I mapped a broken process once and found six silos hiding inside it

That was the moment I stopped treating silo breakdown as a side project and started treating it as the work itself. What I found inside that process map was not one wall between two teams — it was a series of invisible barriers, each one reasonable in isolation, each one contributing to a system that moved slowly, decided badly, and exhausted the people trying to make it function. This page is where I document the patterns, the frameworks, and the hard-won observations that came out of that work and everything that followed. If you are a practitioner who is already inside a similar problem, I built this resource specifically for you. No jargon. No theoretical models that require a PhD to interpret. Just structured thinking about a structural problem, explained plainly enough to act on.

Alex Mercer

Alex Mercer

Lead Process Optimisation Practitioner

Team mapping a cross-departmental process on a whiteboard

The Root Cause

Silos are not caused by difficult people — they are the output of a system that rewards local optimisation

Every team inside a siloed organisation is behaving rationally. Finance protects its reporting cycle because it has learned that sharing data mid-cycle creates reconciliation problems. Operations guards its throughput metrics because they are what it is measured on. Customer service closes tickets quickly because ticket resolution time is the number that appears on the dashboard. Each team is doing the logical thing. The system is producing an illogical result. This is the thesis-antithesis at the heart of silo formation: individually reasonable decisions aggregating into a collectively unreasonable structure. The fix is never about persuading people to behave differently. It is about redesigning the conditions that make the current behaviour the path of least resistance.

Contact me

The cross-functional alignment audit — what it is and why the sequence matters so much

This is the core diagnostic framework I apply when working through a silo breakdown problem with a practitioner team. It is not a checklist. It is a structured sequence of observation and analysis that builds a complete picture before recommending anything.

Most organisations attempt to fix silos before they have accurately diagnosed them. The audit exists to prevent that mistake.

The first phase of the audit is pure observation. I map the real communication pathways inside the organisation — not the org chart, but the actual routes information travels to reach the people who need it. This involves tracing specific decisions backward through the organisation to identify every point at which information changed hands, got delayed, got filtered, or simply stopped moving. The map that results from this phase is almost always different from what leadership expects to see.

The second phase is gap analysis. I compare the observed information flow against the process outcomes the organisation is trying to achieve. Where the two diverge, there is a structural gap — a place where the design of the system is working against the intent of the people operating within it. These gaps are the precise locations of the silos. They are rarely where the symptoms are most visible, which is why diagnosis must precede prescription.
The third phase is intervention design. Armed with a complete gap map, I work with the practitioner team to design the smallest structural changes that will produce the largest improvement in cross-functional flow. The emphasis on smallest is deliberate. Large structural redesigns create resistance and rarely hold. Small, targeted changes to handoffs, shared metrics, and escalation protocols compound over time into a fundamentally different organisational dynamic.

The fourth phase is reinforcement design. This is the phase most organisations skip, and it is the reason most silo-breakdown efforts do not hold beyond the first few months. Reinforcement means identifying the specific incentive structures that would cause the old behaviour to reassert itself — and changing them before they get the chance. It also means establishing the joint review cadences and shared dashboards that keep cross-functional health visible to everyone who needs to maintain it.

I have studied a range of failed integration attempts across different sectors. The failure modes are surprisingly consistent — and almost entirely avoidable once you know where to look.

What the research and observation consistently show about why silo-breakdown efforts fail

The organisations that fail to dissolve silos are not less capable than the ones that succeed. They are making predictable structural mistakes.

The most common failure mode is prescribing before diagnosing. An organisation identifies that two departments are not collaborating effectively and immediately introduces a coordination mechanism — a weekly cross-functional meeting, a shared project management platform, a liaison role. None of these interventions address the underlying structural conditions. The meeting becomes a place where each team reports its own metrics to the other. The platform becomes another silo in software form. The liaison role becomes a bottleneck. The symptom is managed; the cause is untouched.

The second most common failure mode is measuring the wrong things after the intervention. An organisation successfully redesigns a cross-departmental handoff and then continues measuring each team on its own local metrics. Within months, local optimisation reasserts itself and the redesigned handoff starts degrading. The teams are not being obstructive — they are responding rationally to the incentives they are being managed against. Changing the process without changing the measurement framework is like clearing a blocked pipe without addressing the pressure that created the blockage.

The third failure mode is attempting too much at once. Large-scale structural redesigns generate resistance precisely because they are large. Every team that is asked to change simultaneously becomes a potential point of failure. The organisations that sustain cross-functional improvement are the ones that start with the single handoff that is causing the most friction, fix it visibly and completely, and use that success as the evidence base for the next intervention. The compounding effect of sequenced small wins is more durable than any single large transformation.

The fourth failure mode is abandoning the reinforcement phase when early results look promising. A successful intervention produces visible improvement quickly — decisions move faster, data duplication reduces, cross-functional projects start completing on time. This is the moment when external support is most commonly withdrawn and when the conditions for regression are most quietly assembling themselves. Reinforcement is not a formality. It is the phase that determines whether the change holds for six months or six years.

Analogies that make structural problems visible and actionable

I use analogies deliberately in this work. Not as decoration, but as diagnostic tools. The right analogy makes a structural problem visible to someone who has been too close to it to see it clearly — and that moment of clarity is often what unlocks the conversation about how to fix it. The plumbing analogy is the one I return to most often. A siloed organisation behaves like a water distribution network where each neighbourhood controls its own valve and optimises for its own pressure. Each neighbourhood is doing the sensible thing. The overall network is producing an inefficient result — water backing up in some places, going to waste in others, with nobody responsible for the system as a whole. The fix is not to ask each neighbourhood to be less selfish. The fix is to redesign the network so that optimal local behaviour and optimal system behaviour point in the same direction. The ecology analogy works well for explaining why silos persist even after successful interventions. A habitat cleared of an invasive species will be recolonised if the underlying conditions that made it hospitable to that species remain unchanged. Dissolving a silo without changing the incentive conditions that created it produces the same result — the old structure gradually reasserts itself because the environment still favours it. The traffic engineering analogy is useful for explaining the sequencing principle. Traffic engineers do not redesign an entire road network at once. They identify the single intersection that is causing the most system-wide delay, fix it, observe the effect on overall flow, and then identify the next highest-impact intervention. The same principle applies to silo breakdown. Start with the handoff that is causing the most downstream friction. Fix it completely. Measure the effect. Then move to the next one. The compounding effect of well-sequenced targeted interventions is how lasting systemic change is built without overwhelming the people doing the actual work.
Interconnected ecological system illustrating structural conditions
Network of pipes representing an organisational flow system

Key Concepts

Shared metrics design

Most organisations measure teams on what those teams directly control. This is understandable and often counterproductive. When every team is optimised for its own metric, the space between teams — the handoffs, the escalations, the joint decisions — becomes an unmeasured void. Nobody is accountable for what happens in that space, so nothing improves it. Shared metrics design is the practice of identifying the two or three indicators that genuinely reflect cross-departmental health and making them visible to everyone who influences them. Handoff speed. Decision overlap frequency. Escalation resolution time. When these numbers are on the shared dashboard, the incentive to maintain walls weakens naturally. Transparency, applied to the right indicators, is one of the most structurally powerful interventions available.

Handoff mapping

A handoff is the moment when a process crosses a departmental boundary — when information, a decision, or a task moves from one team's ownership to another's. Most handoffs inside siloed organisations have never been deliberately designed. They exist as informal conventions — whoever usually handles this passes it to whoever usually receives it, through whatever channel happens to be available. Handoff mapping is the practice of making these transitions explicit: documenting exactly what information changes hands, in what form, through what channel, and with what expected response time. Once a handoff is mapped, its failure modes become visible. And visible failure modes are fixable ones.

Team reviewing a shared metrics dashboard

Incentive misalignment audit

Before designing any structural intervention, I examine the incentive landscape — the formal and informal rewards that shape team behaviour. What does success look like for each team, as defined by the metrics they are managed against? Where do those definitions of success conflict with each other? Where does optimising for one team's metric create friction for an adjacent team? The incentive misalignment audit makes these conflicts explicit and provides the basis for a redesign that aligns local incentives with collective outcomes. It is not a comfortable process — it often surfaces decisions made at the leadership level that are inadvertently sustaining the silos the organisation is trying to dissolve. But it is the only foundation on which durable structural change can be built.

Cross-functional process handoff diagram on whiteboard

If any of the concepts described on this page match what you are observing inside your organisation, the most useful next step is a direct conversation. Tell me what you are seeing and I will tell you what I have seen work in comparable situations. No formal brief required — just describe the friction.

Six things I have learned about silos that most organisations discover too late

These are not theoretical observations. They are patterns I have encountered repeatedly, across different sectors and organisation sizes, working through real structural problems with practitioners who are already inside them. Each one took me by surprise the first time I encountered it.

    1

    The symptom is never where the silo is

    The place where friction is most visible is almost never the location of the underlying structural problem. A customer complaint about slow resolution usually traces back to a handoff failure two or three steps earlier in the process — a point where information should have moved between teams and did not. Fixing the visible symptom without tracing it back to its structural origin is why so many interventions produce only temporary relief.

    2

    The most dangerous silos are the invisible ones

    A silo between finance and operations is usually visible — the tension is obvious and the friction is complained about openly. The silos that do the most long-term damage are the ones nobody has named yet: the ones that have been normalised into the organisation's working assumptions. These are the hardest to diagnose and the most important to surface, because they have been quietly shaping decisions for long enough to feel like facts rather than design choices.

    3

    Workarounds are the most reliable silo indicators

    When I begin a mapping exercise, I look for the workarounds first. The manual reconciliation spreadsheet. The informal Slack channel that bypasses the official system. The team member who is informally responsible for translating between two departments that cannot communicate directly. Every workaround is a structural gap made visible. The workaround is the organisation's attempt to function despite the silo — and it is the clearest possible signal of exactly where the intervention needs to focus.

    4

    Leadership alignment is necessary but not sufficient

    One of the most common errors I observe is treating leadership agreement as the intervention itself. Two department heads agree that collaboration needs to improve, shake hands, and return to managing their teams against the same local metrics they were managing against before the conversation. Nothing changes. Leadership alignment creates the permission for structural change. It does not produce the structural change. The work still needs to happen at the level of incentives, handoffs, and shared measurement.

    5

    Small wins compound faster than large transformations

    The organisations that sustain cross-functional improvement are the ones that sequence their interventions deliberately, starting with the smallest change that will produce a visible result. That first result — a handoff that now works, a decision that now takes days instead of weeks — creates the trust and evidence base needed to justify the next intervention. The compounding effect of sequenced small wins produces more durable change than any single large-scale redesign, and it does so without overwhelming the people who have to keep the organisation running while the change is happening.

    6

    Regression is a design problem, not a discipline problem

    When a successfully dissolved silo reasserts itself after a few months, the instinctive response is to blame the people who reverted to old behaviour. This is almost always the wrong diagnosis. Regression happens when the structural conditions that created the silo in the first place have not been changed. The teams are responding rationally to the incentives they are still being managed against. Preventing regression is a design problem — it requires changing the measurement framework, not repeating the intervention.

Cookies

This site uses cookies to understand how visitors interact with content and to ensure core functionality works correctly. No cookies are used for advertising or cross-site tracking.
Read our Cookie Policy