Define the seven numbers once
The problem
“I don’t know where I make or lose money.”
Most NEMT companies have exactly one scoreboard, and it arrives from the accountant weeks after the month it describes has closed. A P&L is accurate, and it is the right tool for taxes. It is the wrong tool for deciding this week whether a contract or a rate is still worth running.
A scoreboard is seven numbers, accurate to the week rather than to the cent, on things you can still change. The usual failure is that nobody can reproduce them: two people say profit per trip and mean different things, and the review turns into an argument about the formula. Write each one down once, with its trap named:
| Number | What it is | The trap |
|---|---|---|
| Profit per trip | Revenue minus direct cost, per completed trip | Direct means driver wage, fuel, and the per-trip cost of the vehicle. Leave dispatch salary and rent out of it, or you are looking at net margin twice. |
| Profit per mile | The same profit, divided by miles actually driven | Driven miles, not billed miles. The gap between the two is the deadhead nobody is paying you for. |
| Monthly margin | The trend line, everything in | The only one of the seven allowed to be slow. Read it for direction across months, not as a verdict on one. |
| On-time performance | Completed trips inside the pickup window | Use the window your contracts hold you to, not your own. Theirs is the one that pays. |
| Deadhead percentage | Empty miles as a share of total miles | Count the run back to the yard. Operators who exclude it report a number their fuel bill disagrees with. |
| Days to payment | Trip completed to cash in the bank | Measure from completion, not from claim submission. The days between those two are the ones you control. |
| Driver turnover | Rolling twelve months: separations over average headcount | The leading indicator. It moves before on-time performance does, which is what makes it worth watching while it is still boring. |
None of these come with a target. Targets vary by market, fleet and contract mix. The weekly review reads direction, and direction only means something once the definition holds still.
Count your weeks of cash
The problem
“I don’t know if I’ll make payroll in 3 weeks.”
“Volume drops and I find out a month later in the bank.”
The seven numbers tell you whether the operation works. None of them tells you whether you can make payroll a month from now, and a company can fail that test while all seven look fine.
Profit and cash part ways in three places: loan and lease payments, what the owner takes out, and money you are owed but have not received. An account that pays a month after the trip is holding cash you have already earned.
So keep one more line beside the seven: cash in the bank, thirteen weeks forward. Start from this week’s balance. Add what you expect to collect each week from each payer, using how long each one actually takes to pay. Subtract payroll, fuel, insurance, loan payments and your own pay.
Cash in the bank, thirteen weeks forward
An example shape, not real figures.
If the line drops below zero anywhere in those thirteen weeks, that is the problem to work on now, while there are still weeks to solve it.
Put trips per week by payer on the same page. Payment arrives weeks after the trip, so a drop in one payer’s volume reaches your bank weeks late. Trips per week shows it the week it happens. When a payer’s trips fall two weeks in a row, someone calls them before the review ends.
The thirty-minute review
The problem
“My tech is a mess: dispatch, billing and the books don’t talk.”
Same day, same time, same three people, one page, half an hour. Friday morning works well for many companies, because the week is done and the next one is not built yet.
Read direction, not level: better or worse than last week, for each number. Then pick the one that moved the wrong way, put a name against it, and agree the next step. One number, one owner, one action.
Reviews rarely die of indiscipline. They die of assembly. If producing the numbers takes most of a day in spreadsheets, the review gets skipped in the first busy week and never comes back. So write down where each number comes from and who produces it, then time the slowest one. Fixing that source is worth more than anything else on this page.
Two more habits kill it: producing numbers inside the meeting instead of reading them, and moving the day.