This post explains why even a complete competency model can't tell you who's ready for the next role. It then shows you how to add what's missing, using the example of one employee measured against two roles.
A competency model is a framework that lists the skills, knowledge and behaviors each role needs and describes what each one looks like at different levels of ability, often on a four-level scale such as Basic, Skilled, Advanced and Expert. Most companies build it in layers, with core competencies that apply to everyone, functional competencies that apply to a job family such as finance or engineering, and leadership competencies that apply to people who manage others.
A good model helps in many places. It gives hiring managers the same criteria to interview against, shows employees what good work looks like in their role, and gives your learning team a shared language for designing training.
Everything in the model, though, describes roles rather than people. Each entry answers the question "What does this role need?" and none of them answers "What can this person do today?" Because the model describes the role and not the people in it, answering a readiness question needs a second set of data alongside it:
|
Your competency model |
Your people data |
|
|---|---|---|
|
Describes |
What a role needs |
What a person can do today |
|
Answers |
"What does good look like in this role?" |
"Where is this person now?" |
|
Changes when |
The role is redesigned |
The person learns, practices or changes jobs |
|
Usually owned by |
HR |
The employee and their manager |
To know who's ready for a role, you need both columns, and most companies have only built the first one.
You may have read that companies should replace competency models with skills. You don't have to choose between them, because most competencies can be broken down into the skills behind them and people can be measured on those skills. That way you keep the role definitions your leaders already agreed, and you add the people data the model is missing.
When a leader asks who's ready for a role, the gap between the role side and the people side shows up in five ways.
In most companies, the competency model sits in one place and information about people sits somewhere else. Performance reviews give each person one overall rating, learning records list the courses they finished, and their project history lives in their manager's head. None of that is recorded against the competencies in your model at the levels the model describes, so when a leader asks whether someone meets the "Advanced" level of stakeholder management, all you can check is their manager's impression.
When companies do collect competency ratings, they usually come from one source, either the employee rating themselves or their manager rating them, and neither is reliable enough on its own to decide who moves into a role. A review of 22 meta-analyses found that people's ratings of their own abilities match their actual performance only moderately (Zell and Krizan, 2014). The same research found that people rate themselves more accurately on specific, familiar tasks than on broad ones, which is a good reason to describe each competency in terms people can observe.
A manager's rating has a different blind spot. Managers see how someone handles their current job, but they rarely see how that person would cope with the different demands of the job they want next.
A competency model makes it easy to compare someone with the next role on their own career ladder, such as a senior analyst against the team lead role above them. It does much less to show that the same person is close to a role in another function, because nobody thinks to compare them with that role. Few companies even encourage people to look sideways, and in LinkedIn's 2024 Workplace Learning Report, only 19% of employees said their company encourages them to explore internal role changes (LinkedIn). When nobody looks across functions, people who are ready to move into another part of your business stay invisible to the leaders who could use them.
Most companies update their competency models on a fixed schedule rather than when the work changes. Bersin research cited by Korn Ferry found that even among high-performing companies, 62% to 70% updated their competencies only every two to three years (Korn Ferry Institute). That cycle is slow compared with how fast work is changing, since the World Economic Forum estimates that 39% of workers' current skills will change or go out of date between 2025 and 2030 (WEF Future of Jobs Report 2025).
People's ratings age as well. A rating from last year's review describes who someone was before their last six months of projects, so when both the role description and the rating are a year or two old, comparing them tells you about a situation that no longer exists.
Even with good data on both sides, "ready now" means different things to different leaders. One manager calls someone ready when they meet most of the competencies, while another wants every competency met plus experience in a similar role, so without one shared rule, two people with the same profile can land in different groups depending on who assessed them.
When you can't say who's ready for a role, you usually end up hiring from outside, and outside hires tend to cost more and perform worse in their first years than people promoted from within. In a study of a US investment bank, Matthew Bidwell of Wharton found that outside hires were paid about 18% to 20% more than employees promoted into similar jobs, received lower performance reviews in their first two years, and were more likely to leave (Knowledge at Wharton).
The shortage is easiest to see in leadership and other critical roles. In DDI's 2025 HR Insights Report, companies could fill only about half (49%) of their critical leadership roles from inside right away (DDI), and in a 2026 i4cp study, only 40% of business and HR leaders said their company had enough people ready to step into critical roles (HR Dive).
Your employees pay a price too, because people who could have moved never get the chance, and most of them can't see where they could go next. In October 2026, Gartner reported that only 35% of employees see a clear career path in their company as AI changes their work (Gartner). That matters to you as well, because LinkedIn found that employees who made an internal move had a 75% chance of still being at their company two years later, compared with 56% of those who didn't move (LinkedIn).
You don't need to rebuild your competency model to answer the readiness question. What you need to do is record people's current levels next to the role descriptions you already have, agree on one rule for comparing the two, and make the descriptions of your most important roles specific enough that people can be rated against them. The seven steps below show you how, starting with the roles where the answer matters most.
If you try to measure readiness for every role at once, the project is likely to stall before anyone gets an answer, so start with the 10 to 20 roles where a vacancy would hurt the business most, such as roles that are hard to hire for, roles with high turnover, or leadership roles with no clear successor. Once the process works for those roles, you can extend it to the roles around them.
A description like "shows strategic thinking" is hard to rate because two managers will picture different things when they read it, so for each competency in your priority roles, write down what the work looks like at each level. "Advanced" budget management, for example, might mean that the person builds and owns a yearly budget of more than $1 million and explains any gap from plan to the finance team without help. Descriptions like this also make people's ratings of themselves more accurate, because, as the research in reason 2 showed, people judge themselves better on specific tasks than on broad ones.
Before anyone is rated, decide which competencies are critical for each role and how big a gap you will accept at each readiness level. You also need to decide whether time spent in a similar role counts, and in the rule below it doesn't, because the evidence on each competency already shows what the person can do. A simple rule could look like this:
The exact thresholds matter less than applying the same rule to everyone, because a shared rule is what lets you explain to a candidate and their manager why one person is ready now and another isn't yet.
For each person in the running for a priority role, record their current level on each of the role's competencies, along with who gave the rating and when. The ratings are most reliable when they come from more than one of these sources:
When two sources are more than one level apart, ask the people who gave the ratings to talk it through and agree on a level, rather than averaging the two, because an average would hide what one of them saw and the other didn't.
Most readiness reviews compare a person only with the role above them, so compare each person's profile with similar roles in other functions as well. That is usually where you find people who are ready but overlooked, because nobody has made that comparison before.
A readiness label only helps if it changes what happens next, so for each gap, agree one action that gives the person a chance to show the missing competency in real work, such as a stretch assignment, an internal gig (a short project outside the person's own role), a mentor, or a course followed by a project where they apply what they learned. Then set a date to rate them again, so their readiness label moves when they do.
Update a person's ratings whenever they finish something that produces new evidence, such as a project, a gig or a certification, instead of waiting for the yearly review. Review a role's description in the same way when its work changes, for example when the team adopts a new tool or goes through a restructure, rather than waiting for the next scheduled update of the whole model.
Imagine your company is a distribution business with warehouses in several regions, and the regional operations manager who runs six of those warehouses plans to retire at the end of next year. When your head of operations asks who could replace him, the retiring manager puts forward Priya, a senior operations analyst who has reported to him for four years and whom he describes as high potential.
Your competency model already defines the regional operations manager role with six competencies, each described on the four levels from Basic to Expert, and it marks three of them as critical. Following steps 2 to 4, HR describes each level in terms of real work, collects ratings from Priya, her manager and the site managers she works with, and records who gave each rating and when. Here is Priya's profile against the role:
|
Competency |
Level needed |
Priya's level |
Evidence and who rated it |
Gap |
|---|---|---|---|---|
|
Operational planning (critical) |
Advanced |
Advanced |
Led the region's capacity plan for peak season. Rated by her manager and two site managers, May 2026. |
None |
|
Budget management (critical) |
Advanced |
Skilled |
Tracks spending on a $250,000 project but has never built a yearly budget. Rated by Priya and her manager, May 2026. |
One level |
|
People leadership (critical) |
Advanced |
Basic |
Rated herself Advanced for informally mentoring two new analysts. Her manager rated her Basic because she has never managed a team. Because the ratings were two levels apart, they talked it through and agreed on Basic, June 2026. |
Two levels |
|
Stakeholder management |
Advanced |
Advanced |
Feedback from three site managers, April 2026 |
None |
|
Data analysis and reporting |
Skilled |
Expert |
Built the region's performance dashboard, now used by every site. Rated by Priya, her manager and two site managers, May 2026. |
Above the level needed |
|
Process improvement |
Skilled |
Advanced |
Lean certification and a finished waste-reduction project. Rated by Priya and her manager, March 2026. |
Above the level needed |
Under the rule from step 3, Priya is ready in 1 to 2 years for the regional manager role, because she is two levels below the requirement on people leadership and one level below on budget management, and both are critical competencies.
That answer is more useful than the "high potential" label, because it tells Priya and her manager what she needs to work on before the role opens. To build people leadership, she could lead a six-month improvement project across several warehouses with three analysts reporting to her, and to build budget management, she could share ownership of next year's regional budget with her manager before he retires. A mentor who runs another region could help with both, and you would rate her again after six months to see whether she has moved up a group.
Following step 5, you then compare the same profile with a role in another team. Your company also has an operations excellence lead, who works across all regions to find and remove waste in how the warehouses run and has no direct reports. That role needs Advanced in three critical competencies, which are process improvement, data analysis and reporting, and stakeholder management, and Skilled in its other two, operational planning and budget management. Priya meets or beats the level on all five, so under the same rule she is ready now, even though nobody had thought of her for the role because it sits outside her reporting line.
Your competency model described both roles before you started, so what changed the answer was adding evidence of Priya's current levels and using the same rule for every comparison.
In 2020, Johnson & Johnson picked 41 "future-ready" skills, grouped into 11 capabilities, that its 4,000 technologists would need. It then used a large language model to estimate each person's level on every skill on a 0 to 5 scale, using data from its HR system, recruiting database, learning platform and a project management tool. Employees also rated themselves on the same scale, and a score was treated as usable when the two ratings were within one point of each other, which follows step 4 because the company compared two sources of evidence before it trusted a rating (MIT Sloan).
Technologists could then compare their current skills with the skills needed for a future role they were interested in, while executives could see heat maps of skills by region and business line. Use of the company's development platform rose 20% after the first round of skill estimates, and by March 2024, 90% of technologists had used it.
Fuel50 keeps your role requirements and your people's current levels in one place, so you can answer a readiness question without building a spreadsheet for every role.
On the role side, Skills Architecture builds each role's profile from a skills ontology, which is a library of more than 5,000 skills that also records how the skills relate to each other. Fuel50's people scientists maintain that library, and every skill in it comes with four proficiency levels and suggested ways to develop it.
On the people side, the Skills Inventory brings skills data from across your company into one current view of what each person can do. Succession then groups the candidates for a role into readiness levels such as "ready now" and "ready soon" based on their skills, rather than relying only on the 9-box grid, which places people by performance and potential.
In Priya's case, her ratings would sit in her profile, and Succession would place her in one of its later readiness groups for the regional manager role rather than "ready now". Mobility matches employees to open roles based on their skills, including roles in other functions, which is how the operations excellence role would come up for her. Development and Gigs would then turn her two gaps into a mentor, a learning plan and a short project where she leads a team.
At Lennox, employees who set development goals and worked with mentors through Fuel50 were 21% more likely to make an internal move (Lennox story), and setting a goal and working with a mentor are the kind of actions step 6 asks you to agree for every gap.
See how Fuel50 shows who's ready for your critical roles.
A competency model is a framework that lists the skills, knowledge and behaviors each role needs and describes what each one looks like at different levels. Companies use it to hire, review performance, plan people's development and decide who could replace someone in a key role.
A competency model describes what a role needs, while a skills profile describes what one person can do today. To know whether someone is ready for a role, you compare their skills profile with the role's competency requirements.
Record the person's current level on each competency the role needs, using ratings from more than one source, and then apply a rule everyone has agreed on. For example, "ready now" could mean the person meets the required level on every critical competency and is no more than one level below on the rest.
Review a role's description whenever its work changes, for example after a restructure or when the team adopts new technology, rather than only on a fixed schedule. Update people's ratings whenever they produce new evidence, such as finishing a project or earning a certification.
The most common types are core models, which apply to everyone in the company, functional models, which apply to one job family such as finance, and leadership models, which apply to people who manage others. Many companies also build a separate model for each of their most critical roles.