A firm wins more work, hires to cover it, and a year later margin is flat or worse — despite the top line looking healthy. It is a familiar pattern, and the instinctive explanations (pricing was too aggressive, staff cost too much) are sometimes right and often miss the actual cause.

01. Where the assumption breaks down

The assumption behind “more projects should mean more margin” is that each additional project costs roughly the same, proportionally, to deliver as the ones before it — so fixed costs get spread thinner and margin improves with scale. That assumption holds when delivery is systematised. It breaks down when delivery still depends on manual, repeated, person-dependent steps, because those steps do not get cheaper with volume. They get more numerous.

02. The specific costs that scale badly

  • Rework. Every project run without a proper review-and-approval gate produces some rate of “it comes back”. At low volume that is an occasional annoyance. At higher volume it is a predictable tax on every project, and it compounds with the next issue.
  • Re-keying and reformatting. The same manual step, repeated once per project instead of once per system. Ten projects means the same clerking work done ten separate times rather than built once into how the workflow runs.
  • Knowledge concentrated in one person. A workflow that runs well because one experienced person carries it in their head scales exactly as far as that person’s calendar does. More projects means more demand on the same bottleneck, not less.
  • Approvals with no visibility. Work finished and waiting in an inbox for sign-off. At low volume someone eventually notices and chases it. At higher volume backlog accumulates quietly, and nobody can say from the outside how much work is stuck or for how long.

03. Why this is invisible in most reporting

Standard financial reporting tracks revenue, cost of delivery in aggregate, and margin at the portfolio level. It rarely isolates cycle time, rework rate, or backlog per workflow — the specific measures that would actually show this pattern forming. By the time margin compression shows up in the numbers finance actually watches, it has already been happening for a while.

04. What actually moves the number

Not working faster — that is the wrong lever, and it is usually the first one tried. What moves it is changing the path the work takes: removing the re-keying rather than asking someone to type faster, structuring intake so evidence arrives with the request instead of getting hunted down separately, and building review and approval into the workflow as visible gates rather than an inbox nobody can see into. That is the specific shape of what a connected delivery system does, and why we measure a workflow’s baseline — expert hours, cycle time, rework rate — before proposing anything to fix it.

05. Where to start

Not with a transformation programme across every service line. With the one recurring deliverable that is most economically material to the margin problem — mapped and measured on its own, so the fix is judged against a real number rather than a hope.

Bring the workflow that is actually costing you margin, and we will tell you honestly whether it is worth measuring — what each engagement costs →