Agency KPIs: the few numbers a $2M+ agency owner should watch

Agency owners I advise have told me they're running a million-dollar company on 7% profit. Revenue was up, the team was full, and the owner couldn't say where the money went.

This post is about agency KPIs for the owner: a short scorecard for the growth and profit of your own agency. It covers margin, pipeline, win rate, and how much of your revenue rides on one client. Reporting results to clients (campaign dashboards, ad metrics) is a different job, and none of it is here.

You get five numbers. Each one comes with how to calculate it from your own books or CRM, how often to review it, and what a bad reading should make you do.

What is net profit margin, and what should a bad one make you do?

Net profit margin is what's left of revenue after every cost of running the agency, divided by revenue. Pull the trailing twelve months from your accounting system, subtract salaries, contractors, software, and rent, and divide what remains by total revenue. Pay yourself a market salary before you call anything profit. Otherwise the margin includes wages for a job you're doing.

Review it monthly on the trailing twelve months.

Revenue growth hides this figure better than any other. An agency can add a million dollars in revenue and watch profit stay flat, because each new person arrived before the price went up. I wrote about how that plays out in agency profit margin, including why margins often slip as an agency adds headcount.

A falling margin sends you to two places in this order: the price on your last three projects against what each cost to deliver, and how many change orders you wrote. If projects keep running over, fix the scoping before you touch hiring.

How much pipeline do you need to hit next quarter's number?

Pipeline coverage is the value of your qualified opportunities divided by the revenue you still need to close. If you need $500K and have $1.5M of qualified deals open, coverage is 3:1. In my guide to marketing KPIs, I say that anything under 2:1 usually means you miss your revenue target. The related measures are in marketing KPIs for professional services firms.

The calculation depends on your definition of qualified, so settle it once. A deal belongs in the coverage figure when a buyer with budget has described a problem and asked for a proposal. A name on a list stays out.

Review it every week. Pipeline is the one number here that warns you early, because it moves months before revenue does.

When coverage drops, spend your next week of owner time on outreach and conversations with past clients, ahead of anything in delivery. Hiring and discounting come after that.

What is your win rate, and where do your wins come from?

Win rate is proposals won divided by proposals decided, over a stretch long enough to cover a full sales cycle. Pull it from the CRM and split it by lead source: referrals, inbound content, outbound, events, and paid. The blended number tells you little. The split tells you where to put your hours.

Review it after every proposal decision, and review the split each quarter.

A low win rate on one source means you're spending time on the wrong buyers there. A low rate across the board means you're pitching deals you shouldn't. That second case is a qualification problem, and how to decide whether to bid at all is where I'd start.

A bad reading should make you pause proposals for your weakest category of lead and watch what happens to the rate.

How much of your revenue rides on your biggest client?

Concentration is your largest client's revenue over the trailing twelve months, divided by total revenue. Run the same math for your top three combined. It takes one query in your accounting system.

Review it monthly. A client grows into that share slowly, and then a single budget decision on their side changes it overnight.

It matters twice. If you lose that client, the revenue goes, and so does the margin, because your team is sized for the work. There's also a quieter cost, which I've felt myself. At Foundry, the agency I co-founded, I had to stay directly involved in whoever our biggest client was until we reached about 15 people. Big clients ask for the owner, and the owner says yes.

A high reading sends you to two jobs: win the next client of that size before you need it, and put that client's renewal date on your calendar now. A price increase gets harder to say once one client carries that much of the agency.

What is your effective hourly rate?

Effective hourly rate is project revenue divided by the total hours your team spent on it, whether billable or not. Pull hours from your time tracker and revenue from your invoices, project by project.

Review it when each project closes, against what you quoted.

It shows what the project paid you per hour once the revisions, the extra calls, and the unplanned work are included. Two projects with the same price can land very differently. This number also exposes hourly billing: as your team gets faster with new tools, the invoice shrinks if you're charging for hours.

A bad reading should change how you price the next project of that type. If a service keeps coming in low, write the change-order terms into the contract at the start. Agency growth levers covers where pricing sits among the other ways to lift revenue and protect margin.

How do you put the five numbers together?

Build a single table with five rows. Each row holds the current number, last quarter's number, and the action you've committed to if it's off. Keep the definitions in the same document so the next person who calculates a number gets the same answer.

One person updates the table, and at a $2M+ agency that is rarely the founder for long. Give the updating to someone else once the definitions are stable, and keep the review for yourself.

Begin with whichever number you can't answer today. It's usually the one you've been avoiding.

Key Takeaways

• The owner's scorecard has five numbers: net profit margin, pipeline coverage, win rate, biggest-client share of revenue, and effective hourly rate.
• Each comes from your accounting system, CRM, or time tracker. A new tool is unnecessary.
• Review pipeline weekly, review margin monthly, review concentration monthly, review win rate after each proposal decision, and review effective hourly rate when a project closes.
• Pick the action for a bad reading before you see the reading.