9-Box Grid Talent Management:How to Run a Talent Review That Actually Drives Decisions
The 9-box grid has been a standard HR tool since McKinsey developed it for GE in the 1970s. Most teams know what it looks like. Far fewer use it in a way that changes anything. Here is how to run a talent review that produces real action plans, not just a colorful spreadsheet.
The 9-box grid plots every employee on two axes: current performance and future potential. The intersection of those two ratings produces one of nine talent profiles, each with a different set of implications for development, succession, and retention. Done well, the output drives months of targeted people decisions. Done poorly, it is three hours of meetings and a spreadsheet nobody looks at.
The tool itself is not complicated. The hard part is the calibration. Getting a room of managers to agree on what high potential actually means, and then apply that definition consistently across their teams, is where most talent review processes break down. SHRM research consistently finds that organizations with formal talent review processes have 30-40% better succession bench depth than those relying on informal identification. The process matters even when the output is imperfect.
The 9-box connects directly to your workforce planning and internal mobility strategy. High-potential employees in boxes 1-3 are your promotion and succession pipeline. Strong performers in boxes 6 and 9 may be better suited for lateral moves or specialist tracks than management paths. And employees in box 7 require a decision: is this a performance issue, a role mismatch, or something the business needs to exit?
This guide covers the full process: how to define the two dimensions, how to run the calibration session, what action plans look like for each box, and the six mistakes that make talent reviews useless. For broader context on how talent reviews fit into your overall hiring and retention strategy, see our guide to talent acquisition strategy. On the external hiring side, McKinsey research consistently shows that companies with strong internal talent pipelines outperform peers on total shareholder return by 15-20%.
The Framework
The 9-box grid: what each box means
The grid reads performance on the horizontal axis (low to high, left to right) and potential on the vertical axis (low to high, bottom to top). This produces nine distinct talent profiles. The names vary by organization, but the logic is consistent.
The top-right box, Stars, contains your highest-priority talent. They perform at a high level today and show the potential to take on significantly more. Your retention and acceleration investments should be concentrated here. Stars represent roughly 3-7% of most workforces. If your talent review produces more than that, your criteria are too permissive.
The bottom-left box, Under Performers, contains employees who are both performing below expectations and showing limited growth potential. These are typically performance management situations. The honest answer is that most organizations wait too long to address this group, which drags down team performance and morale.
The middle of the grid, Core Players and Solid Contributors, represents the majority of most organizations. These employees are valuable. They keep the business running. But they are not your succession pipeline, and investing executive development budget here is a misallocation.
Running the Review
The four-step talent review process
Most organizations run talent reviews annually, often in Q4 ahead of compensation planning or in Q1 following performance reviews. The sequence matters. Rushing past calibration is the single most common failure mode.
Define Criteria
Set shared definitions of performance and potential before ratings begin
Manager Ratings
Managers independently rate each direct report on both dimensions
Calibration Session
HR facilitates cross-manager discussion to normalize ratings
Action Planning
Assign development plans, succession targets, and retention strategies
Step 1: Define your criteria (2-4 weeks before the session)
Before a single manager rates anyone, HR must publish clear definitions for both axes. Performance is usually the easier one, anchored in goal attainment and competency data from your review cycle. Potential is harder. Most organizations define it as the capacity to succeed in a role two levels above the current one within the next three to five years. You need behavioral anchors for each potential level. What does high potential actually look like in your specific context? Give managers three to four concrete examples before you ask them to rate.
Step 2: Manager ratings (1 week before the session)
Managers submit their ratings independently before the calibration session. This prevents anchoring, where the first person to speak in a room biases everyone else's view. Ratings should cover every direct report, typically on a three-point scale for each dimension. Managers should also flag anyone they are uncertain about and anyone who may need special consideration due to circumstance (new hire, extended leave, major role change mid-cycle).
Step 3: Calibration session (2-4 hours)
This is where the real work happens. HR facilitates a cross-manager discussion of ratings, starting with outliers: employees rated very differently by different managers who interact with them, and employees in the highest and lowest boxes. The goal is not consensus for its own sake. It is ensuring that a five-year veteran performing well for Manager A is rated on the same standard as a three-year employee performing well for Manager B. Without calibration, the grid reflects managers' opinions about their own teams, not a genuine talent map.
Step 4: Action planning (within 2 weeks of the session)
The calibration output is not the deliverable. The action plan is. Every employee in box 1 (Star) should have a succession target and a development plan with specific milestones. Every employee in box 7 (Under Performer) should have a performance improvement plan or an exit process starting. Everyone in between should have a clear conversation with their manager about what their box placement means for their career trajectory, without necessarily naming the box itself.
Rating Criteria
How to measure performance and potential separately
Conflating performance and potential is the most common 9-box error. A 20-year engineer who consistently delivers excellent work but has no interest in management is not a low-potential employee. They are a high performer with a different growth trajectory. Your criteria need to keep these two dimensions genuinely separate.
What the employee has delivered in their current role over the review period. Backward-looking.
- Goal attainment against agreed targets
- Quality and consistency of outputs
- Impact on team and business results
- Competency demonstration in current role
What the employee could achieve in roles of greater scope or complexity. Forward-looking.
- Learning agility: how quickly they absorb new skills
- Leadership behaviors beyond their current scope
- Adaptability under ambiguity or change
- Motivation to take on broader responsibility
Performance ratings should be grounded in data wherever possible: goal attainment percentages, output quality metrics, peer and stakeholder feedback from your structured review process. The more objective the performance input, the less the calibration session needs to do.
Potential ratings are inherently more subjective, which is why calibration matters more for that axis. Harvard Business Review research on high-potential identification suggests that learning agility is the single strongest predictor of leadership success at higher levels. It outperforms IQ, personality measures, and even past performance in studies of executives. If your potential criteria do not include some version of learning agility, add it.
One practical tip: ask managers to provide a specific behavioral example for each high-potential rating before the calibration session. "She took on the entire product launch when the PM left, restructured the timeline in two days, and shipped on schedule" is defensible. "I think he has a lot of potential" is not. The behavioral evidence requirement alone improves calibration quality significantly.
Action Plans
What to do with each talent profile
Stars (High Performance / High Potential)
- Name them as successors to specific roles in your succession plan
- Assign them to stretch projects that expose them to new domains
- Connect them with executive sponsors who are not their direct manager
- Have an explicit retention conversation: what would keep them here for the next 3-5 years?
- Consider accelerated promotion timelines. Waiting for tenure is how you lose these people.
Stars represent your most significant retention risk. According to LinkedIn Talent Solutions data, high performers are 40% more likely to be actively targeted by competitors. Your talent review should trigger an immediate retention review for this group.
High Potentials (Moderate Performance / High Potential)
- Identify whether underperformance is a skill gap, a motivation issue, or a role fit problem
- If skill gap: targeted development with a 90-day milestone to show improvement
- If motivation: explore whether role redesign or a different manager could shift engagement
- Do not leave them in the same situation hoping it resolves. That rarely works.
- Reassess within 6 months; potential ratings degrade without enabling conditions
This is the box most organizations mismanage. High potential with low performance is a signal that something specific is wrong. Find it and address it, or the person will either leave or slide into a lower box.
Core Players (Moderate Potential / Moderate Performance)
- Recognize their contribution explicitly. They often feel invisible.
- Provide steady development in their current domain
- Offer lateral moves to broaden their experience if they want it
- Do not write them off or overinvest executive development resources here
- Focus on engagement and satisfaction rather than advancement
Core Players are often 50-60% of a workforce. Keeping them engaged is a stability imperative, not a development priority. Recognition and fair compensation matter more than aggressive growth plans for this group.
Under Performers (Low Performance / Low Potential)
- Document performance concerns with specific, behavioral evidence
- Initiate a formal Performance Improvement Plan with clear timelines
- Set explicit checkpoints at 30, 60, and 90 days
- If no improvement is demonstrated, proceed to exit with appropriate documentation
- Do not let this drag for 12+ months. It harms the employee, the team, and your credibility as a manager.
The Under Performer box is uncomfortable, but ignoring it has costs. Research from CEB found that low performers drag the productivity of the employees around them down by 10-15%. Address it early.
Experts (Low Potential / High Performance)
- Build individual contributor career tracks so they do not feel forced into management
- Compensate them competitively. Losing an Expert is expensive to replace.
- Leverage them as mentors and subject matter experts
- Do not assume they want promotion; ask them directly about career goals
- Assign them to high-value, specialized work that matches their depth
Experts are often undervalued because they will never become VPs. That framing is wrong. A 10-year specialist who runs a critical function flawlessly is worth two or three average managers. Treat them accordingly.
Common Mistakes
Six ways talent reviews fail
The 9-box has critics, and some of the criticism is fair. When organizations run talent reviews poorly, they produce biased outputs and no action. When they run them well, they are one of the highest-leverage HR processes available. Here is what separates the two.
All stars, no action
When every manager rates their team as high potential, the grid loses its value. Calibration exists to fix this. If your grid is 40% stars, your criteria are too loose.
Conflating performance with potential
A 15-year specialist who executes their current role flawlessly is not necessarily a future VP. They may be an Expert. That is a valuable classification, not a failure.
Rating without context
New hires, employees returning from leave, and people in the middle of role transitions cannot be fairly rated on the same scale as tenured employees. Flag these separately.
No follow-through
The grid is a planning tool, not an end in itself. If action plans are not assigned after the calibration session, the whole exercise was a waste of an afternoon.
Sharing the grid with employees
Most organizations do not share box placements directly. The insight is useful for HR and managers. Telling someone they are in box 7 without context is almost always counterproductive.
Running it once and forgetting
People move. A Rough Diamond in Q1 can become a Star by year-end with the right development. Talent reviews should run at least annually, ideally twice a year.
Equity and Fairness
Managing bias in the 9-box process
The 9-box is a human rating system, which means it carries human biases. McKinsey research on performance management finds that women and underrepresented employees are consistently rated lower on potential relative to their performance ratings compared to their white male peers, even when controlling for output and experience. This is a structural problem, not an individual one.
Three practices reduce this bias meaningfully. First, require behavioral evidence for all high-potential ratings, not just verbal endorsement. Evidence is harder to distort than intuition. Second, track box distributions by demographic group and flag statistical outliers for additional scrutiny. If women are systematically rated lower on potential despite equivalent performance, that is a signal worth investigating. Third, train managers on specific bias patterns before the calibration session: recency bias, halo/horn effects, and affinity bias are the most common distortions in talent reviews.
For a broader look at how bias enters the hiring and evaluation process, see our guide on reducing unconscious bias in hiring. The same patterns that affect candidate selection also show up in talent reviews, and the mitigations are similar.
My view is that the 9-box does not create bias. It makes existing bias visible in a way that can be measured and addressed. That is actually an argument for the process, not against it.
Succession Planning
Connecting the 9-box to succession planning
A talent review without a succession plan is a diagnosis without a treatment. The 9-box tells you who your high potentials are. Succession planning tells you which roles they are being developed to fill and on what timeline.
The standard output of this connection is a succession slate for each critical role: typically a list of candidates at three readiness levels. Ready now (could step in within 90 days), ready in 1-2 years, and ready in 3+ years. Most organizations target at least two candidates per level for critical leadership roles.
Can step into the role within 90 days with minimal ramp. Typically already performing significant portions of the scope.
Needs one or two targeted experiences or skill developments before they are prepared for the full scope of the role.
High potential earlier in their career with a clear development trajectory but significant experience gaps relative to the target role.
A practical benchmark from LinkedIn Talent Solutions: organizations with two or more ready-now successors for their top 20 critical roles spend 40% less on external executive hires and fill those roles 60% faster when vacancies occur. The bench depth built through talent reviews has a direct cost impact.
The succession plan should be reviewed at the same cadence as the talent review. People leave succession slates through promotion, departure, or changed circumstances. A plan that is 18 months out of date is not a plan.
Tools and Technology
Running the 9-box with the right tools
Most small and mid-sized organizations run talent reviews in spreadsheets or presentation decks. That works for teams under 100 employees. Above that threshold, a dedicated HRIS or talent management module becomes worth the investment. Not because the grid requires it, but because the action tracking and succession planning that follow the review are hard to manage in a static document.
The core requirements for a talent review tool are simple: the ability to capture manager ratings, run a calibration session collaboratively, visualize the grid output, and assign action plans with owners and deadlines. Most modern HCM platforms (Workday, SAP SuccessFactors, Oracle HCM) include this functionality. Dedicated succession planning tools like Avature or SmartRecruiters add more granularity if your succession pipeline is complex.
Whatever tool you use, the process discipline matters more than the software. A well-run calibration session on a shared Google Sheet will produce better outcomes than a sloppy session in a purpose-built platform. For startups and scale-ups where every employee in the talent review is also a critical hire risk, connecting your talent review insights to your recruitment pipeline and quality of hire metrics helps you understand where your internal development is working and where you still need external talent.
Frequently Asked Questions
What is the 9-box grid in talent management?
The 9-box grid is a talent assessment tool that plots employees on a 3x3 matrix based on two dimensions: current performance and future potential. Each of the nine boxes represents a different talent profile, from underperformers to high-potential stars. HR teams use it to prioritize development investments, build succession plans, and identify flight risks.
How do you define potential in the 9-box grid?
Potential is typically defined as the likelihood of an employee succeeding in a role of significantly greater scope or complexity than their current one. Most organizations assess it through four lenses: learning agility, leadership behaviors, adaptability, and motivation for broader responsibility. The key is agreeing on a definition before ratings begin, not after.
Should employees be told their 9-box placement?
Most organizations do not share specific box placements. The grid is a planning and calibration tool for HR and management. However, employees should receive clear developmental feedback and career conversations as an output of the talent review process. The grid informs those conversations without being shared directly.
How often should you run a 9-box talent review?
Annual reviews are the minimum. Twice a year is better, particularly for fast-growing companies where roles and people change quickly. Quarterly is too frequent for meaningful movement to occur. The timing should align with your performance review cycle so data is fresh.
What percentage of employees should be rated as high potential?
Research from CEB (now Gartner) puts the realistic figure at 3-5% of any workforce. In practice, calibrated talent reviews typically identify 10-15% as high potential when criteria are rigorous. If your grid shows 30%+ in high-potential boxes, your criteria likely need tightening or your calibration process is not working.
How is the 9-box grid different from a performance review?
A performance review evaluates how well someone did their job over a defined period. The 9-box grid uses that performance data as one input, but adds a second dimension: potential for growth. A high performer with limited growth potential (an Expert) and a high performer with significant growth potential (a Star) are very different talent investments. The grid makes that distinction explicit.
Resources & Further Reading
Related Guides
- Workforce Planning: How to Build a People Plan That Holds Up
How talent reviews feed into headcount and capacity planning
- Internal Mobility Programs: How to Build One That Actually Works
Turning 9-box insights into internal career movement
- Quality of Hire: How to Measure It and Why It Matters
Connecting talent review ratings to your hiring effectiveness
- How to Reduce Unconscious Bias in Hiring
The same bias patterns appear in talent reviews and sourcing
External Sources
- Harvard Business Review: How to Keep Your Top Talent
Research on learning agility as the strongest potential predictor
- McKinsey: Talent Management Insights
Data on the financial impact of talent pipeline strength
- SHRM: Talent Acquisition & Development
HR research on succession bench depth and talent reviews
- LinkedIn Talent Blog: Talent Strategy
Data on succession planning ROI and high-potential retention
Build the talent pipeline your hiring plan depends on
Prepzo helps you track internal talent, identify gaps, and connect your succession planning to active hiring. Start free.
Try Prepzo free