Organisations don't have culture problems. They have behaviour problems that culture language has made harder to solve.
That's not a provocation for its own sake. It's a diagnosis. And the distinction matters practically for anyone trying to change how an organisation actually works.
The concept of organisational culture entered mainstream management in the early 1980s with a genuine insight: that organisations with similar strategies and structures produced markedly different outcomes, and that the invisible force producing those differences — the pattern of shared assumptions, norms, and behaviours — deserved a name and serious attention.
That insight was valuable. But the naming carried a cost.
What happened when culture became a noun
Once culture became a thing rather than a process, it became something to be diagnosed and fixed. Consultants arrived with surveys. Reports were produced. Workshops were held. Values were articulated, printed on walls, and distributed in onboarding materials.
And in most cases, the behaviour that prompted the original concern continued largely unchanged.
This is not a failure of effort or intention. It is a failure of level. The interventions were operating on the noun — culture as a described entity — while the actual problem was a verb: what people repeatedly do, tolerate, reward, and avoid. You cannot fix a verb by diagnosing a noun.
The culture poster is not culture. It is a description of the culture someone wants. The culture itself is what people do on Tuesday afternoon when no one is watching.
The irony of culture language
Edgar Schein, whose three-level model of culture has dominated the field for four decades, identified the problem precisely: genuine culture change requires intervention at the level of basic underlying assumptions — the unconscious, taken-for-granted patterns that shape behaviour without deliberate intention. Most culture programmes, he observed, never reach below the surface level of artefacts. They change what organisations say without changing what organisations do.
Culture language accelerates this problem. When we describe a challenge as a culture issue, we locate it at the level of shared beliefs and values — the internal states that culture language most naturally describes. But behavioural science has established with considerable consistency that internal states are poor predictors of behaviour. Behaviour is primarily a function of environment, context, and social norm.
Telling people what the culture should be, and asking them to report how aligned they feel with it, changes neither the environment nor the norm. It changes the language people use to describe what they already do.
The question that actually works
The question that unlocks genuine culture change is not "what is our culture?" or even "what should our culture be?" It is: which behaviours are currently present, which would produce the culture we want, and what conditions would make those behaviours more likely to occur naturally?
This reorientation — from diagnosis to dialogue, from measurement to facilitation, from values prescription to behavioural agreement — requires treating culture not as a noun to be managed but as a process to be shaped. Weick described culture as an ongoing collective activity of meaning-making. You don't change it by describing it differently. You change it by creating conditions in which different behaviour becomes the norm.
That is a different kind of intervention. It works at the level of the environment — the systems, habits, structures, and conversations that make particular behaviours easier or harder. It produces specific, observable, accountable commitments rather than aspirational language. And it gives teams genuine ownership of the outcome because they built it themselves.
The problem was never the culture. It was always the behaviour. The difference matters — because you can design toward behaviour in ways you cannot design toward culture.
Start there.