Chasing invoices: the complete process

Chasing an invoice is one process with nine stages, and only three of them involve a reminder. It begins with the terms you agree before any work starts and the way you arrange to be paid, runs through the due date you choose and the moment your customer first has the chance to pay, and ends either with the money in your account or with a decision to stop.

In this article

The short version

• The process starts before the invoice exists, because the terms and the collection arrangement decide how much chasing you're left with.
• 22.0% of customer-initiated payments land on the due date itself and 33.5% of the on-time ones arrive on the day, measured on 59,777 paid invoices (Adfin platform data).
• 5.5% of those invoices went out with a due date already in the past, so not one of them could ever have been on time (Adfin platform data).
• Most late payment behaves like admin. 14.1% of late payments are a single day late and 48.3% are a week or less, while 7.1% run past 90 days (Adfin platform data).
• Escalation works through the medium and the formality, so a phone call after two emails carries more than a firmer email would.
• Stopping belongs to the process too. A debt costing more to pursue than it would ever return is worth writing off deliberately.

One invoice, from terms to cash

This page follows a single invoice through its whole life. Credit control: a complete guide covers the same territory as a function inside a business: who owns it, what it costs you, what to measure and whether to run it in house. Here the subject is narrower and more concrete: one invoice, from the terms being agreed to the money arriving or the debt being written off.

Nine stages, laid out in the order they happen, so you can see how early most of the decisions get made.

The first four stages are yours alone. Everything after them depends on somebody else, and gets more expensive as you go.

Before the invoice exists

Two things are settled here, and they shape everything downstream.

The first is how long your customer gets. Agree nothing and the statutory default gives them 30 days from the later of the day you did the work and the day they had notice of the amount. How long a customer has to pay an invoice sets that out in full. Agreeing something explicit is usually better than inheriting the default, because a default nobody discussed tends to be a default nobody diarises.

The second is how the money will reach you. There's a real difference between an invoice your customer has to act on and an invoice that collects itself from an account they've already authorised. Where a direct debit mandate already existed when the invoice was raised, 97.0% of invoices were paid and 1.3% were still open six or more months later. Without one, 90.3% were paid and 5.4% were still open, so about four times as much money stayed stuck (Adfin platform data, 152,689 invoices created in the seven months to 31 January 2026).

That isn't a controlled comparison. Mandated invoices are smaller and they belong to customers who had already agreed to a mandate, so some of that reliability was there beforehand. Even so, a recurring billing relationship is where your chasing gets designed in or designed out.

Both decisions belong in a written policy instead of in your memory, and writing a credit control policy that gets followed covers what to put in one.

The due date you choose

The date you type into the invoice does more for you than any message you send afterwards, so it deserves more thought than it usually gets.

Three findings from Adfin's own book are worth having in mind. Invoices due on a Saturday were paid on time 60.4% of the time and Sunday 60.6%, against 65.8% to 67.2% Monday to Friday (Adfin platform data, 59,777 paid customer-initiated invoices). Longer terms went with better on-time payment and never worse: 59.5% on same-day terms against 76.4% at 8 to 14 days, flattening around 74% to 76% out to 60 days. And 5.5% of those invoices, 3,271 of them, carried a due date that had already passed when they were issued, so every one of them was late on arrival.

Read the terms finding as an association and not as advice to extend your terms. Businesses that set proper terms tend to run a tighter ledger anyway, and the same table shows the trade-off, since longer terms bring a longer delay on the ones that do go late. The full table, with denominators, is in the UK Getting Paid Report.

The back-dated group is the cheapest thing on this page to look at in your own ledger. Some of it will be ordinary catch-up invoicing raised after the fact, and some of it will be a template quietly putting yesterday's date on today's work. Where a customer is pushing for 60 days, how to invoice a client who insists on 60-day payment terms works through what to trade for it.

Getting it in front of someone who can pay

The gap between sending an invoice and someone with authority to pay it actually reading it is the least visible part of this process, and a surprising amount of delay accumulates there.

Once your customer can pay, many of them do so quickly. 23.0% of customer-initiated payments arrived within an hour of the request, a quarter within two hours, 42.7% within a day and 67.5% within a week, with a median of 53 hours (Adfin platform data, 59,778 paid invoices). Those hours are counted from when the request existed, so any time lost before a person sees it gets added to the end of your cycle instead of absorbed.

Reminders aren't what helps at this stage, but four things about the invoice itself do. Send to a named person and copy the shared inbox, in that order. Put the amount and the due date where they'll be read on a phone. Carry whatever purchase order or reference your customer's system needs. And give them a way to pay from the message itself: that's what payment links are for.

While you're there, find out who approves payment. Discovering on day 30 that your invoice needed a second signature costs you a month.

The due date itself

Of the on-time customer-initiated payments in Adfin's book, 33.5% arrived on the due date itself and 40.5% in the final two days, and 22.0% of all 59,777 payments landed exactly on the day they were due (Adfin platform data). The median on-time payment came three days early.

So payments cluster tightly on the date you set. Adfin's warehouse holds no record of chasing events at all, so nobody here can tell you whether a due-date reminder produced that cluster or whether your customer's own diary did. Either way the date is the trigger and any reminder is downstream of it, and most of your leverage lies before the invoice ever goes overdue.

So if your own on-time rate is disappointing, look at the dates you're setting and the arrangement behind them before you look at the wording or the frequency of your follow-ups.

The first week past due

An invoice that misses its due date has usually not been refused by anybody. Among late customer-initiated payments in Adfin's book the median delay was 8 days, 14.1% were late by exactly one day, 27.3% by three days or fewer and 48.3% by a week or less (Adfin platform data, 20,533 late payments).

Half of late payment, then, is a few days of administrative slippage. The first follow-up is best written as though that's what it is: a short message with the invoice number, the amount, the date it was due, one specific question and the payment route again. Asking whether anything is holding it up gets you an answer. "Please remit at your earliest convenience" doesn't.

Two things hold this stage together. Set the follow-up schedule when the invoice goes out, because a schedule you invent after the due date has already lost the days you were trying to save. And try not to send the same message a third time in the same channel, since a change of medium carries the escalation better than a change of adjective.

Most of the difficulty here is in the timing, so if you want yours written out as dates and actions, how to chase an unpaid invoice lays one out, how often should you chase an unpaid invoice deals with the gaps between the messages, and payment reminder email templates gives you wording you can paste into them.

When it needs a conversation

Somewhere around two to three weeks over, written reminders stop adding information. At that point a call to a named person does two jobs an email can't: it tells you which of two very different situations you're in, and it produces a date.

The distinction to listen for is whether your customer can't pay right now or has chosen not to yet. A customer who can't pay will often take an instalment arrangement and stick to it, and you keep the relationship and most of the money. A customer who has simply deprioritised you responds to a specific commitment and a diary entry. Both go better with the invoice open in front of you and one question ready: when will this be paid, and by what route.

If the invoice is disputed, reminders tend to harden the position instead of moving it, so answer the dispute and restart the clock on the corrected amount. Difficult payment conversations has the sentences for all three cases, including the ones worth avoiding.

Putting the formal position in writing

Around a month over, the tone changes from admin to commercial. On a business to business debt you can charge statutory interest at the Bank of England base rate plus 8%, running at 11.75% as of June 2026, and claim fixed compensation of £40, £70 or £100 depending on the size of the debt.

That right applies automatically on business debts whether your contract mentions it or not, and very few businesses use it: only 5.9% had introduced or increased an overdue penalty on their sales invoices, in research for DBT and the Small Business Commissioner. Applying it consistently from the start of a relationship is easier than introducing it to one client who has annoyed you.

Late payment interest and compensation covers what you can claim and how to present it, and UK late payment rules has the wider legal position. When the formal notice hasn't worked either, the next document is a letter before action, and what to send and when sets out what it has to contain.

Recovery, and the decision to stop

Past about two months, an unpaid invoice stops being a chasing problem. The tail is real: 7.1% of late payments in Adfin's book ran past 90 days and the 99th percentile was 321 days (Adfin platform data, 20,533 late payments). Money at that age is where your write-off risk actually lives, and a decision serves you better there than another reminder.

Recovering overdue debt: the steps in order compares the three routes open to you properly. A debt collection agency takes a percentage and takes the relationship with it. The small claims court is cheap and slow, and works best on an undisputed debt with a paper trail. A payment arrangement recovers less per month and more in total, if the customer is solvent.

The fourth route is stopping, and that counts as an outcome of the process here. Where pursuing a debt would cost more in fees and your own hours than you could realistically recover, the money is better spent elsewhere. Bad debt and write-offs covers the accounting side and how to make the call without agonising over it.

The same process across a whole ledger

One invoice at a time is how the process works. A ledger is how you actually experience it, and two habits make that manageable.

Start from the ageing instead of the inbox. How to read an aged debtor report covers the buckets and what they hide, including why a book collected largely by direct debit makes the 1 to 30 day column look worse than it is.

Then accept that your ledger isn't uniform. Among repeat customer-initiated payers in Adfin's book, 27.4% were never late once, while the 24.3% who were late most of the time accounted for 51.9% of all late payments (Adfin platform data, 12,863 payments across 1,359 business-customer pairs with five or more payments each). Those pairs are a selected group, since a customer who paid promptly by card five times running is exactly the one most billers would have moved onto a mandate, so read the shape and not the exact proportions. How to set different chasing rules for different clients turns that into something operable, and what credit control is actually deciding takes the argument further.

Whether automating the follow-ups earns its keep is answered in do automated payment reminders actually work. The method behind every platform figure on this page, including what the data can't see, is in how we measure UK payment behaviour.

Which parts of this need you

Read back through the nine stages and they split cleanly into two kinds of work.

Some of it needs a person and always will. The call where a client explains they're in trouble, the judgement about whether to extend terms or hold the line, the decision to escalate or to write off: those are commercial calls about your own business and your own relationships, and handing them to software would be an odd thing to do.

The rest is repetition. The same confirmation before the due date, the same short reminder afterwards, the same escalation, against a different name every week. It competes for your attention with everything else in the week, and that's how a schedule set on day one quietly becomes a schedule nobody ran. Businesses affected by late payment already spend an average of 86 hours a year on this, on the Small Business Commissioner's research.

That repetition can be delegated, and delegating it well means more than a fixed schedule. You set the rules of engagement once: which channels are allowed, how persistent to be, what can go out without you looking at it, and where you want to review before anything is sent. An agent then works the sequence customer by customer, from your own email domain, adapting to what each account has responded to before.

There are two questions here and they usually get treated as one. Is there a best channel or send hour across all businesses and all customers? There's no reliable public evidence either way, Adfin can't publish a finding on it either, and any confident universal answer is a guess. Does this particular client reply to a text and ignore email? Their own record answers that, and a person can hold that pattern in their head for ten accounts and not for four hundred.

Common questions

What is the correct process for chasing an invoice? Agree terms and a collection method before the work starts, choose a working-day due date, get the invoice to a named person who can approve it, follow up briefly the day after it's due, again about a week later, then move to a phone call. Around a month over, put the formal position in writing with interest and compensation, then decide between an agency, the small claims court, an arrangement or a write-off.

How long should I wait before chasing an unpaid invoice? A confirmation a few days before the due date prevents more delay than any reminder after it, because it catches routing and approval problems while there's still time. After that, the day following the due date and then roughly weekly for the first month is a defensible pattern. Decide the schedule when the invoice goes out.

Does chasing invoices actually get them paid faster? Nobody can answer that from Adfin's data, because the warehouse holds no record of chasing events. What it does show is that payments cluster on the due date, with 22.0% of customer-initiated payments landing exactly on the day they were due, so the date you set and the way you arranged to collect are doing a lot of the work before any follow-up.

When should I stop chasing an invoice and write it off? When the cost of pursuing it, in fees and in your own time, is more than you could realistically recover. Age is the signal to look at: 7.1% of late payments in Adfin's book ran past 90 days, and debt that old rarely improves on its own. Writing it off is a decision, and it frees the time for invoices you can still collect.

Can I charge interest on a late invoice? Yes, on business to business debts, at the Bank of England base rate plus 8%, running at 11.75% as of June 2026, plus fixed compensation of £40, £70 or £100 by the size of the debt. The right applies automatically even where your contract says nothing about interest.

How is this different from a credit control process? Credit control is the function: the policy, the ownership, the metrics and the decision about resourcing it. This page is one invoice's journey through that function, from the terms being agreed to the cash arriving or the debt being written off. If you're designing the function, start with the credit control guide instead.

Sources

• GOV.UK — claim debt recovery costs on an unpaid invoice (accurate as of August 2026)
• Small Business Commissioner — late payments research (accurate as of August 2026)
• DBT and the Small Business Commissioner — late payments research, impact on the UK economy (accurate as of August 2026)

This article explains how invoice chasing works and is not legal advice. The statutory interest rate moves with the Bank of England base rate, so check the current position before relying on a figure. Last updated August 2026.