What leaders usually conclude
Capacity is part of the story, and it is real — two fewer people genuinely means less throughput. But capacity alone does not explain why delivery on this specific piece of work slowed so much more than the headcount reduction would predict. One of the two people who left had, informally, been the sole reviewer for a particular area of work. That review function has not been replaced, because it was never written down as a distinct role — it was simply something she did, reliably, without a title attached to it.
What the evidence suggests
Erik Hollnagel, Robert Wears and Jeffrey Braithwaite’s work on the difference between “work-as-imagined” and “work-as-done” offers a useful lens here. Organisations tend to believe their documented processes describe how work actually happens; in practice, people make continuous, unrecorded adjustments to keep things running, and those adjustments often carry more of the operational weight than the formal process chart suggests. As the researchers put it, “things go well because people make sensible adjustments according to the demands of the situation” — not because everyone is simply following the documented procedure. When the person making those adjustments leaves, what disappears is not visible in any org chart, because it was never on the org chart to begin with.
This connects to a broader point from Jay Galbraith’s work on organisation design: growing organisations accumulate genuine information-processing and coordination demands, and informal roles — someone who happens to catch errors, someone who happens to know which client relationships are fragile — often carry a meaningful share of that load without any formal recognition. Losing that person removes a function the firm did not know it was relying on.
Nelson Repenning and John Sterman’s research on what they call the “capability trap” helps explain why this informal role is so rarely formalised before it is lost. Building the underlying capability that would let the firm function without depending on one person’s judgement — documenting standards, training a genuine second reviewer, building redundancy into how complex work is checked — takes real, sustained investment of time that competes directly against the next billable deadline. As the researchers put it, “nobody ever gets credit for fixing problems that never happened,” and a firm under normal day-to-day delivery pressure will consistently choose the visible, urgent client work over the invisible, preventive work of building capability, right up until the moment the capability gap is suddenly and expensively exposed by a departure.
Competing explanations
A firm might reasonably conclude the slowdown is simply a documentation gap — write down what the departing person did, and the loss is contained. This is only partly right. Some of what left was tacit judgement built through years of exposure to unusual cases, which resists being reduced to a checklist even when the willingness to document exists. Treating the loss purely as a documentation failure risks understating how much of it was capability, not information.
A further complication is that the two departures happened together, which makes it harder still to separate their individual effects. Where a firm loses one person at a time, the specific function that person carried is at least somewhat easier to isolate retrospectively. Simultaneous departures compress that diagnostic work into a single, more disruptive event, and firms under that pressure often default to the fastest available explanation — capacity — because it points to the fastest available response: hiring. The risk is not that hiring is the wrong response; it is that treating it as the complete response leaves the informal review function unreplaced indefinitely, simply because nobody framed it as a separate thing to solve for.
It is also worth distinguishing this from a simple staffing shortfall. A firm that loses two junior staff performing clearly defined, documented tasks faces a real but comparatively straightforward capacity gap. Losing someone who held an undocumented, informally recognised function is a different kind of loss precisely because the firm cannot easily specify, in a job description, what needs replacing — it can only notice, after the fact, that something is missing, which is a much harder problem to hire against.
A further wrinkle worth naming directly: the departing person themselves often could not have fully specified the role either, even if asked in advance. Much of what she caught, she caught through pattern recognition built over years — a phrasing that seemed slightly off, a client whose tone suggested unease before they had said so directly — and that kind of judgement resists being written into a checklist even by the person who holds it. A firm hoping for a clean handover, where the departing person simply documents what they do before leaving, is often disappointed not because the person was unwilling to help but because a meaningful share of what they contributed was never fully explicit to them either.
Where this shows up
The pattern described here is most recognisable in professional services firms, part of the wider Professional and Knowledge-Based grouping. The same organisational questions appear, under different names, in other sectors.
Questions worth considering
- When someone experienced leaves, has the firm actually separated what was lost — documented knowledge, tacit judgement, or an informal function nobody had named?
- Is there a role in this firm being performed reliably by one person that has never been written down anywhere?
- Would replacing headcount alone restore what slowed down, or only part of it?
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.