Please enter your email in order to download this publication.
“A bad carpenter blames his tools”
The talent review is a well-accepted and increasingly common element of how organizations manage talent. They see the value in understanding an individual’s potential to advance and using that information to better plan for succession. Which tool should guide that discussion, however, is increasingly under debate.
The classic performance x potential gird (9-box or other variant) remains the most popular tool to use in these conversations, based on our client experience. Yet, there are some HR leaders and consultants who deride the 9-box and similar matrices as outdated or no longer applicable in talent review discussions.
Their objections include, “Haven’t we moved beyond the point where we put people in boxes?” or values-based reactions like, “It’s not helpful to label people.”
About The Tool
The purpose of a talent review is to differentiate how far and how fast someone can move in your organization so that you can accurately plan for succession and development. We are regularly asked what the optimal tool is to use in this discussion, if any, and our consistent answer is: it doesn’t matter as long as it facilitates an accurate conversation about an individual’s performance and potential.
The fact that leaders frequently conflate performance with potential in these discussions makes it valuable to have a way to discuss those factors separately. Hence, the performance and potential matrix.
About Performance and Potential Matrices
Given the on-going discussion about the best tool to use, and that the 9-box and its derivatives are the most popular, we share below our thoughts on those tools and other options to guide the potential discussion.
How many boxes is ideal?
We can confidently say that we have seen every permutation of grids from 3 to 5 to 6 to 28 boxes. Every company has a logical reason as to why they made the choice they did. Some try to match the boxes to their performance management scale. Others add potential categories to create finer gradations of movement.
If we focus on the purpose of the tool – to enable an accurate discussion – we don’t believe there is an optimal number of boxes. We also can’t find any research that suggests there’s a “right” number of ways to classify potential and potential. This means that it’s your organization’s preference and judgment that should dictate how many categories there are in your model.
Your choice depends on what you believe is the number of meaningfully different ways that people can demonstrate performance and potential in your organization. If you genuinely believe there are 28 different ways people can do that, then a 28-box grid meets your needs.
In considering the “right” number of boxes, we suggest that the definition of any one box should feel unique and separate from the others. For example, someone we describe as a high potential leader should sound different from a high-performing leader. That high-performing leader definition should sound different than an average performing leader who is well-placed. If there’s not a meaningful difference in how you define two boxes, they’re likely describing the same type of individual and can be combined.
It’s also easy to overvalue symmetry in the tool. We believe that it is not possible to be a consistently low performer and still have high potential to advance. In a typical 9-box model that suggests that the lower right-hand box (and possibly the center-bottom box) should not exist. The resulting seven box grid may look lopsided, but it’s more accurate than including boxes in which no one will ever be rated.
About the Potential Axis
We describe the best science about how to define potential in our recent article, “Potential after the Pandemic,” so we won’t repeat that here. We will reinforce, however, that potential should be measured by the ability to move upward over a set period.
We use that definition because potential is the primary input to succession planning. If a company doesn’t know what talent can move up at what pace, it is literally impossible for them to plan for succession.
The pace at which, and the reasons why, someone can move up in any given organization will differ. This means that your definition for movement over time may differ from other companies.
Potential labels should directly state how far and how fast someone can move up in the organization. How you divide those time periods is up to you, but we recommend starting with what you believe is the fastest possible upward movement that can happen in your organization. After that, the next category should describe what feels like relatively fast movement. If that sounds subjective, it is.
Our “generic” labels for the potential axis of a 9-box grid are Two Levels in Five Years, One Level in Two Years and Well Placed. “Well Placed” is used very intentionally instead of other possible choices. Labeling someone “low potential” is a non-starter in most companies not to mention slightly insulting. Well placed declares that, at this moment in time, you do not see the individual being promoted within 2 years.
About the Performance Axis
Two key factors help the performance axis to deliver value. First, it should reflect an individual’s sustained performance over time – ideally a three-year average. Using only the most recent year risks having wildly different annual ratings based on one unusually good or poor year.
Second, it should incorporate both what someone delivers and how they deliver it. The blending of the what and the how keeps the rating or classification simple but helps to ensure that weak behaviors lower the overall performance rating.
We have some clients who use one axis for performance and the other for behaviors. This approach makes behaviors more prominent than blending them with performance. It isn’t our preferred approach, however, since it requires an additional indicator be added to indicate potential.
While it might seem obvious that you should try to directly map your performance management scale to the performance axis, we typically find that it’s an unhelpful exercise. What matters most on the performance axis is ensuring that only true high performers are in the highest performance category.
Too many performance management ratings show an unreasonably large number of people in the highest performance category. This can feed the wrong information into the rest of the potential discussion.
Your performance axis labels can take many forms – percentiles (80th percentile and above, 50th percentile), labels (highest, average, below average) or other performance indicators (consistently exceeds goals, consistently meets goals). Your primary concern should be that the top category truly differentiates those leaders who consistently deliver at the 75th percentile and above compared to their other smart peers.
Keep the Junk Out of the Boxes
It’s an all-too-common mistake to load up a performance and potential matrix with labels, definitions, colors and other attempts to “help” managers with the tool. This adds complexity without value and removes a manager’s responsibility to compellingly describe why Suzie is a high potential.
If your matrix has labels for potential, it’s your managers’ responsibility to describe why Suzie should be placed in a box. It’s not up to HR to add a lengthy definition into each box so that a manager simply points at it and says, “That’s Suzie right there!”
Similarly, what value is added by coloring the boxes or adding a label to each one? Are you going to communicate to someone that they’re a “yellow” or a “green”? In a typical 9-box, you have three definitions for potential (and hopefully some separate guidelines for potential) and three definitions for performance. That feels like enough information to provide to a manager for her to make an intelligent decision about someone’s placement.
Assessing High Potential without a Matrix
It’s certainly possible to assess potential without using a matrix but the fundamental discussion remains the same. When we design “question based” potential identification processes for clients, they start with the essential question, “Does this person consistently perform at a level higher than their peers?”
That question is the same one you’d discuss when determining where on the performance axis to place someone. You’re simply having the same conversation without a grid in front of you. The benefit of the question approach is that only high performers are eligible to be high potentials, since anyone who isn’t rated a high performer is instantly eliminated the conversation.
After that question, we ask three to five additional questions that help to categorize that person’s ability to move up in the organization.
The criteria used in a question-based approach are no different than those used with a matrix. The conversation might be more structured since you’ll ask the same questions about each individual. The output will be the same since you’re still categorizing employees by different levels of potential.
The drawback of this approach is that it ignores the 75%+ of employees who are not high performers or high potentials. There is no opportunity to discuss them to see if you’ve missed a few “shy-po’s” or to assess the quality and depth of a manager’s overall team on one grid.
Overcome the Fear of Labeling
Some leaders believe that placing a label or rating on people’s performance and potential is somehow deleterious. It’s important to remember that a talent review is a prediction of upward potential at that moment and with the information currently available.
Someone who is rated as being in the right role, right now (Well Placed), might be considered ready for promotion if their performance changes or the company’s needs change. Conversely, someone who is rated as high potential today may have that rating change if they don’t deliver results in their current role or if their capabilities become less valued by the organization. There is no harm done by stating your current view of an individual’s potential.
We feel strongly about the topic of transparency but also recognize that every organization needs to determine their own talent philosophy around transparency. We’ve written extensively about this (see Be Transparent Now) so we won’t revisit that material here.
We will reinforce, however, that it’s going to become increasingly difficult to hide from your employees the organization’s view about how far and how fast they can move.
Link to Succession Planning
The purpose of a talent review is to determine the speed at which people can advance in your company, in order to plan for succession. This means that anyone who is in your high potential box should appear on a succession chart. If they do not, what are they high potential for?
This is also a great way to audit the names that a manager places on their grid. If any high potential talent doesn’t appear on that manager’s succession chart, or people who aren’t high potential appear on the succession chart, you should have some tough questions for that manager.
Link to Development Planning
We’ve asked hundreds of audiences around the globe about their effectiveness in translating talent reviews into development actions. There is a consistent and meaningful drop from the first process to the second one.
One way to ensure development planning happens, at least for your highest potential employees, is to ensure that there is a specific development action agreed-upon in the talent review for every high potential leader.
The discussion about which tool to use in a talent review discussion pales in comparison to the quality of the discussion. If a tool helps to guide the discussion and leads to a more accurate prediction, you should use one. If you find that your predictive accuracy is just as good not using a tool, that’s fine as well.
Your goal in a talent review is to plan for your organization’s succession needs and commit to the investments that will ensure successful growth. That goal will be realized through a well-structured discussion, clear definition of potential and honest assessment of an individual’s ability and willingness to do more.