What is an organisation's capability?

An organisation's capability is its repeatable ability to produce an outcome through people, knowledge, routines, systems, relationships and decision rights together. Individual expertise contributes to it and does not constitute it. The distinction has been load-bearing in this research for two years and has never been stated here properly.

Two consequences follow immediately, and both are counterintuitive enough to be worth the page. An organisation can lose capability while every individual stays. And it can keep capability while individuals leave.

Skill and capability

Start with the smaller distinction, because the vocabulary collapses constantly.

A skill is the ability to perform an activity. A capability is the capacity to achieve an outcome, which requires combining skills with knowledge, judgement and the ability to adapt them to a situation that does not match the one they were learned in.

Someone can hold every relevant skill and still be unable to handle the whole problem. That gap is what the word capability names. A list of trained skills describes what a person or an organisation can actually do only very roughly, for the same reason.

Capability lives in routines

The cleanest account is Nelson and Winter's. Firms are carriers of routines: patterned, repeatable sequences of coordinated action that persist beyond the individuals performing them. Routines are what an organisation knows how to do. An Evolutionary Theory of Economic Change (1982).

Prahalad and Hamel reached a compatible position from strategy: core competence is the collective learning in the organisation, particularly the capacity to coordinate diverse skills. The Core Competence of the Corporation (1990). Their warning was about outsourcing, and it was that cost advantage can be bought while the competence that made the firm able to compete is hollowed out underneath.

Teece, Pisano and Shuen added the moving part. Under rapid technological change what matters is not a fixed stock of competences but the ability to integrate, build and reconfigure them. Dynamic Capabilities and Strategic Management (1997). The construct has been criticised as hard to operationalise, and the criticism is fair, but the point survives it: capability is a verb.

The case that matters: losing it without losing anyone

If capability lives in routines, then a routine that stops running decays regardless of whether the people who used to run it are still on the payroll.

Consider a review step. Work used to pass from a junior to a senior, who read it, found the problems, explained them and sent it back. That sequence produced the output, and it also produced the junior's judgement, the senior's continued familiarity with the detail, and a shared standard nobody had written down. Route the work through a system that produces the output directly and the sequence stops running. Every person is still employed. The routine is gone, along with the three things it was quietly also doing.

This is why headcount is such a poor instrument here. Nothing in a staff list registers the difference.

The reverse case

The same argument runs the other way and deserves equal weight, because it is the reason organisations survive at all.

A genuinely organisational capability tolerates individual departure. When a routine is coordinated, documented where documentable, practised by several people and renewed by newcomers entering it, one person leaving is a loss rather than a rupture. When it is not, the organisation has been running on a person rather than a capability and did not know which.

The useful diagnostic follows from that: a capability that cannot survive the loss of any single individual was never organisational, whatever the process map says.

What AI changes

Nothing in the definition above involves a machine. What follows from the tools is a measurement problem rather than a new mechanism.

Almost all AI measurement is of individual output, and almost all of what is at risk is not individual. A tool can raise what every person produces while the coordination, review and handover through which the organisation delivered the outcome falls into disuse. Those two things can move in opposite directions simultaneously, and only one of them appears on the dashboard.

Argyris and Schon supply the reason it goes unnoticed. Organisations correct errors within their existing assumptions readily, and question the assumptions themselves rarely, because organisational defences prevent it. Organizational Learning (1978). A programme measuring throughput will improve throughput and will not raise the question of what else the removed step was doing, because that question sits outside the loop being measured.

What this page does not establish

The literature here is theoretical and case-based rather than measured. Nelson and Winter founded a school; they did not run an experiment. Dynamic capabilities have been criticised precisely for resisting operationalisation, which means the framework can describe a loss after the fact more easily than it can detect one in advance.

No study establishes that AI adoption degrades organisational routines. The mechanism is available and the measurement has not been done, and this page should not be read as evidence that it has.

The six dimensions this research proposes elsewhere for auditing capability are an operationalisation offered on top of that literature, not a finding from it.

Key sources

• Nelson, R. R. and Winter, S. G. (1982). An Evolutionary Theory of Economic Change. Belknap Press. In the essential works.
• Prahalad, C. K. and Hamel, G. (1990). The Core Competence of the Corporation. Harvard Business Review, 68(3), 79-91. In the essential works.
• Teece, D. J., Pisano, G. and Shuen, A. (1997). Dynamic Capabilities and Strategic Management. Strategic Management Journal, 18(7), 509-533. In the essential works.
• Argyris, C. and Schon, D. A. (1978). Organizational Learning: A Theory of Action Perspective. Addison-Wesley. In the essential works.

Related SuperSkills research

On holding on to the people side of it, keeping expertise in an organisation. On what the loss accumulates as, capability debt. On measuring it, assessing capability rather than output. On the individual faculties underneath, judgement and tacit knowledge. On the entry route into a practice, the missing rungs.

About this research

Rahim Hirji is the author of SuperSkills (Kogan Page, 2026), keynote speaker on AI and human capability, and founder of The SuperSkills Intelligence Company. Routines, core competence and dynamic capabilities are established constructs from evolutionary economics and strategy, and none of them is a SuperSkills coinage.

How this research works  ·  Reviewed quarterly  ·  Found an error? Tell me and it is corrected on the page.