Saoirse Mulrennan manages finance at a digital agency with a team of 55. The agency runs between 35 and 45 active client projects at any point. For years, overall agency profitability looked acceptable. Project-level analysis told a different story. We spoke with her about what they found and what they changed.
The gap between agency profit and project profit
How did you discover the project-level problem?
We started tracking time-to-budget ratios more carefully. A project might be billed at €60,000 and show a healthy gross margin on paper. But when we mapped actual hours logged against the fee, some projects were consuming 40% more time than the estimate. The margin was being absorbed by unrecovered labour, not visible in the top-line figures.
The monitoring system
What does project-level tracking look like now?
Every project has a budget in hours and in euros, broken down by phase. Project managers update their hour estimates weekly in our PM tool, which feeds into a finance dashboard. We track three figures for each project: budget hours, hours logged to date, and hours estimated to completion. When those three numbers do not reconcile, we investigate before the project closes.
We also introduced a mid-project budget review at the 50% completion mark. If a project has consumed 65% of its hour budget at the halfway point, that triggers a conversation with the client about scope or with the team about efficiency. Neither conversation is comfortable, but both are easier to have mid-project than at invoice stage.
What changed
Has this affected how you price new work?
Significantly. We now use historical actuals from similar projects as the basis for estimates, not optimistic assumptions. The estimates are less competitive in some cases, but the projects finish where we expect them to.