The limits of competency frameworks (and where they still earn their place)

A framework is a map, not the territory

Most large organisations run on a competency framework, and for good reason. It lists the behaviours the business expects of a leader, sorts them into levels, and gives everyone a common vocabulary for what good looks like. Used well, it makes promotion fairer, feedback clearer, and development more deliberate. The trouble starts when a useful map gets mistaken for the territory itself. A framework describes the observable behaviours a leader should display. It does not, on its own, tell you whether those behaviours will hold when the situation turns and the easy options run out.

What competency frameworks genuinely do well

It is worth being fair about this before discussing the limits, because the limits only matter once you respect what the tool is for.

A good competency framework gives an organisation three things that are hard to get any other way. The first is clarity. When the expectation for a senior leader is written down as a set of named behaviours, people stop guessing what the organisation values. A manager preparing for promotion knows what is being looked for. The person assessing them is working from the same list. Ambiguity is expensive, and a framework removes a great deal of it.

The second is fairness. Without an agreed standard, leadership decisions drift toward whoever is most visible, most confident, or most like the people already in charge. A framework anchors the conversation to evidence against a shared bar rather than to impression. It will not eliminate bias, but it makes bias easier to spot and harder to defend, which is genuine progress. We have written separately about where bias still enters leadership assessment and how to guard against it.

The third is a shared language. A framework lets a board, an HR function, and a line manager discuss the same person using the same terms. It connects recruitment, development, and succession into one vocabulary instead of three. That coherence is valuable, and dismissing competency frameworks wholesale, as some commentary does, throws away something organisations genuinely need.

So the case is not that frameworks are flawed. It is that they answer a specific question well, and are routinely asked a larger one they were never built to answer.

The reduction problem: behaviour is the surface, not the source

Here is the limit, stated plainly. A competency framework measures what a leader does that you can see. It infers capability from the presence of observable behaviour. That works well when conditions are stable, because stable conditions let almost anyone perform the listed behaviours. It works far less well when conditions harden, because the same visible behaviour can rest on very different foundations underneath.

At CapabilityFX we draw a firm line between leadership skills and leadership capability. Skills are specific and learnable, and a framework catalogues them well: running a meeting, giving feedback, setting direction. Capability is the broader ability to apply sound judgement across situations a leader has not faced before, especially when the stakes are high. Skills are a subset of capability. A framework that stops at the catalogue of observable behaviours stops short of the thing that decides whether those behaviours survive pressure.

This follows from how leadership actually changes. The decade of doctoral research underpinning our work points to a consistent finding: lasting change happens inside-out, at the level of who a leader is, not only what a leader can do. A framework reads the outer layer, the visible behaviour. It is largely silent on the inner capability that produces that behaviour and decides whether it holds. Two leaders can tick the same competency. One does so because the behaviour is genuinely theirs and will hold under load. The other has learnt to perform it in calm weather and will lose it the moment the weather changes. The framework, by design, cannot tell them apart.

Why the gap widens under pressure

Pressure is the discriminator. The behaviours a framework lists are easiest to demonstrate precisely when they matter least, in calm conditions that forgive a great deal. The questions that actually separate leaders arrive only under load: whether judgement holds when the information is incomplete, whether the difficult conversation still happens when deferring it would be easier, whether the person stays present rather than retreating into procedure or bravado. A framework can name "manages ambiguity well" as a competency. It cannot, from a behavioural rating in a stable quarter, tell you whether the capability behind it is real or rehearsed.

This is not a criticism of any particular framework or of the people who build them. It is a property of behavioural description itself. The method reads the surface accurately and infers the source. When source and surface line up, which is most of the time, the inference holds. When they come apart, which is exactly when leadership is tested, the framework reads the surface and misses the source. That is the gap a capability lens is built to close. For the fuller version of this argument, see why most leadership assessments measure the wrong thing.

Complement, do not replace

The right move is not to discard the framework. It is to add the layer the framework cannot reach. A competency framework tells you whether a leader currently displays the behaviours the organisation expects. A capability lens tells you whether those behaviours rest on something durable. You need both, and they answer different questions.

This is where the assessments CapabilityFX is licensed to use earn their place alongside a framework, not in place of it. Ennea International's Five Lens Development Platform reads leadership at the level of motivation, identity, and relational style, the patterns that sit beneath the visible behaviour and decide whether it holds. Tomorrows Compass's future-readiness assessment looks forward, at whether a leader is equipped for the demands coming rather than the ones already mastered, which a present-tense framework cannot ask. Both work with a framework, deepening a behavioural picture rather than competing with it. You can see how we combine them on the assessments page.

If you want a structured way to see where, in a real decision, behaviour either holds or gives way, our DUAL model sets out the sequence: Discover, Understand, Accept, Lead. It maps the move from describing a situation to leading through it, which is precisely the territory a competency list cannot chart. Our 4D method shows how that deeper capability is built, not merely catalogued.

What it looks like in practice

The difference between framework-present and capability-present is easiest to see in specific people doing specific work.

The high-rated successor. Consider a divisional head in a wholesale and retail group who scored well against every competency in the leadership framework. The behaviours were all there on paper, and the framework had done its job honestly: it described an accomplished manager. When a key supplier collapsed and the division had to be replanned in a fortnight, a different pattern surfaced. The listed behaviours, decisiveness, communication, holding others to account, thinned out as the pressure rose. The competencies had been real in calm conditions and rehearsed rather than rooted. The framework was not wrong. It had simply read the surface, which was all it was built to read. A capability lens, run alongside, had already flagged a pattern around control that explained exactly where the behaviour would give way.

The under-read operator. Now consider the opposite. A plant manager in the same group rated middling against the framework, short on the visible markers the business associated with leadership presence. On the competency picture alone, an obvious development case and nothing more. When a serious safety incident hit, that manager ran a calm, sequenced response, held the hard conversations, and named honestly where the system had failed. The capability was plainly there. The framework had under-read it because the manager did not perform the listed behaviours in the expected register, not because the underlying judgement was missing. The behaviour the team needed showed up in the room when it counted, in the small moments where judgement is either present or it is not. A framework tuned to calm-weather presentation missed it.

Across both, the lesson repeats. A competency framework is a fair and useful reading of the surface. It overrates the first leader and under-reads the second for the same reason: it measures observable behaviour and infers the capability beneath, and that inference breaks at exactly the moment leadership is tested. Add the layer that reads the source, and both leaders come into focus.

The reader's next step

You do not need to abandon your framework to ask sharper questions of it. Before you trust a competency rating with a development or promotion decision, put it to a short test.

• Is the behaviour rooted or rehearsed? Ask whether the competency was demonstrated under real pressure or only in stable conditions. A behaviour that has never been tested under load is a hypothesis, not a verdict.
• Does the framework explain or only describe? A framework tells you a competency is present or absent. It rarely tells you why, or whether it will hold. If you need the why, you need a layer the framework does not contain.
• Is it forward-looking? Most frameworks are written around the demands of the current role and the present organisation. Ask whether yours reads readiness for the demands coming next, or only mastery of the ones already faced.
• What happens at the edges? Picture your two most demanding scenarios of the last two years. Ask whether the framework would have told you, in advance, which leaders would hold through them. If you cannot say yes with confidence, you are reading fair-weather competence and calling it capability.

Keep the framework, add the lens

The organisations that get the most from a competency framework are not the ones with the most detailed one. They are the ones who treat the framework as the clear, fair, shared-language layer it is, and who add a capability lens for the question the framework cannot answer. If you are weighing how to pair the two for your own leadership population, our services set out how an engagement works, and the assessments overview shows what the deeper layer reads.

Measure the source, not only the surface

A competency framework is not the problem. Treating it as the whole picture is. Used for what it does well, clarity, fairness, and a shared language, it is one of the more useful instruments an organisation has. Asked to predict whether a leader will hold when the stakes are highest, it is reading the surface and inferring the source, and the cost of that gap is high: leaders promoted on rehearsed behaviour who falter under load, and capable leaders overlooked because they do not present in the expected way. Keep the framework. Add the layer that reads what sits beneath it. If you would like to see what that pairing looks like for your team, start a conversation.

The leaders described here are representative composites drawn from patterns we observe in practice, not identifiable individuals. The research claim refers to the doctoral work underpinning the CapabilityFX approach to leadership capability.