This series has made one argument in five different ways.
A freight forwarder is really an unpaid bank. The timing of its five payment clocks was never designed. Growth makes the gap worse rather than better. Most of the standard advice assumes a leverage that small companies do not have. And the haulage decision is really a risk decision that nobody prices.
Every one of those problems has the same property, and it is the property that lets them survive for decades in company after company.
None of them appear on the profit and loss account.
The P&L is not lying. It is answering a different question. It tells you what you earned and what you spent. It has no column for how long your money was somewhere else, and duration is the whole of this problem.
So this final essay is the practical one. Seven numbers. None of them come out of your accounting system without some work, all of them are calculable from records you already keep, and together they make the invisible thing visible. As throughout this series, any figure is invented and rounded.
One: the cash conversion cycle
The number of days between your first naira leaving and the client's naira arriving.
This is the headline. If you only ever calculate one thing from this series, calculate this.
Measure it per file and then average it per client. You will find the spread between clients is enormous, far wider than the spread in their margins, and that the ranking of clients by this number bears very little relationship to the ranking by revenue.
A reasonable target for a well run import file is under forty five days. Many forwarders discover theirs is over seventy and have never known it, because no report in the building was capable of showing it.
Two: cash fronted per naira of margin
Divide the money you put out on a job by the margin the job earned.
In the example I used at the start of this series, that ratio was about thirty to one. Yours will vary with the duty rate and the value of the goods.
What makes this number useful is that it turns an abstract worry into a comparison. A job at ten to one and a job at fifty to one might have identical margins and identical invoices, and they are completely different propositions for the business. One of them you can do twenty of. The other you can do four of.
This is the number that tells you how much growth your balance sheet can actually carry, which is the question the third essay in this series was really about.
Three: peak exposure per client
Not what a client owes you today. The most you have ever had out on them at one moment.
Almost every forwarder can tell you their debtor balance. Almost none can tell you their peak, and the peak is the number that determines whether a late payment is inconvenient or fatal.
The peak is usually two or three files overlapping, which means it is considerably larger than any single invoice and considerably larger than most people's intuition. Work it out for your three largest clients. Then ask the only question that matters: if that client paid ninety days late, would we survive it?
Set a limit and write it down while you are calm. A limit decided during a crisis is not a limit, it is a negotiation.
Four: days from delivery to invoice issued
The gap between finishing the job and being able to ask for the money.
I have put this fourth but it might be the one to fix first, because it is the only number on this list that is entirely within your own control and costs nothing to improve.
Where terms run from the invoice date rather than the delivery date, every day you take to invoice is free credit you have handed over without being asked for it. Five days is common. Ten is not rare. On a business doing volume, that is weeks of funding donated annually for no commercial reason at all.
Target one day. Put the number on the weekly report. It will fall the moment somebody is watching it, which tells you something about why it was high.
Five: invoice rejection rate, with reasons
How many invoices come back, and specifically why.
Log every rejection with a reason code for one quarter. The value is not the total, it is the split, because the split separates two completely different problems that currently look identical from the inside.
Some rejections will be your own errors: the wrong reference, a missing document, a template that has not been updated. That is a defect rate, it is yours, and it is fixable in an afternoon once you can see the pattern. You will usually find three causes account for most of it.
The rest will be requirements nobody told you about. Those are not a defect, they are a gap in your knowledge of the client's system, and once documented they stop costing you anything at all.
Without the reason codes, both of these feel like the same vague frustration and neither gets fixed.
Six: financial charge earned against financial charge invoiced
What your own policy says you should charge for advancing money, against what actually appeared on invoices.
Many forwarders already have a disbursement or financial charge in their terms. A good number of them apply it inconsistently, and some have contractual charges that are reserved in the quotation and then never billed at all.
That gap is the purest money on this entire list. No new client, no new cargo, no new negotiation, no new cost. Just invoicing something already agreed.
Target one hundred percent. If it is not close, you have found the cheapest improvement available to your business this year, and it is sitting in the difference between what sales agreed and what finance billed. Between departments again.
Seven: client concentration in receivables
What share of the money you are owed sits with your largest client.
Above about thirty percent, a single late payer can stop the company. That is not a prediction, it is arithmetic.
This number is worth watching over time rather than in a single reading, because concentration usually creeps. You win a big account, you serve it well, it gives you more work, and two years later most of your exposure is with one organisation and nobody ever made a decision to let that happen.
The one sheet that changes the conversation
Seven numbers is a programme, and programmes are easy to postpone. So here is the single exercise to do first, and it takes an afternoon.
Take last month's completed files. For each one, write down three things: how much of your own money went out, how many days it was out, and what the job earned.
Multiply the money by the days. Apply whatever your borrowing actually costs you. Subtract the result from the margin.
You will find files that lost money after funding costs. They will not be the ones anybody suspected. Several will be jobs the company was pleased to win, and at least one will be from a client everybody considers a good account.
That single sheet of paper is the most persuasive document you can put in front of a managing director, and the reason is that it does not argue with anybody. It does not blame sales for the terms, or operations for the delay, or finance for the chasing. It does not propose a strategy. It just shows, quietly and in the company's own numbers, that some of what has been recorded as revenue was actually lending, at a rate of interest below zero.
Nobody argues with that sheet. They ask what to do about it, which is a much better conversation to be in.
The thread through all six essays
If there is one idea I would like to leave behind from this series, it is not any of the seven numbers.
It is that in this industry the money is almost never lost inside a department. It is lost in the space between two of them.
The payment term agreed by sales and paid for by operations. The classification chosen by operations and never priced by commercial. The free time negotiated by one party and consumed by another. The haulage rate beaten down in one office while the demurrage it caused lands in a different one. The charge reserved in a quotation and never carried onto an invoice.
Every one of those sits in a gap. Every gap belongs to nobody. And nobody's problems are the ones that survive longest, because they never appear on anybody's report.
You do not need a new system to find them. You need to start asking, about any number that matters, who owns the distance between the two ends of it.
That question has paid for itself more times than any tool I have ever built.