Walk into most service businesses doing $500K to $3M and you'll find a founder who tracks everything and acts on almost nothing. Revenue dashboards. Website traffic. Instagram followers. Time-tracked hours by project. Twenty metrics deep, three inches of insight.
The problem isn't that these founders don't care about data. It's that nobody told them which numbers actually predict growth and which ones just feel productive to check. So they end up watching the wrong scoreboard while the game moves on without them.
Here are the five KPIs that reliably move a service business forward, how to calculate each one, and the vanity metrics you can stop tracking this week.
Why most service business dashboards are broken
A KPI is only useful if it changes what you do next week. That's the test. If your revenue chart went up and you'd do nothing differently, or went down and you still wouldn't know where to intervene, it isn't a KPI. It's decoration.
Service businesses have a specific problem here. Unlike SaaS or e-commerce, where funnel math is clean, service revenue is lumpy, project-based, and tied to human capacity. That means a lot of standard business metrics you read about online don't apply cleanly. Monthly recurring revenue looks great on a chart, but if you sell one-off engagements, it tells you nothing.
The KPIs below are chosen because they work for service businesses specifically. They connect cause and effect. When one moves, you know where to push.
KPI 1: Gross margin per delivery hour
This is the single most underused number in service businesses. It answers: for every hour your team spends delivering work, how much gross profit do we keep?
The formula is simple. Take your gross profit for the month (revenue minus direct delivery costs, including labor). Divide by total hours your team spent on client work. That's your gross margin per delivery hour.
Why this matters more than revenue: two agencies can both do $150K a month. One is drowning at 60 percent utilization with a $45 margin per hour. The other is calm at 45 percent utilization with a $110 margin per hour. Same revenue. Radically different businesses.
Track this monthly. If it's dropping, one of three things is happening: your scope is bloating, your pricing is stale, or you're using expensive labor for cheap tasks. Each has a different fix. Scope bloat means better project boundaries. Stale pricing means a rate increase on new clients. Expensive labor on cheap tasks means you need to move that work down to a lower-cost seat.
That last point is where most founders leak the most money. A $140 an hour senior editor doing thumbnail exports. A $90 an hour operations manager scheduling meetings. Every hour of that mismatch is money you already earned and then set on fire.
KPI 2: Sales cycle length by lead source
Most founders track leads and closed deals. Very few track how long it takes a lead to become a deal, broken out by where the lead came from.
This one number changes your marketing budget. If referrals close in 11 days and cold outbound closes in 74 days, and both convert at similar rates, referrals are worth roughly seven times more per lead in cash flow terms. Not because they close at a higher rate, but because your money comes back faster and you can reinvest it.
To calculate: for every deal closed in the last 90 days, log the date of first contact, the date of close, and the source. Average the days by source. You now have a real picture of which channels are actually funding your business versus which ones look busy.
The action this drives is uncomfortable but important. Most founders discover their favorite channel (the one they enjoy working on) isn't their most efficient. LinkedIn content might feel productive. If it produces leads that take 90 days to close while your partner referrals close in two weeks, you know where to spend Monday morning.
KPI 3: Client concentration risk
What percentage of your revenue comes from your top three clients? If it's above 40 percent, you don't own a business. You own a job with extra steps.
Track this monthly. Calculate top client as a percentage of trailing 90-day revenue. Then top three. Then top five. Watch the trend.
Concentration risk is the metric that quietly kills growing service businesses. A founder gets a big anchor client, revenue jumps, they hire aggressively, and 18 months later that client changes strategy or gets acquired. Suddenly 55 percent of revenue is gone and payroll is due Friday.
The fix isn't to fire big clients. Big clients are great. The fix is to make sure your business development pipeline is producing new logos fast enough to keep the top client under a threshold you can survive. Most operators set that at 25 to 30 percent for the top account.
If you're above the threshold, the next 90 days should be spent adding smaller accounts, not chasing more work from the anchor. That feels counterintuitive because the anchor wants to spend more. Take the extra work if you must, but count it as risk, not growth.
KPI 4: Delivery capacity utilization
Utilization is billed or client-facing hours divided by available working hours. For most service businesses, healthy utilization sits between 65 and 80 percent for delivery staff. Below 65, you're overstaffed or underselling. Above 80, you're burning people out and quality is about to drop.
Here's what most founders miss. Utilization isn't just a staffing metric. It's a leading indicator of everything else. When utilization creeps above 85 percent for two months, you'll see it in the third month as missed deadlines, client complaints, and a resignation you didn't see coming.
Track utilization per person, weekly. Not as a surveillance tool, as an early warning system. If your best account manager has been at 92 percent for six weeks, don't wait for them to break. Move work off their plate now.
This is also the metric that tells you when to hire. The mistake most operators make is hiring reactively, after a bad month of burnout and missed work. By then you're behind. A better rule: when a role hits 80 percent utilization for four consecutive weeks, start recruiting for a second seat. The lead time on getting someone hired, onboarded, and productive is usually 60 to 90 days. Start when the signal appears, not when the fire is already going.
KPI 5: Revenue per full-time employee
This is your efficiency ceiling. Total revenue divided by total headcount, including yourself and any full-time contractors. Track it quarterly.
Healthy service businesses land somewhere between $180K and $350K in revenue per employee, depending on the vertical. Below $150K, something structural is wrong. Above $400K, you're likely running lean in a way that's hard to sustain without breaking someone.
The reason this metric matters: it's the number that determines whether growth actually improves your life. You can double revenue and halve your revenue per employee, which means you now have twice the headaches for the same profit. Or you can grow revenue 40 percent while holding headcount flat, and your margins expand meaningfully.
The path to improving this number isn't working people harder. It's structural. Move low-value tasks off high-cost seats. Systematize the repetitive parts of delivery. Use lower-cost roles for coordination, admin, research, and first-draft work so your expensive seats spend more time on the parts clients actually pay for.
This is where most operators we work with find their next 20 points of margin. Not by cutting people, but by making sure every seat is doing work that matches its cost.
The vanity metrics you can stop tracking this week
Now the fun part. Here's what to take off your dashboard.
- Total leads. Meaningless without source, close rate, and cycle length. A thousand bad leads is worse than ten good ones because bad leads consume sales time.
- Website traffic. Unless you sell to a wide audience through content, raw traffic tells you nothing. Track qualified inbound inquiries instead.
- Social media followers. If your last ten clients didn't come from social, follower count is a hobby, not a metric.
- Hours worked. Founders wear this like a badge. Nobody pays you for hours. They pay you for outcomes. Track output.
- Proposals sent. Proposal volume without win rate and cycle length is just activity theater.
- Gross revenue in isolation. Revenue without margin context is the most misleading number in business. You can grow revenue while going broke.
None of these are useless in every context. But if you're tracking them without connecting them to a decision you'll actually make, they're taking up space where a real KPI should sit.
Building your five-number scorecard
Take one hour this week. Open a spreadsheet. Put these five numbers at the top:
- Gross margin per delivery hour
- Sales cycle length by lead source
- Client concentration (top account as percent of revenue)
- Delivery capacity utilization
- Revenue per full-time employee
Fill in this month's numbers. Then last month's. Then the same month last year if you have the data. You now have a scorecard that tells you where the business actually is, not where it feels like it is.
Review it on the first Monday of every month. Thirty minutes. Ask one question for each metric: what would I do differently if this number moved 20 percent in the wrong direction? If you can't answer, the metric isn't earning its spot.
The businesses that compound over five and ten years aren't the ones with the most sophisticated dashboards. They're the ones where the founder knows five numbers cold, checks them on a rhythm, and adjusts. That's the whole game. Everything else is noise dressed up as diligence.