Chasing a late payment without wrecking the relationship

The instinct when an invoice goes past its due date is to read it as a signal about the customer. Usually it is not. In a business of any size, an invoice is paid when it reaches a payment run, and it reaches a payment run when somebody enters it, someone approves it, and a date comes around. Any of those steps can stall for reasons that have nothing to do with willingness to pay: the invoice went to a person who has left, it is sitting in an approval queue behind a query, the purchase order number is missing so it cannot be matched, or it arrived two days after the cut-off for this month's run.

This matters because it determines the tone that works. A first follow-up written as though non-payment were a choice invites a defensive reply and puts the relationship on a footing you will regret if the cause turns out to be a wrong email address. A first follow-up written as though something has probably gone wrong administratively is both more likely to be accurate and more likely to get the invoice moving, because it makes it easy for the person reading it to help.

The deliberate non-payer exists, and the timeline below reaches them eventually. Starting from the assumption that this is who you are dealing with is expensive when it is wrong, which is most of the time.

A schedule works better than judgement exercised message by message, for the same reason a checklist works better than memory: it removes the decision about whether today is too early, which is the decision that causes chasing to be skipped entirely.

Before the due date, send a short reminder that the invoice falls due shortly. This is the highest-value message in the sequence and the one most often omitted, because it is the only one that can prevent lateness rather than react to it. It also surfaces problems while there is still time to fix them: a missing purchase order number found three days early is an administrative correction, and found three weeks late it is an argument.

Around three days after the due date, a short enquiry. Assume an administrative cause and ask directly whether the invoice has been received and entered. At roughly two weeks, a firmer message: state the amount, the number of days overdue, and ask for a specific payment date rather than an assurance. A commitment to a date is checkable and an assurance is not. At around thirty days, escalate, meaning the conversation moves to someone senior on both sides, and the message states what will happen next. The exact intervals matter less than that they are fixed in advance and actually followed.

Every message in the sequence should contain enough for the recipient to act without replying to ask a question. That means the invoice number, the invoice date, the amount, the due date, and the invoice attached again. Attaching it again each time feels redundant to the sender and removes the most common reason for delay, which is that the recipient cannot find it and will get to it later.

What should change across the sequence is the ask, not the volume of words. The pre-due reminder asks for confirmation that it is in hand. The three-day enquiry asks whether it has been received and entered, which is a question about a process rather than an accusation. The two-week message asks for a payment date. The escalation states a consequence.

Two things should never appear. The first is a threat you have not decided to carry out, because a deadline that passes without the stated consequence teaches the reader that your deadlines are decorative, and that lesson is difficult to unteach. The second is an implication of bad faith before you have established any, since it is unrecoverable in a way that a firm request is not. Firmness comes from specificity: a named amount, a named number of days, and a request for a named date is firmer than any adjective.

Chasing is not free. It consumes the time of whoever writes the messages, and it consumes the relationship at a rate that increases with each round. At some point the expected recovery falls below the cost of pursuit, and continuing is a decision made by inertia rather than by arithmetic.

The inputs to that judgement are the size of the amount, the age of it, what you know about the customer's situation, and whether the relationship has future value. A small sum from a long-standing customer going through a visible cash squeeze is a different case from the same sum owed by someone who has stopped responding entirely, and the second is closer to hopeless despite looking more like a case worth fighting. Age is the most predictive single factor available to you: the older a receivable, the lower the proportion that is generally recovered, and any collector will tell you the same. Precisely how much lower depends on your customers and your sector, and it is a number you can only get from your own ledger by looking at what happened to the last two years of overdue invoices.

Stopping is a decision, and it should be recorded as one. Writing an amount off is not an admission of failure; it is the removal of a figure that has stopped being an asset from a report you use to make decisions.

Beyond a firm request, the routes available depend on facts about the debtor and the contract, and they get costly quickly.

A formal demand from a lawyer costs a fee and sometimes works purely by signalling that the matter has changed character. Interest on delayed payment may be contractually available if your terms provide for it and are part of the agreement, and terms printed on an invoice nobody accepted are weaker than terms in a signed contract. Where both parties are registered enterprises there is a statutory framework around delayed payments to micro and small enterprises, including a facilitation council mechanism, and whether it applies turns on the classification of the supplier and the specifics of the transaction. That is a question for a professional who can see the documents, not one an article can answer.

Suing for a commercial debt is available and is usually disproportionate for small sums once the time and cost of pursuing it are counted honestly. The realistic value of understanding these routes is mostly upstream: knowing what a contract would have given you is the argument for writing terms into the next one. The most effective collection work happens before the sale, in agreeing payment terms explicitly, taking a deposit where the risk warrants it, and setting a limit on how much any single customer can owe at one time.

A receivables ledger will tell you exactly what is owed, by whom, and for how long. That is genuinely valuable and it is where any collection effort should start, because chasing from memory means chasing the customer you spoke to most recently rather than the one who owes the most.

What it cannot tell you is whether an amount will be paid. Nothing in a ledger distinguishes an invoice that is thirty days late because of a slow approval chain from one that is thirty days late because the customer has run out of money. Those look identical in a report and require opposite responses. The information that separates them comes from the conversation, and it lives in someone's head unless it is written down against the record.

So the useful discipline is to record not just the amount and the age but what was said and what was promised. A note reading customer committed to pay on the twelfth is a fact you can check on the thirteenth. Without it, each follow-up restarts from zero, the customer is asked the same question repeatedly, and the sequence loses the one thing that gives it force, which is the demonstration that you are keeping count. Automating reminders can send the messages on time; it cannot judge which silence is a process delay and which is a warning, and treating a schedule as a substitute for that judgement is how a good customer gets a fifth automated notice during a genuine crisis.