Your CFO asks a simple question: which of last quarter’s projects actually made money?
Nobody can answer with one number. Hours live in your time tracking system. Costs live in accounting. Invoices and payments live somewhere in between. Three systems, three teams, no shared definition of “project margin”.
That gap is where your profitability hides. It’s also why finance is the right team to own operations analytics in a services firm.
The Wrong-Sponsor Problem
In most services firms I see, operations analytics is owned by IT or by the operations team itself. Both groups produce dashboards. The dashboards rarely change behavior. The reason is structural: IT optimizes for system uptime, operations optimizes for utilization in this week’s plan. Neither has the standing to insist that data crosses department lines.
Finance does. Finance already touches every billable hour as revenue, every cost as a P&L line, every payment as cash. If you want one team that can credibly demand a single project margin number — and defend it — that team is finance.
What Finance Brings That Others Don’t
Three things, specifically:
- Definitional discipline. Finance is used to definitions that don’t shift. In a services firm, billable, billed, recognized, and collected are four different things, and people use them interchangeably. Finance won’t let that slide.
- End-to-end reach. Project profitability requires connecting hours to costs to invoices to cash. Only finance has the standing to pull all four into one model.
- Accountability for outcomes. Finance owns the P&L. When operations analytics works, the result shows up on the books. When it doesn’t, finance feels it first.
This isn’t solo work — your delivery leads know the projects, sales knows the pipeline. Finance leads because finance has to defend the numbers in the end.
Three Places to Start
If finance is going to own operations analytics, the first projects should pay for themselves. In a services firm, three usually do:
1. Project profitability. Hours times cost rate against billings. Net of bench time, scope changes, write-offs. Most firms have a feel for this; very few have one number per project they trust.
2. Utilization, properly calculated. Billable hours over capacity, broken down by team, role, and project. The hard part isn’t the formula — it’s getting everyone to agree on what counts as capacity and what counts as billable.
3. Order-to-cash for services. Time entered → invoice issued → cash received. The cash conversion cycle directly affects your bank balance. Shorten it by a few days and you free up real money — money you don’t have to borrow.
Each of these requires the same foundation: a single integrated dataset connecting operations to finance.
The Three-System Problem
For most services firms, that integration spans:
- Time tracking system — for hours, projects, and client assignments
- Accounting — for cost rates, revenue, and the ledger
- Invoicing and payment tracking — sometimes inside accounting, sometimes outside
Each system has its own keys, its own definitions, its own version of “completed”. Stitching them together is a data engineering job, not an analyst’s job. You need a data warehouse where one project, one consultant, and one client mean the same thing in all three sources.
The Messy Middle
Services firms are full of exceptions, and you can’t pretend they don’t exist.
Time entries corrected three weeks late. Fixed-price projects where hours don’t tie to invoices. Retainers drawn down unevenly. Scope changes someone agreed to in a meeting and never wrote down. Bulk invoices covering multiple projects. Write-offs to keep a client happy.
Some of this you can model. Some you can’t, because the source data simply doesn’t carry enough information. When that happens, don’t force the model. Step back and look at the process. In my experience, the data work surfaces process problems that have been quietly costing money for years.
What This Means For Your Next Move
If your operations analytics keeps stalling, look at who owns it. If it’s IT or operations, that’s probably the issue. Move sponsorship to finance. Pick one of the three starting points — project profitability, utilization, or O2C — and build the integrated dataset that all three eventually need anyway.
This is the work I do with services firms. Data engineering for the integrated dataset, BI strategy for the finance-led roadmap. Time tracking, accounting, AR — one model, numbers you can defend.
Want to talk about who should own operations analytics in your firm? Book a 30-minute call.