Before a professional services firm rolls out an AI tool, its CFO should decide where the saved hours will go, whether into more billable work per person, fewer contractor hours or slower hiring, and put that choice on the monthly review.
BCG's AI at Work survey of close to 12,000 frontline employees, managers and leaders in more than a dozen global markets, published in June 2026, found that among frontline employees who use AI regularly, 42 percent report saving eight hours a week. In the same survey, 66 percent of those regular users still receive limited or no guidance on what to do with the time they save, and more than half say they are not reinvesting time saved into more strategic work.
Close to half or more of respondents also report spending more time reviewing and correcting AI output or managing and directing AI, so the hours a team reports saving should be counted net of that time before anyone decides where they go. BCG's advice to CEOs is to watch business outcomes rather than usage: "The time that individuals save leaks out of the organization unless it is tracked and deliberately reinvested."
Each destination already has a line in the monthly pack: for more billable work, utilization in the teams using the tool, meaning billable hours as a share of available hours; for contractors, contractor hours and spend; for hiring, heads against the hiring plan. The CFO picks one for each team before launch, records it for the three months beforehand, and reports it every month after, next to what the software costs.
On hourly engagements, an hour the tool saves on client work is an hour the firm cannot bill unless it is refilled with other client work. More billable work therefore needs a pipeline to fill the hours, and the partner who holds the pipeline and sets pricing signs off on that choice. Fewer contractor hours become a saving only once fewer are bought, which puts that choice with whoever approves contractor time. Slower hiring needs a hiring plan written before rollout with the tool left out of it, because a plan that already assumed slower hiring would count the same saving twice.
For a consulting firm rolling out an AI tool for research and proposal drafting, the choice might be more billable work, with the proposal hours the tool saves moving to client work. The monthly entry reads: the software's cost, utilization in the teams using the tool and proposal hours per bid, each against the average of the three months before launch, and the partner who staffs those teams as the owner. Counting proposal hours per bid keeps a busy month of bidding from hiding the tool's effect. Utilization is also read against the same months a year earlier, because holidays and year-end move it whatever the tool does. If neither figure has moved by the third review, the rollout stops expanding until that partner can say where the time went.
Slower hiring is the destination many budgets already assume: in Gartner's October 2025 survey of more than 300 CFOs and finance leaders across industries, published in February 2026, expected headcount growth fell from 6 percent in 2025 to 2 percent in 2026.
