FrameworkTalent ReviewsSuccession Planning

What the 9-Box Talent Grid Actually Tells You - and What It Doesn't

The 9-box talent grid is one of the most widely used tools in talent reviews. It is also one of the most consistently misread. This article explains what it actually measures, where it breaks down, and how to use it as a genuine strategic tool rather than an annual calibration exercise.

10 min read
Peopletree Group
HR Leaders & Executives

The 9-box talent grid is one of the most widely adopted tools for assessing talent. It plots employees across two axes - performance and potential - and produces a snapshot that is supposed to guide succession planning, development investment, and leadership pipeline decisions.

In practice, the grid is frequently misread, misapplied, or treated as an end in itself rather than a starting point. This article explains what the 9-box actually measures, where it breaks down, and how to use it as a genuine strategic tool rather than a calibration exercise that ends in a spreadsheet.

What the 9-box measures

At its core, the 9-box grid categorises employees across two dimensions:

Performance

How consistently someone delivers results in their current role. It is backward-looking - a record of what has been achieved.

Potential

Capacity for growth. Most reliably assessed through learning agility - how quickly someone absorbs new information and adapts to complexity. It is forward-looking and, by definition, harder to measure than performance.

The 9-box talent grid: performance vs. potential
The 9-box talent grid: performance × potential. Each cell represents a distinct talent profile.

The real value of the 9-box grid is not in labelling individuals - it is in showing you the shape of your leadership supply relative to your organisation's demand.

Using the grid strategically

The most common mistake organisations make with the 9-box is treating it as a static snapshot. Used well, it is a leadership forecasting tool - one that should shift in response to your business environment.

Consider two scenarios. A company entering a rapid growth phase needs a deep bench of high-potential leaders ready to step into expanded roles. A company in a contraction phase does not - over-developing leaders in a declining market creates flight risk, frustration, and wasted investment.

The grid allows you to calibrate your development focus accordingly. The three scenarios below illustrate what the target distribution should look like under different business conditions.

Rapid growth

When your business is scaling quickly, aim for 20–25% of your talent in the top-right quadrant (high performance, high potential) and 65–70% in the middle tier. You need to build leadership capacity ahead of demand, not in response to it.

9-box distribution for rapid growth: 20-25% high potential, 65-70% medium potential
Target distribution for rapid growth: 20–25% high potential, 65–70% medium potential

Stable, predictable growth

In a stable environment, the right distribution shifts to approximately 15% high potential, 75% medium potential, and the remainder in lower-demand development zones. The goal is to keep your bench strong without over-investing in development that outpaces your actual leadership demand.

9-box distribution for stable growth: 15% high potential, 75% medium potential
Target distribution for stable growth: 15% high potential, 75% medium potential

Contraction or negative growth

During a contraction phase, scale back to 5–10% high potential, with the majority of your talent in medium potential areas. Too many high potentials in a shrinking organisation creates a retention problem - people who are ready for more will leave if more is not available.

9-box distribution for contraction: 5-10% high potential, 80-85% medium potential
Target distribution for contraction: 5–10% high potential, 80–85% medium potential
Business ConditionTarget % in High PotentialWhy It Matters
Rapid Growth20–25%You need to scale leadership capacity ahead of demand.
Stable / Predictable15%Keep your bench strong, but don't over-invest.
Contraction / Decline5–10%Too many high potentials = flight risk.

Two misunderstandings that undermine the grid

1. Potential is not the same as promotability

The most common error in 9-box calibration sessions is conflating potential with promotability. They are not the same thing.

Promotability

Readiness to step into a new role now. A near-term assessment based on current capability gaps.

Potential

Capacity to grow into larger, more complex roles over time. A longer-range assessment based on learning agility.

An employee can be highly promotable today - ready for the next role - but have limited upward potential beyond that. The grid helps you see this distinction and avoid two common errors: over-promoting people who are ready now but will plateau quickly, and overlooking people who are not yet ready but have significant long-term upside.

2. The opposite of high potential is not low potential

What happens to the employee placed in the top-left cell - high performance, low potential? In most organisations, that label carries an implicit message: "We value what you do, but we don't see a future for you here."

"The opposite of high potential is not low potential. It is high professional."

The High Professional

High professionals are not on a leadership trajectory, but they deliver deep expertise, build long-term client and internal relationships, and accumulate institutional knowledge that is genuinely difficult to replace.

Not racing to the top - they are specialists
Building deep expertise and long-term relationships
Accumulating institutional knowledge across transitions
Critical to day-to-day business operations

A well-designed talent review recognises both groups. High potentials need development investment and stretch assignments. High professionals need recognition, retention strategies, and roles that leverage their depth rather than pushing them toward breadth they do not want.

What the grid cannot do alone

The grid is a useful diagnostic. It is not a talent strategy.

If your talent review ends with a completed 9-box and no follow-up action plan, you have run a calibration exercise, not a talent process. The grid tells you where people are. It does not tell you what to do about it, how to develop them, or whether your pipeline is actually aligned to your business strategy.

To move from a snapshot to outcomes, the 9-box needs to be embedded in a broader process that includes:

1
A clear link between the grid output and your succession plan
2
Individual development actions for each quadrant, not just the top right
3
A mechanism for tracking whether development investments are producing movement on the grid over time
4
Calibration discipline to reduce the manager bias that makes most 9-box outputs unreliable

How Peopletree Group applies the 9-box

At Peopletree Group, the 9-box is one input into a broader talent review process - not the centrepiece of it. We use it alongside behavioural assessment data from Talentprint, capability frameworks from the Talent Genome, and people analytics from DataHub to produce a view of talent that is grounded in evidence rather than manager perception alone.

The six capabilities required to make a talent review process actually work are:

1Talent process design

A structured process that connects the grid to business priorities and succession decisions.

2Technology and data

A platform that collects, stores, and surfaces talent data consistently across the organisation.

3Data quality

Clean, comparable data that allows calibration across managers, teams, and geographies.

4Facilitation

Structured calibration sessions that surface bias and produce consistent assessments.

5Insight and action planning

Translating grid outputs into individual development plans and pipeline actions.

6Ongoing measurement

Tracking movement on the grid over time to assess whether development investment is working.

The 9-box is a starting point. What you do with it determines whether it becomes a genuine leadership forecasting tool or an annual exercise that produces a chart and little else.