An agency owner told me his firm had grown 60 percent year over year and he was exhausted and could not figure out why cash was tight. We spent two hours in his numbers. Three of his eleven accounts were consuming 55 percent of delivery hours and producing 19 percent of revenue.
He had never measured hours by account. He had a revenue dashboard and no profitability dashboard, which is the single most common condition in this industry.
These are the six numbers I check, what each one tells you, and the threshold where I get concerned.
One: gross margin per account
Revenue from an account minus the fully loaded cost of the people delivering it, divided by revenue.
Fully loaded means salary plus taxes plus benefits plus software plus a share of overhead, not the hourly rate you imagine someone costs. In practice that is usually between 1.25 and 1.4 times base salary.
Healthy for a service business is 50 to 60 percent gross margin per account. Below 40 percent and the account is subsidized by your other work. Below 30 percent and you are paying for the privilege of doing it.
The reason this is number one: it is the only metric that identifies which specific relationship is the problem. Aggregate margin hides everything, and the average of one great account and one disastrous one looks fine.
To calculate it you need hours by account, which means time tracking. Every agency owner hates this and every agency owner who does it makes better decisions. You do not need six minute increments. You need a weekly estimate per person per account, which takes ninety seconds and is accurate enough.
Two: utilization, with a target well under 100
Billable hours divided by available hours, per person.
The mistake is targeting high utilization. At 90 percent, people have no capacity to think, no time to improve process, no room for the account that suddenly needs attention. Quality degrades and you cannot see it in the numbers until clients leave.
My target range is 65 to 75 percent for delivery staff. The remaining quarter goes to internal work, learning, and slack. Agencies that run at 85 percent and above look efficient for two quarters and then have a turnover problem, which resets you further than the extra billing ever gained.
If utilization is under 55 percent, you have a sales problem, not a delivery problem, and hiring is the wrong response.
Three: revenue concentration
Percentage of revenue from your largest client, and from your top three.
My rule: no single client above 20 percent, top three below 50 percent. Above those lines you are not running an agency, you are an outsourced department with an existential dependency.
The reason this is a profitability metric and not just a risk metric: concentration destroys your pricing power. When a client is 35 percent of revenue you cannot raise their fee, cannot enforce scope, and cannot end the relationship, so their margin drifts down every year while you absorb it. Everything I wrote about firing a client is impossible to act on if that client is a third of your revenue.
Four: effective hourly rate
Total revenue divided by total hours worked, including non billable time, including yours.
This is the single most honest number in an agency and almost nobody calculates it, because it is frequently humbling. An agency doing 900,000 in revenue with eight people working real hours is producing something like 55 dollars an hour of enterprise value against fully loaded costs that may be similar.
Track it monthly and watch the direction. Rising means your pricing, scope discipline, or process is improving. Falling while revenue grows means you are buying growth with hours, which is the most common way profitable agencies become unprofitable ones.
It is also the fastest way to compare service lines. When I have run this, the results have consistently surprised me about which work is actually worth doing, and it usually argues for narrower positioning rather than broader.
Five: cash conversion cycle
Days between paying for delivery and getting paid for it.
Agencies die of cash, not profit. You can be profitable on paper and insolvent in practice if you pay salaries on the first and collect on net 60 from clients who treat that as a suggestion.
Three levers, in order of effectiveness. Bill in advance, monthly, for retainers. Take a deposit of 40 to 50 percent on projects. Enforce a work stoppage clause, in writing, on overdue invoices, and actually use it once so everyone knows you mean it.
The best change I ever made to agency cash flow was moving every retainer to invoiced on the first, due on the first, before work. Half of clients did not blink. The ones who objected were mostly the ones already paying late. The full argument is in cash flow for small business.
Six: pipeline coverage
Value of qualified opportunities divided by the revenue you need to replace or add in the next quarter.
Agencies experience churn constantly, even healthy ones, because clients get acquired, budgets shift, and champions leave. If your coverage ratio is 1 to 1 you are already behind, because not everything closes.
I want three to four times coverage. Below two, stop optimizing delivery and go sell, because in ninety days it will be a crisis with fewer options. The methods I trust for filling this without ad spend are in client acquisition without ads.
What I deliberately do not track
Billable hours as a performance metric for individuals. It rewards slowness and punishes efficiency. Track it at the account level for margin purposes and never in a performance review.
Vanity revenue milestones. Hitting a number with 22 percent margins is worse than being smaller with 55.
Headcount as a proxy for success. Every agency owner who has told me their team size before their margin has, in my experience, had a margin problem.
Where pricing model interacts with all six
These metrics move together depending on how you charge. Hourly billing caps effective rate and punishes efficiency. Fixed project pricing exposes you to scope risk but rewards process improvement. Retainers stabilize cash conversion but tend to accumulate scope creep unless actively managed. Performance pricing can produce excellent margins and terrible predictability.
I go through the tradeoffs in detail in retainer versus project versus performance pricing, but the short version: your pricing model determines which of these six numbers you have to watch most closely. If you cannot say which one, you have not chosen a model deliberately.
For performance based arrangements specifically, get comfortable modeling the economics before you sign. A ROAS calculator will tell you in five minutes whether the deal you are being offered can produce a fee worth having at realistic performance levels, and I have talked myself out of two bad deals that way.
Build the one page, then look at it every month
Six numbers. Gross margin per account, utilization, concentration, effective hourly rate, cash conversion, pipeline coverage.
Put them on one page. Update on the first Monday of every month. It takes about an hour once the time tracking habit exists.
The value is not any single number. It is that you will spot the direction three to six months before it becomes a crisis, and at that distance every problem is solvable with a conversation. At the point where cash is tight and a big client is leaving, your options have narrowed to bad ones. The agency owner I opened with is fine now, and the only thing that changed was that he started measuring hours by account. That was the whole intervention.