The 9-Box Talent Review: How to Use It Without It Becoming a Caste System
The 9-box performance-vs-potential grid is the most-used and most-misused tool in talent management. The discipline that makes it useful — and the rules that…
On this page▾
- The 9-box is a discussion tool, not a verdict.
- Performance is past; potential is forward-looking — different evidence bases.
- Use anchored definitions for each axis, refreshed annually.
- Never share an individual's box placement with them — share the development plan that came out of it.
- Re-grid annually; people move, especially at the edges.
Developed at GE under Jack Welch and popularised across the Fortune 500, the 9-box plots performance (Y) against potential (X) on a three-by-three grid. The frame is durable; the failure modes are also durable, and HR experts spend a lot of time managing them.
What the 9-box actually measures
Performance is observed past delivery against the role's current scope. Potential is the assessed capacity to operate at greater scope or complexity — a forward bet. Conflating them is the most common error. A high performer at their current ceiling is not the same as a stretched performer with headroom.
Anchoring the axes
| Level | Performance (past) | Potential (forward) |
|---|---|---|
| Low | Inconsistent against role expectations | Best fit at current scope; unlikely to grow scope materially |
| Medium | Meets role expectations consistently | Could grow one level with development |
| High | Exceeds role expectations; positively influences peers | Could grow two+ levels; demonstrates scope-stretching behaviors today |
The nine cells and the moves
| Cell | Label (informal) | Standard action |
|---|---|---|
| High perf / High pot | Stars | Stretch, sponsor, retention focus, succession candidate |
| High perf / Med pot | Solid contributors | Recognize, broaden scope, retain |
| High perf / Low pot | Expert specialists | Recognize, deepen, dual-ladder if available |
| Med perf / High pot | Diamonds in rough | Coach, clarify expectations, accelerate |
| Med perf / Med pot | Core | Steady development plan |
| Med perf / Low pot | Solid but capped | Recognition, lateral moves |
| Low perf / High pot | Wrong role / wrong manager | Investigate fit before action |
| Low perf / Med pot | Underperformers | Performance plan, decide in 60 days |
| Low perf / Low pot | Action required | Exit conversation |
The rules that prevent misuse
- Never publish individual placements. Share the development plan, not the box.
- Calibrate across managers in the same room. Single-manager grids are unreliable.
- Re-grid every year — people move at the edges most.
- Bring evidence to potential ratings (specific behaviors, projects, stretch demonstrated).
- Audit the grid distribution by gender, ethnicity, age, tenure — bias hides in the corners.
Read next
All playbooksHow modern talent teams build succession plans that survive contact with reality — beyond the named-successor spreadsheet, into readiness, development, and…
Why ratings without calibration are leniency-and-stringency lotteries, and how to run a 90-minute calibration that produces decisions managers can defend to…
The system around the review matters more than the review itself. A modern approach to goals, feedback, calibration, and the conversation.