7 Transformation Signals That Suggest Your Operating Model Is Breaking
These operational patterns often emerge long before transformation programmes visibly fail.
- Operating models rarely break in one moment. They drift, until the cumulative drift becomes visible. The cost of catching the drift early is materially lower than the cost of catching it late.
- The seven signs below appear in clusters. Recognising two or three is normal in growing organisations. Recognising five or more usually means structural intervention, not optimisation.
- The decisive test is whether the executive team can describe the current operating model in their own words — and whether their descriptions agree.
A transformation programme rarely fails suddenly. It fails gradually, in patterns the executive team can see months before the board does — patterns that quietly suggest the operating model itself is no longer doing what it used to.
The signs below describe those patterns. None of them are catastrophic on their own. None require a programme to be in crisis. They are the operating-model warning signs that emerge while the formal scorecards still show acceptable performance — and they are the ones CEOs of UK mid-market organisations recognise immediately, often before they have language for them.
The pattern matters because the moment when these signs are visible is also the moment when the cost of intervention is at its lowest. By the time the formal indicators have shifted — revenue softening, employee engagement declining, customer NPS dropping — the cost of remediation has compounded several times over. Recognising the soft signals early is the discipline that protects the harder ones.
What the seven signs share is a structural origin. They are not individual performance failures. They are not market headwinds. They are symptoms of an operating model that has accumulated complexity without the architectural discipline to keep up. They tend to appear together, in clusters, in the same organisations.
Decisions that used to be obvious are now slow and contested
Three years ago, a category of decision — a hiring authorisation under a certain threshold, a marketing campaign approval, a vendor selection in a familiar category — took a few days and one or two people. Today it takes weeks. More people are involved. More versions of the proposal are required. More questions are asked. The decision itself, when it finally arrives, often looks similar to what was proposed at the start.
This is the most universal early sign. The organisation has not lost the ability to make decisions. It has lost the agreement about who is allowed to make them, on what evidence, against what criteria.
The mechanism is unintentional. As organisations grow, decisions accumulate consultees. Each consultee was added for a good reason — a previous mistake, a stakeholder request, a compliance requirement. None of the additions are evaluated against each other. The decision-rights architecture is never rewritten. It just accretes. The fix is structural, not procedural. Adding more process to a decision-rights problem makes it worse.
Functional KPIs trend positively while business outcomes do not
The most diagnostic sign for the CEO seat. Marketing reports hitting its targets. Sales reports hitting its targets. Customer service reports hitting its targets. Operations reports hitting its targets. The aggregate business performance — revenue, margin, share, customer lifetime value — is flat or declining.
This is the signature of a decoupled operating model. The functions are each optimising for their local metric, but the metrics no longer aggregate to the business outcome they were designed to drive. Marketing’s lead volume is high, but the leads convert poorly. Sales’ close rate is high, but the deals churn early. Customer service resolves quickly, but the resolutions don’t address the root issue.
The pattern is structural. The functions are not failing. The connections between them have eroded. The handoffs have lost integrity. The metrics that span functions — customer lifetime value, true conversion economics, end-to-end cycle time — are not being measured because no single function owns them. The fix is operating-model design at the seams between functions. Not new departmental KPIs.
The signal that hurts most: every function is hitting its number, and the business isn’t.
The same problem keeps surfacing under different names
A customer complaint becomes a “service issue”. The service issue becomes a “data quality” issue. The data quality issue becomes a “platform integration” issue. The platform integration issue becomes a “vendor relationship” issue. Each version of the problem is logged separately. Each version is owned by a different function. None of them are resolved at root.
Eighteen months later, the same problem — visible to the customer, at least — is still there. It has just been re-named more times. The organisation has expended significant operational effort on it, none of which has touched the underlying cause.
This pattern is hard to spot because each individual instance is reasonable. The service team really did receive the complaint. The data team really did see the quality gap. The integration team really did need to fix the pipeline. The vendor team really did need to renegotiate. Each function did its job. None of them was structured to fix the architectural cause. The diagnostic is to trace a single recurring customer complaint backwards. If it has been logged in four different categories under four different owners over two years, the issue is architectural.
Workarounds are being formalised into process
The first time a team builds a workaround, it is a workaround. By the third time, it has become unofficial process. By the tenth time, it is being trained into new hires as if it were the operating model. The original system or process the workaround was avoiding is still there. Nobody references it anymore.
This is the most quietly corrosive sign. The organisation has stopped fixing things. It has started building around them. The energy required to do this is enormous, and it is dispersed across hundreds of small accommodations that nobody owns at executive level.
The cost shows up as friction, not failure. New hires take longer to become productive because the real operating model exists nowhere but in people’s heads. Compliance and audit become harder because the documented process and the actual process have diverged. AI deployment becomes unreliable because it is being trained against documented process while operations are running on the workaround. When the workaround has become the operating model, the documented process is fiction — and most leadership teams are reviewing the fiction.
Recognising the signals is one thing. Naming where your operating model actually sits is the next.
The free Commercial Readiness Assessment positions your organisation on the Architecture Map across six dimensions of commercial operating model. About ten minutes. No payment. No sales call.
Take the Free Assessment →Executive time is consumed by exception handling, not strategy
The CEO finds themselves resolving operational details that would have been resolved two levels down five years ago. Pricing exceptions. Vendor disputes. Customer escalations. Staffing decisions on individual hires. Each request is reasonable. Each one is sub-strategic.
Cumulatively, they consume the executive bandwidth that should be invested in strategic horizon-setting. The leadership team becomes a senior expediting function rather than a strategic governance function. The board notices something is wrong but cannot easily name it. Decisions made at executive level under time pressure tend to be more conservative than decisions made by the people closest to the customer — and the organisation gradually becomes less ambitious.
The mechanism is decision-rights erosion combined with risk-aversion compounding. Each escalation that the executive accepts confirms that this category of decision belongs at the executive level. The next person facing a similar decision escalates earlier. The pattern reinforces itself. Adding “empowerment” language without rewriting the decision-rights map does nothing — people respond to who actually decides, not to who is told they may decide.
Visible structure. Decisions are clear. The operating model and the documented model agree.
Workarounds emerging. Decisions slowing. Friction visible but absorbed by extra effort.
Outcomes diverging from plans. Functions hitting targets while the business does not.
Programme reset required. Board involvement. Strategic optionality already lost.
The annual planning process produces lower-quality outputs each cycle
The annual planning exercise still happens. Plans get written. Targets get set. Budgets get allocated. But each year, the plans get less specific. The targets get more abstract. The accountability becomes more diffuse.
This is the late-stage sign. By the time annual planning has eroded, the organisation has lost confidence in its ability to predict outcomes. The participants know — without saying it — that the plan does not really drive what happens. Real decisions get made through other mechanisms, and the planning exercise becomes increasingly performative.
The board notices the symptom before they understand the cause. They challenge the plan. They ask for more detail. They request scenario modelling. The exercise becomes more elaborate. The outputs become no more reliable. The relationship between plan and reality has decoupled because the operating model that the plan describes is no longer the operating model the business runs. The fix is not to invest in better planning tools. It is to recognise that planning quality is a downstream measurement of operating-model integrity. Where the operating model is precise, plans are precise.
Three diagnostic tests you can run this quarter
- Take a single recurring decision and time it now versus three years ago. If the cycle time has more than doubled, decision rights have eroded.
- Ask three operational team leads to describe one end-to-end process — in their own words, without referencing documentation. If their answers conflict on first-order questions, the workaround has become the operating model.
- Count the decisions on the executive team’s desk this quarter that would have been resolved at director level five years ago. If the trend is upward, the decision-rights map needs explicit rewriting.
New hires take longer to become productive — even good ones
The strongest hires are still being made. The candidates look excellent in interviews. References check out. The package is competitive. But six months in, they are underdelivering. Eighteen months in, they are at risk of leaving — sometimes citing reasons that the leadership team finds it hard to argue with.
This is the sign that is hardest to attribute to operating-model issues, because the conversation usually runs through individual performance. Each underperformance is reviewed against the individual, not against the system they joined. The pattern is only visible when the executive team aggregates across hires over a two-or-three-year window.
The cause is structural. New hires can only learn the documented operating model. The actual operating model exists in workarounds, in tribal knowledge, in informal escalation paths, in the unwritten rules about who really decides what. Senior hires usually figure it out within a year. Some never do. Either way, the productivity ramp lengthens with every additional layer of workaround — and the cost is paid in talent the business cannot afford to lose.
What this means for the executive team
These seven signals do not arrive alone. They cluster. An organisation showing two or three is normal in any growing business — friction is the price of growth, and some friction is healthy. An organisation showing five or more is signalling structural drift, and structural drift compounds.
The instinctive response to operating-model drift is to apply more management — more meetings, more dashboards, more KPIs, more reviews. The instinct is wrong. Adding management to a structural problem produces more reporting, not less drift. The actual work is upstream. Define the operating model precisely. Map the decision rights explicitly. Name the metrics that span functions. Document the customer journey end to end, in agreement.
This is architectural work, not optimisation work. It is the precondition for everything else — for new platform implementations, for AI deployment, for transformation programmes that survive their first contact with reality. Where the operating model is precise, transformation runs faster and cheaper. Where it is imprecise, transformation amplifies the drift.
The starting point is naming where you currently sit.
Is the pressure on your operating model structural or transitional?
The free Commercial Readiness Assessment positions your organisation across six dimensions of commercial architecture. You receive a personalised report naming where your operating model is strongest, where it is most exposed, and which of the seven signals above are most likely to be present.
Take the Free Assessment →About 10 minutes · No payment · No contract · No sales call