Collaboration Reimagined: Challenging the Status Quo
Collaboration Reimagined: Why You Can’t Train What the Structure Punishes
In any complex system, behaviour and performance are expressions of structural design, and an organisation is exactly such a system. What people do is produced by its architecture: the decision rights, the incentives, the information flows, and the consequence pathways.
And beneath that architecture sits the culture it expresses, the organisation’s real answer to what actually pays off here.
Behaviour follows from that stack.
It is not, at root, a matter of individual character.
Hold that principle against one of the most sought-after and most elusive behaviours in organisational life: collaboration.
Collaboration is usually treated as a mindset – a generous disposition that appears when people share expertise across boundaries out of empathy and shared purpose. Organisations declare it a core value, put it on the wall, and then find it stubbornly refuses to materialise. So they reach for interventions: team-building, shared workspaces, collaboration software, and campaigns urging people to break down silos. These help at the margin and change nothing durable, and the reason is the one the mindset framing can’t see.
Collaboration isn’t failing because people lack the disposition. It’s failing because the architecture is paying them not to.
Collaboration can’t be mandated, because it can’t be exhorted
Try to enforce collaboration directly, and you get one of two things: superficial compliance, or a quiet breakdown as people perform sharing while protecting what matters, and not because people are resistant.
People are quite willing and ready to collaborate in the right conditions and often more freely in informal settings than in formal teams where it’s demanded.
The variable isn’t the person’s willingness. It’s whether the structure around them makes collaborating rational or costly.
Consider the designer at an ad agency who refused to share his expertise, certain that giving it away would erode his competitive edge.
It’s tempting to file this under ignorance, that he doesn’t understand the benefits of collaboration.
But look at what the story actually shows: he understood collaboration perfectly. He also understood his incentives perfectly and read them correctly. In a structure where individual distinctiveness is what gets rewarded, protected, and promoted, sharing your edge is a rational thing to avoid. He wasn’t lacking a competency. He was responding, accurately, to what his architecture paid off. Train him in the virtues of teamwork all you like; send him to every workshop; the moment he returns to a structure that rewards individual edge, he will protect his edge, because he’s not confused, he’s rational.
This is why the “collaboration as competency” reframe, despite improving on “collaboration as mindset,” still fails.
It relocates the problem back into the individual, something they lack and can be trained to acquire, when the behaviour was never primarily about what the individual possessed. Collaboration is neither a mindset to inspire nor a competency to install.
It’s a behaviour that an architecture either rewards or punishes, and no amount of training overrides an incentive pointing the other way.
What actually suppresses it
If collaboration is structural, then the barriers are structural, and the biggest one is usually the performance management system.
A KPI-driven model, especially when tied to compensation, is close to purpose-built to destroy collaboration. When reward depends on hitting individual targets, and even worse, on relative standing against colleagues, sharing knowledge becomes a direct cost. Helping a colleague improves their number, possibly at the expense of yours. Admitting a problem lowers your standing. The structure has made the collaborative act personally expensive, and then leadership wonders why people won’t collaborate.
This is the deeper reason you cannot incentivise collaboration through a KPI. A target designed to “drive collaboration” reduces a relational behaviour to a metric, and the moment collaboration becomes a number people are measured on, they optimise the number rather than the thing it stood for, producing the appearance of collaboration while the substance drains away. You can’t fix a structural suppression of collaboration by adding another target on top of it. You have to remove the suppression.
Fear does the rest of the work, and it too is structural rather than dispositional.
Where the architecture punishes error and rewards individual standing, people rationally protect themselves – guarding knowledge, avoiding the exposure that sharing creates, declining the risk of visible contribution. This gets read as a trust problem or a mindset problem. It’s an accurate response to a structure where openness costs more than caution. Fear isn’t the saboteur. It’s the symptom of an architecture that made self-protection the rational strategy.
What actually produces it
So the fix isn’t cultivating collaborative individuals.
It’s building an architecture in which collaborating is the rational thing to do – where the structure rewards the behaviour, the values statement merely requests.
Concretely, that means performance systems that reward contribution to shared outcomes rather than individual standing against colleagues, so that helping someone stops being a cost and becomes part of the job. It means decoupling reward from relative ranking, so that a colleague’s success no longer threatens your own. It means consequence pathways where sharing a problem or a mistake is safe rather than career-lowering, so that knowledge circulates instead of being hoarded.
And beneath all of it, a cultural foundation whose real answer to “what pays off here” is aligned with the collaboration the organisation claims to want, because where the foundation still rewards individual edge, every structure built on it will quietly reward hoarding, no matter what the workshops say.
Do those, and collaboration stops needing to be mandated, inspired, or trained.
It appears on its own because you’ve stopped paying people to avoid it.
Training and mentorship have a place, but as programs that flourish inside a structure that already rewards sharing, not as substitutes for building one. A mentoring culture across departments is a wonderful thing; it will not survive a performance system that punishes the people who mentor.
The point
Collaboration isn’t a value you declare, a mindset you inspire, or a competency you train. It’s a behaviour your architecture produces or prevents.
The organisations that “can’t get their people to collaborate” almost always have structures that reward the opposite – individual targets, relative ranking, punished error, guarded knowledge – and no declaration of collaboration as a core value competes with an incentive that pays the other way.
Stop asking why your people won’t collaborate.
Look at what your structure is rewarding them to do instead.