Payment performance is a board-level conversation in 2026. Boards reading the finance pack quarterly need a small, defensible set of metrics that show what payment behaviour actually looks like inside the business, not what the contract terms say it should look like. DPO does not do that job. Neither does DSO. Both can hide late payment, supplier churn and bank-detail switches behind a clean-looking average.
The trouble with DPO and DSO
Days payable outstanding is a ratio. It tells you, on average, how long invoices sit before they are paid. Days sales outstanding tells you the mirror image for receivables. Both numbers serve a purpose, but neither was designed to surface risk.
An average smooths over the long tail. A business with a clean DPO of 32 days can be running a four-week structural delay on its most strategic suppliers and a clean 14 days for everyone else. The average is fine. The relationship is not. Boards reading DPO in isolation will miss that distinction every time. We have written before about the hidden cost of late payments that nobody measures; the same dynamic plays out at board level when the headline ratio looks healthy.
Both metrics also lag. They are calculated at the end of a period, against the period's invoices. By the time the number is on a slide, supplier confidence has already shifted, fraud has already cleared, and any concentration risk is already three months stale.
The four metrics that should sit alongside
The scorecard below is the smallest defensible set that gives a board a real read on payment behaviour. None of these are exotic. Most can be sourced from the ERP, the bank, and an industry benchmark, with one caveat that finance teams usually discover halfway through implementation.
1. Payment timeliness against terms
This is the percentage of invoices paid on or before the contracted due date, weighted by invoice value. It answers the question DPO does not: are we paying when we said we would? Track the figure by supplier tier and by business unit. A blended 88 per cent is meaningless if your top-quartile suppliers sit at 64 per cent.
2. Supplier concentration risk
Calculate the share of total AP spend held by your top ten and top fifty suppliers. Then layer on the share of those suppliers paid late in the last quarter. Concentration on its own is operational. Concentration plus lateness is a relationship risk that surfaces in renewals, price negotiations and supply continuity. Boards should see both.
3. Exception rate on bank-detail changes
This is the AP fraud canary. Track the count and value of bank-detail changes against the count and value blocked or rolled back after exception review. Pair it with the same figure for new supplier additions. If you cannot source the underlying data, that is the answer. The reporting gap is the risk. For the wider context, see why AP fraud is about to scale rapidly in the AI era.
4. Payment behaviour deviation against industry benchmark
This is the only metric on the list that requires data you do not own. The UK Industry Benchmark gives sector-level payment behaviour numbers that let a CFO read their own performance against peers. Sitting at the median is a defensible starting position. Sitting two quartiles below the median in a sector where supplier flight is rising is something the board needs to hear.
How to source each metric
Three of the four can be assembled from a clean ERP and a disciplined AP team. The fourth, behaviour deviation, requires network-level data because a single buyer's history is statistical noise. Most teams discover this at the modelling stage. The pragmatic answer is to pull the first three from the systems you already run, then layer in benchmark data through a connected accounts payable network that sees the supplier across all of its buyers.
Reading the scorecard
Three patterns should trigger action at the board level. First, when payment timeliness drops below 85 per cent and supplier concentration sits above 50 per cent in the top fifty. That is a relationship risk hiding behind operational language. Second, when bank-detail exception rate spikes above the trailing six-month average. Fraud attempts often cluster, particularly around year-end and after a public funding announcement. Third, when behaviour deviation against benchmark widens by more than five percentage points in a single quarter. That signals a structural shift, not a one-off bad month.
What boards should ask for next
The scorecard is a starting point, not a destination. Once it lives in the finance pack, the next question is whether the data is current, verifiable, and aligned across business units. Most CFOs reach for a network layer at that point, not because they want another tool, but because the data needed to keep the scorecard honest sits outside the four walls of the business.
.jpg)