Run your eye down an aged debtor report and the work looks uniform: the same reminder, the same escalation, a different name each week. Each of those lines carries a judgement about one particular customer, and look at where late payment actually comes from and most of the variation belongs to the payers instead of the procedure.
In this article
The short version
- Among repeat customer-initiated payers in Adfin's book, 27.4% were never late once and 10.3% were late on every single payment (Adfin platform data, 1,359 business-customer pairs).
- The 24.3% who were late at least 80% of the time accounted for 51.9% of all late payments in that group (Adfin platform data).
- Those pairs qualified by making five or more payments themselves, so they're a selected group and the shape travels further than the exact proportions do.
- Pooled comparisons can measure who uses a thing instead of what the thing does. Apple Pay's 6 percent on-time advantage over card falls under 1 percent inside a single business's own book (Adfin platform data).
- Your procedure can be identical across a ledger where the customers behave nothing like each other.
Two hundred names and one procedure
Most credit control advice, ours included, is written as a sequence: confirm before the due date, remind after it, change medium, escalate, stop. That's useful, and it's how the chasing process is best explained to somebody starting out.
What the sequence hides is that you're applying it to a population you already know a lot about. You know which client queries every invoice and pays on day 32 anyway. You know which one pays the morning it arrives. You know the one where a call to the office manager works and email disappears. None of that knowledge appears in a procedure, and all of it appears in what you actually do on a Tuesday afternoon.
So the interesting question about a credit control routine isn't how well designed it is, but how much of the outcome it can move at all, given who's on the other end.
Where does the lateness come from?
Adfin can measure this on its own book. Take every business-customer pair with at least five payments the customer made themselves, by card, wallet, open banking or bank transfer, and you have 1,359 pairs covering 12,863 payments (Adfin platform data).
| Repeat payers | Share of pairs |
|---|---|
| Never late once | 27.4% |
| Late on every single payment | 10.3% |
| Late 80% of the time or more | 24.3% |
Adfin platform data, 12,863 payments across 1,359 business-customer pairs with five or more customer-initiated payments each.
That worst quarter produced 51.9% of every late payment in the group (Adfin platform data). On a two hundred client ledger that's roughly fifty names generating half your chasing work, and about fifty-five paying on time whatever you send them.
A real limit belongs with those figures. A pair only enters the cut after five payments the customer initiated, and a customer who paid promptly by card five times running is exactly the one most businesses would have moved onto a mandate, so the sample leans towards customers who resisted that. The concentration is solid inside the group; the exact proportions won't transfer to your book. The population, the definitions and the exclusions are all in how we measure UK payment behaviour, and the wider set of findings is in the UK Getting Paid Report.
What a single chase is weighing up
If a quarter of your customers never miss and a quarter miss almost always, the same third reminder is doing two different jobs. To one it's a courtesy. To the other it's the fourth month of an arrangement that hasn't worked.
Four judgements tend to be buried inside one apparently routine follow-up.
- How much of your attention this name deserves this week, given what it did the last five times.
- Whether the delay is about approval routing, a query, cash, or a habit, because those want four different responses.
- Whether the collection arrangement behind your invoice suits this customer at all, a bigger decision than any message you could send.
- What the relationship can carry, since a client you also advise is not a supplier's customer.
Not one of those is a scheduling question, and you're making all four anyway, from memory, in the gaps in your week. Writing them down is roughly what a credit control policy is for, and writing a credit control policy that gets followed covers how much to fix in advance.
Where this lands practically is per-customer rules: a shorter leash on the names with a record of missing, a longer one on the names without. How to set different chasing rules for different clients works through the mechanics, and the ranking you'd need is already sitting in your ageing. How to read an aged debtor report covers pulling it out.
A pooled difference can be a difference in who
A pooled difference between two groups can be a difference in who's in them, and that move explains a lot of what passes for insight in payments data.
Across Adfin's whole book, invoices paid by Apple Pay arrived on or before the due date about 6 percent more often than invoices paid by card. That reads like an argument for turning wallets on. Hold the business constant, though, and across 71 businesses with at least twenty payments in both arms the typical gap falls under 1 percent, with 39 doing better on Apple Pay against 32 on card (Adfin platform data). The pooled figure was largely telling you which businesses enable wallets and which of their customers reach for one.
Your own late rate is a mixture in the same way. Move your steady payers onto mandates and the rate for whoever is left paying manually gets worse while your collections improve. Adfin sees exactly that: on-time payment inside the customer-initiated channel drops about six percent over the first year of a relationship, while overdue rates at business level roughly halve over a comparable window (Adfin platform data). Both are right, and both are measured on populations shifting underneath them.
So your dashboard can move for reasons unconnected to how anybody chased anything. Keep the per-customer view in front of you even when the total looks tidy.
What none of this settles
Adfin's warehouse holds no record of chasing events at all. Nothing here knows what was sent, when, or through which channel, so this page can't tell you that chasing by segment collects more than chasing everybody the same way. The supported claim is narrower: your customers vary a lot, and the variation holds steady enough across five or more payments to be visible.
The 27.4% who were never late didn't necessarily get there on their own. Some are well-run finance functions and some had a supplier who made paying easy, and this cut can't separate them.
And these figures come from a selected slice of one platform's book, weighted towards small practices billing small recurring fees, so test the shape against your own ledger.
What follows for a real ledger
If the variation lives mostly in your customers, two things follow.
Start with measurement at customer level. Rank every client with five or more invoices by how often they missed a due date and you'll have your own version of the table above by this evening. It'll tell you more about your week than any ledger-wide figure.
Then there's the awkward implication for your policy: one identical to everybody is tuned for nobody. Uniformity has a defence in fairness and in being simple to run, and you probably already break it quietly on exactly the history described here. Written down, your exceptions survive somebody's holiday.
All of it runs into your capacity. You can hold a per-customer pattern in your head for ten accounts and not for four hundred. That part suits software better than memory: you set what's allowed per client, how persistent to be, what goes out unreviewed and where you want to look first, and an agent works the ledger name by name from your own email domain, leaving the four judgements above with you. Two questions get muddled here. Is there a best channel or send hour across all businesses and all customers? No reliable public evidence settles it and Adfin can't publish a finding either, so any confident universal answer is a guess. Does your client reply to a text and ignore email? Their own record answers that, a far smaller claim than a rule for everybody.
Common questions
Is credit control a process or a set of judgements? Both, and the split matters for where your effort goes. The sequence of confirmations, reminders and escalations benefits from being written down. Inside each step is a judgement about one customer's history and what the relationship can carry, and that part doesn't generalise.
Do most late payments come from a few customers? In Adfin's book they concentrate heavily. Among 1,359 business-customer pairs with five or more customer-initiated payments, the 24.3% who were late at least 80% of the time accounted for 51.9% of all late payments, and 27.4% were never late once. Those pairs are a selected group, so use the shape and not the exact figures.
Should I chase every client the same way? Uniformity has a genuine defence in fairness and in being simple to run. Against it, your customers' payment records differ enormously and most businesses already make quiet exceptions on that basis. Written down, those exceptions outlast whoever happens to be chasing.
Why does my on-time payment rate get worse when collections improve? Because a rate for one channel describes a mixture of customers, and the mixture changes. As your reliable payers move onto direct debit, whoever is left paying manually is a harder group, so that channel's on-time rate can fall while your overall overdue rate falls too.
What should I measure if I want to chase by customer? The share of invoices each customer paid after the due date, over their last five or more invoices, and how many days late they typically were. Both come out of your ageing and a payment history, and a ledger-wide average hides the concentration completely.
Does chasing customers differently get you paid faster? No data here can answer that. Adfin's warehouse holds no record of chasing events, so nothing measures what was sent or what followed. The supported claim is that customers vary widely and consistently in whether they pay on time.
This article describes patterns in Adfin's own platform data and is not legal or financial advice. The figures are observational: they describe one platform's book of mostly small-ticket UK invoices, and no figure here establishes cause. Last updated August 2026.
