Imagine a mid-sized Chennai software company, three hundred people, and its annual talent review. The head of engineering, Suresh, has a printed grid on the table. Beside him sit the HR business partner, Divya, and the head of product, Karthik. Twelve names from the senior-engineer and team-lead population need to be placed on a nine-box grid. Within ten minutes the meeting stalls on one name: Meera, a senior engineer.
The people, company and scene here are invented, but the argument is a very ordinary one. Let us use it to understand the tool.
What the grid actually is
The nine-box grid is a three-by-three matrix. One axis is current performance (low, moderate, high), judged on results delivered against what was expected in the present role. The other axis is potential (low, moderate, high), meaning the capacity to grow into broader or more complex roles. Each person lands in one of nine cells. Top right is the high performer with high potential, often called a star or future leader. Bottom left is low on both. The cell in the middle, moderate on both, is where the bulk of any healthy workforce sits, and that is fine.
It is a conversation tool. It does not calculate anything, and different companies label the cells differently, so do not treat any label as standard. What is standard is the idea of separating two questions that people tend to blur: how well is this person doing today, and what could they do tomorrow?
Back to Meera
Suresh puts Meera in the top-right box. Her delivery is excellent, her code reviews are sharp, and she mentors juniors without being asked. Karthik disagrees. He has worked with her on two product launches and says she avoids ambiguous problems and prefers well-defined tasks. High performance, certainly, but is that high potential?
Here the grid does its real job. Divya asks both men a plain question: what evidence do you have of how she behaves when the problem is unclear? Suresh has only seen her on assigned work. Karthik has seen her step back twice, but one of those times the requirements changed late and nearly everyone struggled. After ten minutes of specifics, they agree that Meera is a high performer with moderate-to-high potential, and that the gap in evidence can be closed by giving her a stretch assignment, not by arguing further.
That outcome is better than either original opinion. The box matters less than the shared evidence that produced it. This is the whole point of calibration: managers who each see only a slice of a person compare notes so that ratings mean the same thing across teams.
How potential gets misjudged
Potential is the harder axis, and where the grid is most abused. A few patterns to watch for in any room like this one:
- Performance masquerading as potential. A strong current performer gets a high-potential tag purely because results were good. Doing the present job well does not prove capacity for a bigger one.
- Visibility bias. People who speak confidently in meetings or sit near leadership get rated higher than quiet colleagues with similar ability.
- Recency. The last quarter dominates the discussion.
- Similarity. Managers favour people who resemble them in background or style.
Some organisations try to define potential through observable indicators, for example learning agility, ability to handle ambiguity, and drive to take on larger scope. Whichever framework you use, write down the behaviours and ask for examples. “She just feels senior” is not evidence.
What happens after the boxes are filled
A grid with no follow-up is wall decoration. Each region suggests a different kind of action. People in the high-potential cells may need stretch roles, mentoring by a senior leader, or exposure to other functions. High performers with limited potential for promotion are often the backbone of the business; they need recognition, fair pay, and interesting work in their area of expertise, not a promotion they may not enjoy. Someone with high potential but low current performance may be in the wrong role, or lacking support, so find out before judging. And low performance with low potential is a conversation about improvement with clear time frames, or a role change.
Note the word “suggests”. The grid does not decide a promotion or a pay rise. Those should rest on role requirements, business need and a proper assessment.
Handle with care
Two cautions. The grid is usually a confidential management document. Telling people which box they are in, especially a lower box, tends to cause hurt and gaming. Many companies instead share development feedback without the box label. Whether you disclose or not, be consistent and decide beforehand.
Second, the picture is a snapshot. Someone placed in a middle box this year may be somewhere else in eighteen months after a new manager, a new role or a family situation resolved. Review it at least annually and let people move. If the same names sit in the same cells for five years, the grid has become a label rather than a tool.
As for Meera, she gets a six-month assignment leading a vague integration problem with Karthik’s team, and Divya notes a check-in date. The grid went back in the folder; the stretch assignment is what changes anything.
Figures and names in the examples are illustrative.
Leave a Reply