Professional services analytics: utilization, project profitability and pipeline
A services firm sells time, so three numbers decide its profit: how much of the available time is billed, how much of the billed value is collected, and whether projects cost more to deliver than they earn. Formulas, examples and a setup plan.
A consulting firm, agency or accounting practice sells time. Its finances look simple on paper:
Revenue = billable hours × average rate collected
Profit = revenue - (people costs + overhead)
In practice three levers decide whether the margin is 5% or 25%: how much of the available time is billed (utilization), how much of the standard rate is actually collected (realization), and whether projects cost more to deliver than they earn (project margin). This guide covers each, plus the pipeline metrics that tell you whether next quarter will be busy or quiet.
Utilization
Utilization = billable hours / available hours
Available hours are the hours someone could work after holidays, leave and company closures. A worked example: 40 hours a week for 46 working weeks is 1,840 available hours a year. Someone who bills 1,380 hours runs at 1,380 / 1,840 = 75%.
Not everyone should aim for the same figure. Common rules of thumb (they vary by firm and field, so compare with your own history):
| Role | Typical target | Why |
|---|---|---|
| Consultants and delivery staff | 75 to 85% | Main delivery role, with some time for training and internal work |
| Managers and project leads | 60 to 70% | Delivery plus sales support and people management |
| Partners and principals | 35 to 50% | Mostly client relationships, sales and running the firm |
Firm-wide figures help with planning, but the useful signals come from utilization by person, team and month.
What to watch:
- Bench time rising. More people between projects points to a pipeline gap or a skills mismatch.
- Sustained over 90%. Short-term revenue, long-term burnout, mistakes and resignations.
- Big gaps between teams. One practice at 80% and another at 55% is a staffing or demand imbalance.
- Seasonal dips. August and December are predictable. Plan for them rather than react.
Realization
Billing a full timesheet is not the same as collecting for it.
Realization = amount billed / (hours worked × standard rate)
If a team logs 200 hours at a standard rate of $150 ($30,000) and the client is invoiced $25,500 after a discount and a write-off, realization is 85%. Low realization on one client or one service line usually means scoping or pricing needs attention. Track write-offs by client and project type so the pattern is visible.
Project profitability
A project that brings in $200,000 is not profitable if it consumed $220,000 of labor and expenses.
Project margin = (project revenue - project costs) / project revenue
Costs include loaded labor (salary, benefits and a share of overhead for every hour on the project), direct expenses (travel, software, subcontractors) and unbilled scope.
Worked example: a fixed-fee project at $60,000 takes 400 hours. The loaded cost is $80 an hour, so labor is $32,000, and there are $3,000 of expenses. Margin = ($60,000 - $35,000) / $60,000 = 42%. The effective rate is $60,000 / 400 = $150 an hour. If the same project had taken 550 hours, labor would be $44,000, margin 22%, and the effective rate $109.
Where projects lose money
- Scope creep without change orders. Compare estimated with actual hours every week, not at the end.
- The wrong staffing mix. A senior person doing junior work at three times the cost, or a junior person taking three times as long on senior work.
- Poor estimates. Track estimate accuracy by project type so proposals improve.
- Unbilled work. Hours over a fixed-fee budget, goodwill write-offs, invoices sent late.
A project health view shows, for each active project: budget used against work completed, hours forecast at completion against the estimate, projected margin at the current burn, and a status for anything over budget or behind.
Pipeline and demand
Services pipeline has an extra twist: you forecast demand for specific skills at specific times, not only revenue. A $500,000 win that needs five data engineers when you have two is a delivery problem.
Weighted pipeline = sum of (deal value × stage probability)
With a $200,000 proposal at 50% and a $100,000 early conversation at 20%, the weighted pipeline is $100,000 + $20,000 = $120,000.
Pipeline coverage = weighted pipeline for the period / revenue still needed in the period
Other numbers worth tracking:
- Win rate by service line, so business development time goes where it converts.
- Sales cycle length, from first call to signed contract. Longer cycles mean building pipeline earlier.
- Proposals per win. Many proposals per win suggests you are chasing the wrong work.
- Backlog in months = contracted but undelivered revenue / monthly delivery capacity.
Then match demand to capacity: hours already committed per role per month, plus hours the weighted pipeline would need, against hours available. The gap tells you whether to hire, bring in contractors or slow sales.
Key ratios in one table
| Metric | Common healthy range | What it shows |
|---|---|---|
| Firm-wide billable utilization | 65 to 75% | How much capacity is sold |
| Realization | 85 to 95% | Pricing and discount discipline |
| Average project margin | 30 to 50% | Pricing and delivery efficiency |
| Pipeline coverage, current quarter | 2 to 3 times the gap | How predictable revenue is |
These are rules of thumb, not standards. Your trend over time matters more than any benchmark.
Where the data lives, and how clariBI fits
Services data is usually spread across four places: a CRM for pipeline, a project tool for delivery, a timesheet for hours, and billing for cash. clariBI connects most of these directly:
- Pipeline: HubSpot, Pipedrive, Close, Copper, Insightly, Nutshell, Affinity or Folk.
- Delivery: Asana, monday.com, Jira, Linear, Notion or Airtable.
- Meetings: Calendly or Cal.com bookings, useful for "how many discovery calls turned into proposals".
- Billing: Stripe or PayPal, or invoice exports from your accounting tool.
There is no direct connector for time tracking, PSA or accounting tools such as Harvest, QuickBooks or Xero, so hours and invoices come in as exports or through a Google Sheet that you or a script keep updated. The help article on connecting Google Sheets and Drive covers that.
Joining CRM, project and billing data works when they share something to join on, usually the client name. Keep client names identical across tools ("Acme Ltd" in one and "ACME" in another will not match). The post on multi-source dashboards explains how joins across sources work.
With the sources connected, partners and project leads can ask questions such as "Which projects are past their due date with open tasks?" or "Revenue per client this year, next to hours logged?" and get a chart and table computed from the synced rows. Goals per service line (for example "Q3 proposals sent: $500,000") keep targets next to the numbers, with progress recorded as the quarter goes. Client-facing public links and scheduled email reports need Professional. The professional services use case lists what to connect first, and agencies may also find proving client ROI with data useful.
Next step
Export last quarter's timesheet and calculate utilization by person using the formula above. If the spread between your highest and lowest team is more than 20 points, that imbalance is the first thing to look at, before any new tooling.
