Contractor Finance KPIs Ranked by Profit Impact (2026)

Most contractors run their business on two numbers: revenue and jobs booked. Both feel like progress. Neither tells you whether you are actually making money. A more useful approach is a smaller set of numbers that tie directly to profit and cash, reviewed in order of profit leverage. The order below is editorial judgment informed by the Level Index benchmarks cited in each section, not a measured ranking of owner outcomes.

Here are the eight, in the priority order we use.

1. Gross profit per billable hour

Why it ranks first: skilled labor, not demand, is the constraint for most trades. So the truest measure of the business is how much gross profit each field hour produces. It normalizes for job size and mix in a way that revenue and margin percentage do not. A shop can grow revenue and still make less per hour worked.

How to compute it: (revenue minus direct job costs, including loaded technician labor) divided by field labor hours. Pick one denominator and keep it. Billable hours (hours on jobs) give a higher figure than total paid field hours, which include drive, admin and idle time. Do not add technician cost back or recover it separately, because it is already inside direct job costs. Track it weekly by job type against your own trailing average and target. The Level Index does not publish a trade median for this metric. A persistent decline against your own baseline, not a trade comparison, is the signal that you may be busy but underpriced for the hours worked. Run yours in the gross profit per hour calculator.

2. Job-level gross margin

Benchmark: median job gross margin is 44.3% across 1,747,089 completed jobs from 1,791 companies (The Level Index, measured per job across all job types; 6.2% of those jobs had negative margin). Use it only after labor, materials, and subcontractor costs are attached to the same job as revenue.

Why it ranks this high: you cannot price the next job correctly if you never knew the true margin on the last one. The profitable jobs quietly subsidize the losers, the company-level P&L looks fine, and the individual losers stay invisible until cash gets tight. In our judgment, fixing job costing is often the highest-leverage operational change available. See how revenue-only job reports create phantom margin.

3. Collection rate

Benchmark: median collection rate is 85.1% (n=464 contractors, defined as cash collected as a share of billed revenue). Top decile reaches 96.0%; the bottom decile collapses near 39%. The published definition does not state the measurement window, so a low rate can mix timing (invoices still open at period end) with credits, disputes and write-offs.

Why it ranks this high: the gap between what you bill and what you collect is work already performed, but not all of it is recoverable cash. Fictional example: a contractor bills $10.0M in a year and collects $8.5M against those invoices (85%). At 96% it would collect $9.6M, a $1.1M difference. The full billed-but-uncollected amount is $1.5M; the 96% comparison still leaves $400K uncollected. For this illustration only, suppose the remaining $1.1M comparison gap breaks down as $700K still open in AR at year end, $250K of credits and disputed amounts, and $150K written off. Collecting the $700K faster is a one-time release of cash tied up in receivables, not new annual profit. Preventing the $150K of write-offs recovers cash on revenue already booked, and it recurs only if the root cause is fixed. The $250K of credits may point to pricing, scope or documentation problems rather than collection effort. Split your own gap the same way before estimating what is recoverable. Cash collected is not accrued revenue, so track this alongside AR aging, not in place of it. See the full DSO and collections playbook.

4. Billing speed

Benchmark (n=733 total): median billing speed, measured as days from job completion to invoice issued, is 1 day when progress billing is included, but among post-completion invoicers the median is 7 days; across all 733 companies the slowest decile waits 30+ days.

Why it matters: every day between finishing the work and sending the invoice is an interest-free loan to a customer who did not ask for one, and older invoices are commonly harder to collect (treat this as operating judgment; the Level Index does not measure that relationship). In our judgment, billing speed is one of the cheapest levers on DSO because it mostly costs discipline. A DSO reduction is a one-time change in the receivable balance and the timing of cash, not a recurring profit gain equal to the cash released.

5. Service-agreement gross margin

Benchmark: median service-agreement gross margin is 37.9% (n=259 companies with at least $10K in service-agreement revenue); the top quartile clears 53.5%. The bottom decile runs negative (the 10th percentile is -30.4%).

Why it matters: service agreements can be a high-leverage line item in a service contractor's book, not because they are the biggest revenue line, but because the spread between doing them well and badly is wide. In the Level Index, the 10th to 90th percentile range runs from -30.4% to 70.3% gross margin, about 100 points. A book priced years ago without cost adjustments can quietly run at a loss while looking like healthy recurring revenue. See service-agreement profitability.

6. Change-order capture rate

Benchmark: none measured. The Level Index does not publish a change-order capture rate, and we have no named source for a capture percentage, so this page does not give one. Measure it yourself: capture rate equals scope changes priced and approved before the work, divided by scope changes identified in the field. Count it both by number and by dollar value over the same period.

Why it matters: unpriced scope creep never appears on the P&L as a loss, it appears as nothing, so it is the most invisible margin leak in contracting. The margin effect depends on how much unpriced work you do, so calculate it rather than assume a range. Fictional example: a contractor has $2.0M of project revenue and $1.4M of direct cost, a 30.0% gross margin. Field crews performed extra work that would have billed $100K, and its cost is already inside the $1.4M. Billing it makes revenue $2.1M on the same $1.4M of cost, a 33.3% gross margin, or about 3.3 points. A shop with less unbilled extra work would see less. See change orders as a margin lever.

7. Quote conversion rate

Benchmark: median quote conversion is 73.9% on decided quotes, those where the customer actually said yes or no (n=794 companies with at least 20 decided quotes). Top quartile converts 83.2%. This is deliberately measured on decided quotes: won divided by won plus lost, excluding pending quotes. Measured on all quotes issued, including quotes that never reach a decision, a Level Index sample (n=794 companies with 20+ quotes) has a median of 38.1% (25th to 75th percentile 28.0% to 47.7%). Compare your own rate only to the matching definition.

Why it matters: in our judgment, quote speed is one of the biggest controllable levers on conversion. In the Level Index, customers accept winning quotes a median of 2 days after they are sent, while lost quotes are marked lost a median of 29 days after sending. That timing pattern does not prove that faster quoting causes higher conversion, so test it in your own data. Compare decided-quote conversion for quotes sent within 24 hours of the request against quotes sent after 7 days. Conversion sits below the margin and cash metrics because winning more of the wrong-priced work makes the top three metrics worse, not better.

8. Customer concentration

Benchmark: in the Level Index, the median share of total revenue from a company's largest single customer is 31% (25th percentile 18.5%, 75th percentile 54.6%; n=959 companies with more than $100K in annual revenue). We have no named source for a specific percentage at which buyers or lenders apply a discount. Treat any threshold as a diligence question for your own advisors, not a rule. Buyers and lenders commonly ask about concentration in diligence.

Why it matters: concentration is a risk metric more than a profit metric, which is why it ranks last, but it caps the value of everything above it. A highly profitable business with one dominant customer can be valued lower if a buyer prices in the risk of losing that customer. The effect depends on contract terms, customer tenure and the buyer. Worth watching quarterly even when nothing is wrong.

The two numbers that are not on this list

Revenue and jobs booked. Both are activity, not outcome. A contractor can grow both while gross profit per hour, job margin, and collection rate all decline, which is exactly how a shop ends up busier every year and no richer. Track the eight above, in this order, and the two activity numbers become much easier to interpret.

Related reading:

• Gross profit per hour: the complete guide
• Why your job margins aren't what your software shows
• The DSO and collections playbook
• The Level Index (all contractor benchmarks)