Written by Michiel R. De Boer
The silos between the Sales team and the Product team had been a known problem for three years. The handoffs were consistently poor. The Sales team committed to things the Product team could not deliver. The Product team-built things the Sales team could not explain to customers. The communication between them was formal and adversarial rather than collaborative and frequent. Everyone acknowledged the problem. Leadership had discussed it in at least four operational reviews. The problem persisted.
The reorganisation was designed to solve it. The two functions, along with Marketing, were brought into a single Commercial unit under a newly appointed Chief Commercial Officer. The logic was elegant: shared leadership, shared budget, shared accountability. No more “that’s your problem, not mine” across a functional boundary. One unit, one direction, one P&L.
Six months after the reorganisation, the silos were different. The Sales team and the Product team now had a shared CCO, but the teams sat in different buildings, were managed by different directors with different backgrounds and loyalties, and had different performance metrics that were still, in practice, evaluated separately. The old handoff problems had not disappeared. They had been renamed. The arguments that used to happen across a functional boundary now happened within the Commercial unit, with the same underlying dynamics and the additional complication that the people involved now had to pretend, in official forums, to be on the same team.
Three new silos had emerged: between the combined Commercial unit and Finance, between Commercial and Operations, and between the former Sales leaders and the former Product leaders, who were now formal peers under the same umbrella and informally competing for the CCO’s attention and confidence.
The reorganisation had cost the organisation significant disruption: twelve weeks of productivity lost across the affected teams, a 14 percent increase in voluntary turnover among people who had not wanted to be moved, and four months of the CCO’s time spent on structure and governance design that might have been spent on the customer and market challenges that had been the original motivation.
And the silos, the actual problem, remained. In a different shape, but with the same root cause.
What Silos Actually Are
The vocabulary of organisational silos describes a structural phenomenon: departments that do not communicate well, that withhold information, that optimise for their own metrics at the cost of the organisation’s broader goals. And the instinctive response to a structural problem is a structural solution: change the structure.
But silos are not, at their root, a structural phenomenon. They are a behavioural and cultural one.
Silos exist because the people in different functions have different identities, different incentive structures, different information, different priorities, and, most fundamentally, different mental models of what success looks like and who is responsible for it. A reporting line change does not alter any of these things. It changes the position on an organisational chart. The identities, incentives, information asymmetries, and mental models move with the people into the new structure and reconstitute themselves around the new boundaries.
This is why Jay Galbraith’s foundational work on organisational design is clear that structure is only one of five elements that determine how an organisation actually operates. The others are processes (how work actually flows and how decisions are actually made), rewards (what behaviours are incentivised and recognised), people (the capabilities, skills, and orientations of the individuals), and lateral capability (the ability to work across boundaries, which is a cultural and capability question, not a structural one). A structural change that does not address the other four elements will produce a different chart without producing different behaviour.
The silo problem is a lateral capability problem. It exists because people have not been developed, incentivised, or culturally supported to work across functional boundaries effectively. Moving the boundary does not fix the capability gap. It relocates it.
The Enormous Cost Of Reorganisation: And Why It Is Consistently Underestimated
Reorganisations are the most expensive transformation lever available to senior leadership, and they are consistently deployed with inadequate understanding of what they actually cost.
The direct costs are visible: change management, communication, new roles, potential redundancy. These are budgeted and tracked.
The indirect costs are much larger and almost never measured. Productivity drops during a reorganisation as people spend time in uncertainty, managing their own anxiety, seeking clarity about their new role, navigating new relationships with new managers and new colleagues. Research by Bain & Company suggests that productivity in the affected functions typically falls by 20 to 30 percent during the transition period, which for large reorganisations can be six to twelve months.
Voluntary turnover increases. The people with the most options, which generally means the most capable, leave at a higher rate during periods of structural uncertainty. They are the least tolerant of ambiguity about their future and the most in demand in the external market. The departure of these people is measured as attrition; the cost of their loss: the institutional knowledge, the relationships, the capability, is rarely quantified but consistently significant.
Customer-facing impact, in customer-serving functions, is frequently the most material cost and the least tracked. Customers experience the reorganisation as disruption in their relationship with the organisation: new contacts, changed processes, inconsistent communication during the transition. Customer satisfaction data in the period following a reorganisation of commercial functions consistently shows deterioration. In competitive markets, some of that deterioration translates into customer attrition.
These costs are real, recurring, and largely invisible in the decision-making process that approves the reorganisation. The business case typically models the structural efficiencies to be gained. It rarely models the disruption costs of getting there.
The Cultural Problem That Structure Cannot Fix
Edgar Schein’s model of organisational culture distinguishes between three levels: artefacts (the visible expressions of culture: structures, processes, stated values), espoused beliefs and values (what the organisation says it believes), and basic underlying assumptions (the deeply held, often unconscious beliefs that actually govern behaviour).
Most structural interventions operate at the artefact level. They change the visible expression of how the organisation is arranged. They leave the underlying assumptions entirely undisturbed.
The assumption that typically underlies a silo problem is something like: “My primary responsibility is to my function. My success is measured by my function’s metrics. The success of other functions is their responsibility. When their work intersects with mine, the default assumption is that they will not fully understand my constraints, and I will not fully understand theirs.”
This assumption is not irrational. It is a learned response to the incentive structures and cultural norms of most functional organisations. It is reinforced every time performance is evaluated by functional metric, every time resource competition is resolved in favour of the better-resourced function, and every time cross-functional friction is resolved by hierarchy rather than by collaborative problem-solving.
A structural change does not alter any of these reinforcing conditions. The assumption persists beneath the new chart. The silo reforms around the new boundary.
What changes the underlying assumption is different: a sustained period in which the incentives, accountability structures, and cultural norms actively reward lateral collaboration and penalise functional insularity. This is slower, harder, less visible, and cannot be announced at a town hall meeting. It is also the only approach that actually works.
What Actually Fixes Silos
The research on cross-functional collaboration effectiveness identifies a consistent set of conditions that reduce silo behaviour, none of which is structural reorganisation.
Shared accountability for shared outcomes. The most powerful mechanism for breaking down silos is a performance management system that includes shared metrics, held jointly by the leaders of interdependent functions, that reflect outcomes neither can produce alone. When a Sales director’s performance evaluation includes a customer success metric that depends on Product’s delivery quality, and the Product director’s evaluation includes a revenue metric that depends on Sales’ commitment accuracy, the incentive to collaborate is direct and personal. The silo behaviour is irrational when the incentive structure rewards collaboration.
Regular, structured cross-functional forums with decision authority. Not information-sharing meetings: decision-making forums with clear scope, clear accountability, and clear authority to resolve the specific categories of disagreement that currently escalate to senior leadership. The forums create lateral coordination without requiring structural change. They are the process-level equivalent of the structural change, and they produce the lateral coordination that the structural change promises but rarely delivers.
Cross-functional capability development. The people who are expected to coordinate across functional boundaries need specific capabilities: the ability to understand the priorities and constraints of functions other than their own, to communicate in ways that are legible across functional cultures, and to negotiate competing priorities constructively rather than escalating them. These are learnable capabilities, but they are rarely developed deliberately. Building them: through cross-functional rotations, joint training, shared projects, is more effective, and considerably cheaper, than restructuring.
Leadership behaviour that models lateral collaboration. If the senior leadership team resolves cross-functional conflict by adjudication: by senior leaders making calls on behalf of their functions in competition, the message it sends to the organisation is that escalation to hierarchy is the legitimate mechanism for cross-functional disagreement. The culture that produces silos is maintained. If the senior leadership team visibly models cross-functional collaboration, naming it when it happens, publicly recognising it, resolving cross-functional tensions in collaborative rather than adversarial ways, the culture begins to shift. Senior leaders who collaborate authentically with each other give their organisations permission to collaborate across function at every level below.
Before The Next Reorganisation: The Questions Worth Asking
I want to be precise here, because the recommendation is not “never reorganise.” Structures should evolve with strategy, with scale, with changed operating realities. There are moments when a structural change is both necessary and appropriate.
The questions worth asking before committing to one are these.
Is the problem actually structural, or is it behavioural? If you replaced every person in the affected functions with different people in the same structural arrangement, would the problem persist? If yes, it is structural. If not, it is behavioural, and a structural change will not solve it.
Have you addressed the incentive misalignments that are producing the silo behaviour? Before moving boxes on a chart, examine whether the performance management system, the budget process, and the decision rights structure are creating the conditions for the cross-functional collaboration you want. If they are not, changing the chart will not fix them.
What will the reorganisation cost, and is that cost being measured? Build the disruption model: productivity loss, voluntary turnover risk, customer impact during transition. Compare it to the benefit model. If the benefit model is compelling and the disruption cost is adequately resourced, proceed. If the disruption cost is larger than the modelled benefit, you are spending more to stay in the same place.
Is there a lighter intervention that could achieve the same outcome? A governance change, a shared metric, a cross-functional forum with decision authority, a targeted capability development programme: any of these may produce the lateral coordination you need for a fraction of the disruption cost. They should be exhausted before a structural solution is committed to.
The Most Expensive Way To Stay In The Same Place
The reorganisation is the most visible and most dramatic option available to a leadership team that wants to demonstrate decisive action on a structural problem. It changes what the organisation looks like. It produces activity, conversation, and the temporary sense that something significant is happening.
It does not change the underlying culture. It does not change the incentives that produce silo behaviour. It does not develop the lateral collaboration capability that cross-functional coordination requires. It does not alter the basic assumptions that govern how people in different functions relate to each other.
It is, at its root, a structural answer to a cultural question. And structural answers to cultural questions are not wrong. They are simply insufficient. The structure does not cause the culture. The culture causes the structure, and maintains it through every structural change.
If the cultural work is done, the structural change may not be necessary. If the structural change is done without the cultural work, the problem will wait patiently in the new structure for the organisation to notice it again.
Usually within eighteen months.
The Question Worth Sitting With
Think about the last reorganisation your organisation ran, or the one currently being considered.
Here is the question that most reorganisation discussions never ask: if you replaced every person in the affected functions with different people, but kept the same incentive structures, the same performance metrics, and the same cultural norms: would the problem persist?
If yes, the problem is not structural. And a structural solution will not find it. It will simply give it a new address.
This is Piece 10 of an eleven-part series exploring the patterns behind organisational dysfunction: and what the GITO® Approach reveals about addressing them at the system level.
The final piece: The organisation that keeps solving the same problem every two years: the series culmination, where all eleven stories converge into a single pattern, and a single choice.
