The Symptom
Operational consistency is generally assumed to erode through neglect — controls loosen, standards slip, attention wanders. That pattern exists. But a second, less intuitive pattern erodes consistency just as effectively, and it looks like the opposite of neglect: a control that keeps growing.
Consider an organisation where a pre-delivery checklist has expanded from nine items to thirty-one over six years. Each addition followed a genuine incident. Each item addresses a real, identifiable failure mode that actually occurred.
What Leaders Usually Conclude
Two opposite conclusions tend to compete for attention here, and both can look plausible from a distance. The first is that this reflects increasing rigour — a control system maturing in response to hard-won experience, and therefore something to be reassured by. The second, more common as a checklist keeps growing, is that it reflects bureaucratic accumulation — a control that has drifted past the point of usefulness and now exists mainly to be seen to exist, at the cost of the attention it demands from the people using it.
Why It Happens
The evidence available in a case like this can distinguish between the two, and it rarely gets consulted before a judgement is made. In the case above, completion data shows no decline in attention across the list as it grew, and defect rates have fallen steadily over the same period. That is evidence for the first explanation, not the second: the checklist grew because the work justified it, and growth in a control is not, on its own, proof of bureaucratic accumulation.
This matters because the two explanations point toward opposite responses. Treating a justified, well-used control as bureaucratic accumulation risks removing protections that are actually working — a form of erosion that arrives disguised as simplification. The quiet part of “quiet erosion” is precisely this: consistency can erode either because attention slips, or because a reasonable-looking correction removes something that had, in fact, earned its place. Both look the same from outside; only the evidence distinguishes them.
The more general principle is that length or complexity of a control tells an observer almost nothing on its own. What distinguishes a control that has grown appropriately from one that has merely grown is evidence of use and evidence of outcome — whether the items are actually being engaged with, and whether the metric the control exists to protect has moved. Growth in a control absent that evidence is a different situation entirely from growth accompanied by it, even though both can look identical on paper.
There is a real tension underneath this that deserves acknowledgement rather than resolution: every control, however well justified individually, imposes a cost in time and attention on the person using it, and that cost is cumulative across the whole list, not just the newest addition. A thirty-one item checklist that is fully justified item by item can still, in aggregate, represent a meaningful burden that a nine-item checklist did not. Individual justification for each addition does not automatically mean the cumulative total remains proportionate — that is a separate question, worth asking on its own terms, and one that reviewing items individually will never surface.
What It Looks Like in Practice
A control — checklist, sign-off, review step — has grown steadily over time, with each addition traceable to a specific prior incident.
Completion or engagement data exists but is rarely reviewed against the outcome the control is meant to protect.
Length alone is used as a proxy for either “too rigorous” or “well developed,” without checking which is actually true.
Periodic simplification exercises target the longest-looking controls first, regardless of whether they are the ones showing signs of genuine erosion.
The cumulative time cost of a well-justified control is rarely assessed in aggregate, even when every individual item is defensible on its own.
The same pattern appears across many kinds of organisation; see Other Sectors for how it is read elsewhere.
Questions worth considering
- Which controls in the organisation have grown the most over time, and is there evidence of whether each addition is still being engaged with?
- Has a control ever been simplified or trimmed based on its length alone, without checking what outcome data said about it?
- What would distinguish, in this organisation’s own data, a control that has grown appropriately from one that has simply grown?
- Where a control sits in a highly regulated or safety-critical operation, an auditor or regulator can often confirm whether an addition was incident-driven — in less regulated, more discretionary areas, what internal discipline makes that same distinction?
- Has any control’s reputation — as “just bureaucracy”, or as rigorous and well-earned — become self-reinforcing regardless of the underlying evidence?
- When was the evidence on a given control last checked directly, rather than relying on the organisation’s settled opinion of it?
More on this topic: Operations.
Your team is facing similar challenges?
Subscribe to The Riddle Ledger
Perspectives are sent as they are published, with no promotional email, and you can unsubscribe at any time.