The Numbers an Owner Should Be Able to Recite
Why this matters
Almost every consequential decision an owner makes happens away from a report. A customer asks for a discount while you are standing in their driveway. A tech asks for a raise in the truck. A supplier offers terms on a call. A lender asks a question in a meeting. In each case the answer depends on a number, and you either carry it or you do not.
Carrying a number is not about impressing anyone. It is that a number you can recite is a number you have a feel for, and a feel for it is what lets you notice when it moves. An owner who has to open a file to know their gross margin will find out about a four-point slide the month after it stops mattering.
The test
Someone stops you in a parking lot and asks seven questions. No phone, no file. Whatever you can answer within one guess is what you are actually running the business on. Everything else is a report you read and forgot, which is not the same thing as knowledge.
The list
Seven numbers, chosen because each one governs a decision you make more than once a month, each one moves slowly enough to be memorable, and each one is one you can feel drifting.
| Number | The decision it governs | How to hold it | Refresh |
|---|---|---|---|
| Gross margin, percent of revenue | Whether to hold price, discount, or walk | One percentage, plus your target | Monthly |
| Break-even, in billable hours per week | Whether a slow week is a problem or noise | Hours per week the whole shop must sell | Quarterly |
| Billable ratio, billable hours over paid hours | Whether to hire, and whether the schedule is the problem | One percentage | Monthly |
| Average ticket, as a multiple of a standard call | Whether a job type is worth the truck roll | A multiple, not a currency figure | Monthly |
| Receivables over 60 days, percent of total owed | Whether to keep working for an account | One percentage | Monthly |
| Weeks of payroll held in cash | Whether you can absorb a bad month | A count of weeks | Weekly |
| Booked work ahead, in working days | Whether to push marketing or push capacity | A count of days | Weekly |
Two of the seven refresh weekly, and both are cash or capacity. Those are the ones that can change under you in days, which is exactly why they are the two you should be able to answer without pausing.
Why a multiple for the average ticket. Expressing it against a standard call, say a routine diagnostic visit, gives you a number that survives price changes and is instantly comparable across job types. "This job type runs about three standard calls, and it costs me four calls' worth of time" is a complete decision. A currency figure has to be re-learned every time you reprice.
Why break-even in hours rather than revenue. Hours are what you schedule. A break-even you can compare against next week's board is one you will actually use, and it converts directly into a staffing decision.
What is deliberately not on the list
This is where most lists go wrong, and the exclusions teach more than the inclusions.
Monthly net profit is not on it. It is the number owners most want to recite and the worst one to carry in your head. It lags, it is noisy at small volume, and a single timing difference can swing it enough to change your mood without changing anything about your business. You read net profit; you do not run on it.
Total revenue is not on it. It is the easiest number to remember and the most flattering, which is precisely the problem. Revenue tells you how busy you were, not whether being busy was worth doing, and an owner who carries revenue and not margin will grow their way into trouble with total confidence.
Per-technician utilization is not on it. It is a genuinely useful number and it belongs to whoever runs the schedule. The shop-level billable ratio is yours; the per-person breakdown is a management report you read when the shop-level number moves. Carrying six people's utilization in your head is how an owner ends up doing the dispatcher's job.
Anything you cannot compute without a report is not on it. If answering requires pulling a file, it fails the parking-lot test by definition, and putting it on the list only trains you to fake it.
Anything with a tax answer is not on it. What you owe, what is deductible, what your effective rate is: these are answered by the person who signs your return, and a remembered figure is exactly how an owner talks themselves into a position they should have asked about.
Carry each number as a pair
A number on its own is trivia. "Gross margin is 41%" is a fact you can state and cannot use. Carry every one of the seven as a pair: the current value and the line it is being measured against. "Margin is 41% against a 45% target." "Cash covers 4 weeks of payroll against a floor of 8." "Receivables over 60 days are 22% against a ceiling of 10%."
The pair is what converts recall into a decision in the driveway. Asked for a discount, an owner carrying only the value has to do arithmetic on the spot and will guess generously. An owner carrying the pair already knows they are four points below target and that this particular conversation is not where the four points come back.
Pairing also fixes the most common self-deception in the whole exercise, which is drifting your target toward your actual. Write the target down once, date it, and change it only as a deliberate decision with a reason attached. If you cannot remember the last time your gross margin target changed, that is a good sign. If it has moved twice this year and always downward, it is not a target, it is a description.
Two of the seven take a floor rather than a target, and it is worth being explicit about which. Weeks of payroll in cash and booked days ahead are both floors: a level you refuse to go below, not a level you aim at. Aiming at a floor guarantees you spend half your time under it. A workable starting floor for cash is 8 weeks of payroll, tuned to how seasonal your work is, since a shop with a genuinely dead quarter needs more and a shop with flat year-round demand can hold less.
How the numbers get into your head
Not by memorizing them. By being the person who says them.
Pick a fixed moment that already exists in your week and attach the recital to it. The end of the close, the start of the Monday huddle, the drive back from the last call on Friday. Say the two weekly numbers out loud at that moment, every week, whether or not anything changed. Saying an unchanged number aloud is not wasted; it is how you build the sense of what normal feels like, which is the thing that lets you notice abnormal without checking.
The four monthly numbers ride on the close, and break-even hours gets refreshed alongside them once a quarter. When you finish reading the statements, write them down from memory first, then correct them from the page. That is the recital test run at low cost, twelve times a year, and after two or three rounds the gaps shrink on their own without anybody trying to memorize anything.
The recital test, worked
Do this once a quarter. Write your seven answers down from memory, then open the statements and fill in the actuals. Score each one, and score it by type, because a percentage and a count do not drift the same way.
- Percentages score in percentage points: within 2 points is good recall, more than 2 and up to 5 is stale, more than 5 means you are not tracking it.
- Counts, hours, and multiples score in relative terms against the actual: within 10% is good, more than 10% and up to 25% is stale, more than 25% means you are not tracking it.
Here is a real-shaped result from a six-person shop.
| Number | Recited | Actual | Gap | Band |
|---|---|---|---|---|
| Gross margin | 45% | 41% | 4 points | Stale |
| Break-even, billable hours per week | 96 | 104 | 8 hours, about 7.7% | Good |
| Billable ratio | 60% | 52% | 8 points | Not tracking |
| Average ticket, multiple of a standard call | 2.5x | 2.4x | about 4.2% | Good |
| Receivables over 60 days | 10% | 22% | 12 points | Not tracking |
| Weeks of payroll in cash | 6 | 4 | 2 weeks, 50% | Not tracking |
| Booked days ahead | 9 | 8 | 1 day, 12.5% | Stale |
Two landed in the good band, two were stale, and three fell in the not-tracking band. That distribution is unremarkable for a first attempt. What is not unremarkable is the direction.
All seven misses ran optimistic. Margin higher than real, billable ratio higher, break-even lower, aged receivables lower, cash deeper, booked work longer, tickets bigger. Not one error ran the other way. That is not a coincidence and it is not dishonesty. Memory of a business updates on the good days, when a big job closes or a full week clears, and quietly declines to update on the ordinary ones. Any owner running this test should expect the same one-directional drift and should treat a recital that is uniformly optimistic as the normal, correctable result rather than as a character finding.
The three not-tracking misses cluster, and the cluster is the diagnosis. Billable ratio, aged receivables, and weeks of cash are the three numbers this owner never touches personally. They live in a timesheet summary, an aging report, and a bank view that somebody else assembles. The two he recited accurately, break-even hours and ticket multiple, are the two he uses in conversation every week. You recall what you handle. If three of your misses share a source, the fix is not memorization, it is putting yourself back in contact with that source once a month.
Reading the gaps
The size of a gap tells you something, but its shape tells you more.
A single number wildly off, the rest tight. Usually a definitional problem rather than a memory problem. You are carrying a number computed one way and the statement computes it another. Aged receivables measured from invoice date versus from due date can differ by a full band on their own. Before concluding you have lost track of it, check that you and the statement are measuring the same thing.
Everything stale by roughly the same amount, in the same direction. You are carrying a snapshot from a real moment that has since moved. That is a cadence problem: your refresh interval is longer than the numbers' drift rate. Shorten the interval on the two weekly items first.
Good recall on percentages, poor on counts. Common, and it means you think about the business in ratios and not in capacity. It shows up as chronic overcommitment in the schedule, because ratios do not tell you how many days of work you have.
Perfect recall on all seven. Worth a moment of suspicion. Either you genuinely run tight, or you are reciting your targets rather than your actuals, which feels identical from the inside. The check is simple: were any of the seven numbers you recited different from what you want them to be? If none of them were, you recited the target.
References
- U.S. Small Business Administration (SBA), small business financial management and performance measurement guidance
- Generally Accepted Accounting Principles (GAAP), definitions of gross margin and revenue recognition
- See related: Financial KPIs for a Service Business; Reading Your Profit and Loss Statement; The Weekly Cash Number to Watch