tell people about the plan. A high-potential employee who never hears about it may go to a company that talks about growth openly.
FROM MY PRACTICE · STB

A TV channel is, roughly, thirty television projects that work like thirty independent companies: different conditions, different production technology, different managers.

I saw this situation many times: a person was an A-player and slid down to a C-player, or was a C-player from the start. They move to another project — the same job title — and there they simply blossom. Some people moved several times before they blossomed.

If a person has potential but performs poorly, the cause is often the environment, not the person. Try another role before you draw conclusions.

— Mike Pritula

Mistakes and limits of the 9-box grid

  • Ratings by feeling, without discussion. One manager alone sees only part of the picture. That is why you calibrate as a group.
  • Replacing potential with loyalty. “She has been with us for a long time, so let's grow her” is tenure, not potential.
  • Ignoring current weak spots. It is easy to get carried away with forecasts and forget that the person does not deliver today.
  • Not discussing the ratings openly. Then the grid turns into a formality: values in the cells of a spreadsheet that change nothing.
  • Treating a box as a label. Labels stick to people. The grid is a dialogue about potential, opportunities and risks.
  • Forgetting that potential is a guess. You will rate someone high, start preparing them, and it will not work out. You will miss someone, and later they will show more than you expected. Both mistakes are allowed, as long as you come back and correct the grid.
One question to take to your next review: who in your team shows an average result today but could grow if you developed them properly?
I teach the whole chain — assessing potential, the talent pool programme, succession planning and career management — in the Talent Management Course.
FAQ

Frequently asked questions

What is the 9-box grid?

It is a 3×3 matrix that places every employee by two ratings: performance in the current cycle and potential for a bigger role. Each of the nine boxes has its own action, so the grid works as a map of what to do with whom.

What are the two axes of the 9-box grid?

Performance and potential. Performance looks back: results this cycle, taken from the calibrated rating. Potential looks forward: the ability to take a bigger role in two to three years, judged from evidence.

How do you assess potential for the 9-box grid?

From evidence, not impressions: how fast the person learns, how they handle ambiguity, whether they lead others, and whether they want a bigger role. Every potential rating should come with a concrete example.

How often should you run a 9-box talent review?

In my succession planning course the rhythm is a talent review every six months and a full audit of the plan once a year. A potential rating is a guess, so it has to be reviewed regularly.

Should employees know their box in the 9-box grid?

Tell people the plan, not the label. A high-potential employee who never hears about the plan may leave for a company that talks about growth openly, while a box name on its own sticks to a person and explains nothing.