5 early signs of late payments (before DSO increases)

Most finance teams rely on DSO to understand cash flow performance. But DSO is a lagging indicator. By the time it increases, the problem has often already been there for weeks, and teams are reacting to symptoms rather than causes.

“Late payments rarely appear suddenly. They start with signals in the invoice-to-cash process, signals that credit teams can spot earlier. Late payments are rarely the real problem. They are the result of something in the process breaking down,”

…explains Willand Brienen, Product Manager at Onguard and former credit manager.

Understanding the early signs of late payments allows finance teams to intervene before overdue invoices start to affect DSO. Below are five indicators credit teams can use to predict late payments earlier.

1. Early signs of late payments: invoices that are not first-time-right

Every credit manager knows the scenario: a customer calls to say the invoice is incorrect. Wrong PO number, wrong pricing, wrong address. The invoice gets corrected and resend and the payment clock often restarts from zero. Every correction restarts the payment clock.

A useful question for credit teams is: How many invoices are correct the first time they are sent?

The first-time-right (FTR) rate is one of the most underused metrics in AR. Most teams measure collection performance in detail but have no idea what percentage of their invoices are accepted without correction. That gap matters because invoice errors are not random. They cluster around specific customers, product lines, or internal handovers. Fixing the source eliminates the delay structurally, not just once.

Improving invoice accuracy, through better templates, source data and validations, often reduces delays more effectively than sending additional reminders, says Willand. 

“The stakes are rising. With mandatory e-invoicing rolling out across the EU (Peppol standards), the tolerance for invoice errors is about to drop sharply. An incorrect invoice will no longer just delay payment; it will be technically rejected before it reaches a human. Organisations that treat invoice accuracy as a back-office issue may find it becoming a compliance issue sooner than expected.”

What to measure: FTR ratio per customer segment and per invoice origin. Track correction cycles per invoice and time lost per correction. Even a rough monthly view reveals patterns that are otherwise invisible.

2. Dispute ratios that signal future payment delays

Another early sign of late payments is a high dispute ratio. 

Most organisations log disputes somewhere, but the data rarely goes anywhere useful. Credit manages the follow-up, sales manages the relationship, and operations manages the delivery. Nobody owns the dispute itself. 

Two indicators matter, and and both need to be tracked together:

• dispute volume (tells you something is wrong)
• dispute resolution time (tells you how badly it’s affecting cash)

If disputes remain open for weeks, invoices effectively remain unpayable during that period. Credit teams that reduce time-to-cash focus on solving the cause of disputes, not only managing collections, explains Willand. 

“A practical starting point: categorise every dispute at the moment it’s opened: by type (pricing, delivery, data error, duplicate) and by origin (internal or external). Teams that do this consistently gain something most don’t have: a monthly root-cause report that shows operations and sales exactly where the process breaks down. That conversation is often more valuable than any collection reminder.”

What to measure: Dispute volume by category and by origin (internal error vs. customer claim). Average resolution time per category. Month-on-month trend. A monthly root-cause report shared with operations is a starting point most teams don’t have but quickly find indispensable.

3. Invoice approval latency

This approval latency is one of the most overlooked early signs of late payments.

In many organisations, invoices are not rejected. Unlike disputed invoices, they don’t trigger alerts or workflows. So, they simply wait. Waiting for approval, for the right contact person, for order confirmation, while DSO keeps climbing.

A useful metric to monitor is: How long does an invoice wait for customer approval?

Clear routing, correct PO information and known contact persons often prevent invoices from getting stuck before the payment process even begins. However, there is an underestimated issue, Willand points out: 

“One underestimated issue: invoices stuck in approval limbo still count as outstanding in DSO calculations. They don’t show as disputed, they don’t trigger collection workflows, they just wait. Tracking invoice age from send date to first customer action reveals a bottleneck that most dashboards make invisible.”

What to measure: Time from invoice send date to first customer action (confirmation, query, or payment). Segment by customer and by invoice type. Anything consistently over ten working days without a response warrants a process review, not just a reminder.

4. Promise-to-pay behaviours that predict late payments

Every credit manager has a mental model of which customers they trust. The ones who always pay on time. The ones who promise to pay by Friday and mean it. And the ones who promise to pay by Friday and have never once actually paid by Friday.

The problem is that this mental model rarely gets formalised. Teams follow up based on invoice age, a lagging indicator that tells you how long something has been overdue, not how likely it is to be resolved. 

Promise-to-pay reliability, tracked as a simple fulfilment percentage per customer over a rolling period, turns this intuition into a usable signal. 

“Tracking promise-to-pay reliability helps credit teams segment customers more accurately. A customer who has kept every promise for three years deserves a different approach than one who has missed four in a row. That distinction is where earlier action begins.”

What to measure: PTP fulfilment rate per customer, rolling 12 months. Customers below a meaningful threshold (teams typically find 70–75% a useful marker) should trigger earlier contact, not post-due follow-up.

5. Low auto-match rates in incoming payments

Cash has arrived. The customer has paid. But in the books, the invoice is still outstanding. Why? Because the payment couldn’t be matched automatically.

This scenario is more common than most teams realise. Unmatched payments create a false picture of outstanding receivables and can trigger collection reminders for invoices that have already been paid. A customer who settled an invoice three days ago receives a formal follow-up asking for payment. That’s a trust problem.

The root causes are almost always internal: missing or inconsistent payment references, invoice data that doesn’t match what’s in the customer’s system, or bank integrations that process in batches rather than in real time. Customers aren’t failing to pay; the organisation is failing to recognise that they have.

“In many cases, this is not a customer payment problem but a data and process problem inside the organisation. Improving payment matching often reveals that time-to-cash problems start internally, not with the customer,” says Willand. 

What to measure: Auto-match rate as a percentage of total incoming payments, split by customer and by payment channel. Track unmatched payment volume weekly. The pattern in the mismatches (i.e. which customers, which reference formats, which channels) usually points directly to the fix.

Why these 5 signs matter more than DSO 

These signals describe a sequence:

1. An invoice that isn’t first-time-right either triggers a dispute or gets stuck in approval. 
2. A dispute that isn’t resolved on time produces a broken promise-to-pay. 
3. A payment that eventually arrives can’t be matched because the reference was lost somewhere in the correction cycle. 
4. DSO is the final score, but the game was decided much earlier.

Credit teams that focus on the early signs of late payments (i.e. where the process slows down, rather than on how overdue the results are) operate differently. They:

• escalate earlier, on the right accounts. 
• bring data to conversations with operations and sales that would otherwise never happen. And 
• find that most of what looks like a collections problem is actually a process problem with a process solution.

As Willand summarises:

“Credit teams that shorten time-to-cash don’t just monitor DSO. They monitor where the process starts slowing down, and they make that visible to the people who can actually fix it.”

Conclusion

Late payments are rarely about customers unwilling to pay. More often, they result from invoices that contain errors, disputes that remain unresolved, approvals that take too long, or data that does not flow correctly between systems.

Credit teams that focus on the early signs of late payments gain the insight needed to act before delays become visible in DSO. And that is where real improvements in time-to-cash begin.

“A useful starting point: take these five signals and ask, for each one, whether your team currently measures it and where that data lives. Most teams find that two or three signals are tracked somewhere, but not consistently, and not in one place. That gap is usually where time-to-cash improvements begin,”

…concludes Willand.

Want to explore what this looks like in practice? Let’s talk.

About the expert

Willand Brienen is Product Manager at Onguard and previously worked as a credit manager. His experience in both operations and software development gives him a unique perspective on how finance teams can detect payment delays earlier in the order-to-cash process.