.svg)
Most AP teams can tell you how many invoices they processed last month. Fewer can tell you what it cost to process each one, how long the average invoice took from receipt to payment, or what percentage passed through without generating an exception.
Those are the accounts payable KPIs that actually reveal whether a finance function is running efficiently or carrying hidden costs. This guide covers which metrics are worth tracking, what the benchmarks look like for mid-market businesses, and how to read the numbers to identify where the process needs attention.
Invoice volume is easy to measure. It is also one of the least informative metrics for assessing AP performance. A team processing 1,000 invoices a month with a 3% error rate and a five-day approval cycle has a different problem profile than a team processing the same volume with a 0.5% error rate and same-day approvals.
Accounts payable metrics give finance leaders a structured way to answer three questions:
Without the metrics, these questions get answered by gut feel. With them, they get answered by data.
Cost per invoice is the total cost of running the AP function divided by the number of invoices processed in the same period. It includes staff time, software costs, and any costs associated with error correction and exception handling.
What good looks like: Manual AP processing in the UK typically costs between £8 and £25 per invoice, depending on complexity and error rate. On well-implemented automated platforms, this figure falls below £2.
What it tells you: If your cost per invoice is above £15, you are carrying significant manual processing overhead. If it has stayed flat or risen as invoice volume has grown, your process is not scaling.
How to calculate it:Total monthly AP function cost (staff + software + overhead) ÷ Total invoices processed that month
Invoice processing time measures the average number of days from invoice receipt to payment authorisation. It is distinct from payment timing (which also includes when the actual payment runs), but it is the metric that most directly reflects how efficiently the AP workflow operates.
What good looks like: In a manual environment, invoice processing time typically runs three to seven business days. On automated platforms, invoices that match without exceptions move through in hours. An overall average of one to two days is achievable when straight-through processing rates are high.
What it tells you: Long processing times indicate bottlenecks, usually in the approval stage or exception handling. If processing time spikes at month-end, that is a sign that reconciliation and matching are still largely manual.
Straight-through processing (STP) rate is the percentage of invoices that move from receipt to payment approval without any manual intervention: no exception raised, no human matching step, no approval chase required.
What good looks like: In a manual AP environment, the STP rate is effectively zero. On a well-implemented AI-native platform, rates above 80% are achievable within the first 90 days, rising further as the system learns from your data.
What it tells you: STP rate is the single clearest indicator of how much work the AP function is genuinely automating. A platform that claims to automate AP but has an STP rate of 40% is still routing 60% of invoices through manual steps. That gap matters when you are trying to understand what you are actually buying.
Exception rate is the percentage of invoices that generate a match failure, validation error, or approval query that requires manual resolution.
What good looks like: In a manual environment with no formal exception tracking, exceptions often go unmeasured. On automated platforms, a well-configured system should generate exceptions on fewer than 20% of invoices at go-live, falling further over time as matching logic adapts to your supplier base.
What it tells you: A high exception rate indicates problems upstream: poorly configured matching tolerances, supplier data that does not match the PO, or a high proportion of non-PO invoices entering the system without the context needed for automatic matching. Exception rate should be tracked over time, not just as a point-in-time figure. If it is not falling, the system is not learning.
On-time payment rate measures the percentage of invoices paid within the agreed payment terms. It is the metric that most directly affects supplier relationships and, for businesses that use early payment discount programmes, the metric that determines whether those discounts are being captured.
What good looks like: Finance teams with well-controlled AP functions typically achieve on-time payment rates above 95%. In manual environments, rates of 75% to 85% are common.
What it tells you: A low on-time payment rate is almost always caused by approval delays rather than payment processing speed. If invoices are arriving on time, being matched correctly, and still not being paid on time, the approval workflow is the constraint.
Duplicate payment rate measures the percentage of invoices that result in a payment being made twice for the same transaction.
What good looks like: In manual environments, KPMG benchmarking data puts duplicate payment rates at 0.1% to 0.5% of total invoice spend. On platforms with automated duplicate detection at the point of invoice receipt, this figure should be effectively zero.
What it tells you: Any measurable duplicate payment rate represents recoverable money that the business has paid out unnecessarily. On an annual AP spend of £10 million, a 0.2% duplicate rate is £20,000 in overpayments. This is also one of the metrics that most clearly justifies the investment in AP automation: the saving on duplicate payments alone often exceeds the platform cost.
This metric measures the percentage of suppliers for whom a monthly statement reconciliation is completed. It is less common than the other KPIs on this list but particularly valuable for businesses with complex supplier bases.
What good looks like: Finance teams with manual processes typically reconcile 30% to 50% of supplier statements monthly, limited by the time available. Automated platforms that match statements automatically can achieve close to 100% coverage.
What it tells you: Gaps in statement reconciliation coverage are where undetected discrepancies live. If you are only reconciling half your suppliers, the other half are unreviewed for billing errors, credit notes, and potential fraud.
Early payment discount capture rate measures the percentage of available early payment discounts that the AP function actually takes. When suppliers offer terms such as 2/10 net 30 (a 2% discount for payment within ten days), the question this metric answers is: how many of those opportunities is the business converting into savings?
What good looks like: In manual AP environments, organisations typically capture between 20% and 40% of available discounts, mainly because the invoice processing cycle runs longer than the discount window. On automated platforms where invoices are matched and approved quickly, capture rates above 80% are achievable.
What it tells you: A low capture rate means the business is leaving money on the table. On annual AP spend of £5 million where 15% of invoices carry a 2% early payment discount, a capture rate of 30% versus 85% represents a real cash difference of around £41,000 per year. This is one of the few AP metrics where improvement translates directly into a measurable pound saving rather than a cost reduction. It also signals whether the approval cycle is fast enough to support supplier relationship programmes built around early payment.
How to calculate it:(Value of discounts captured in the period ÷ Total value of discounts available in the period) × 100
Benchmarks are useful context, but they are not targets in themselves. The more useful approach is to track your own metrics over time and measure improvement from your own baseline.
A business processing 200 invoices a month with a cost per invoice of £18 that reduces to £6 over twelve months has demonstrated real improvement. Whether that £6 is above or below the industry benchmark matters less than the direction and magnitude of the change.
The metrics that matter most are the ones you can act on. If processing time is high but the exception rate is low, the bottleneck is probably in approval routing. If the exception rate is high but processing time is low, the matching configuration needs attention. Reading the KPIs together tells a more complete story than any single number.
Finance leaders who want to track these AP performance metrics systematically need them in one place, updated in real time rather than compiled in a monthly report.
A basic AP dashboard should show:
Most AP automation platforms provide this natively. If yours does not, or if the data lives across two systems that are not integrated, the dashboard becomes a manual exercise that is only as current as the last update, which defeats its purpose.
Dost provides real-time visibility into the AP pipeline and the metrics that matter: processing time, exception rate, STP rate, and approval cycle time, all visible without pulling a report or waiting for a monthly close.
Because Dost integrates in real time with SAP, Microsoft Business Central, Sage (200, Intacct, X3), and Oracle, the numbers in the dashboard reflect the current state of the AP function, not yesterday's batch. Finance leaders working from Dost data are making decisions on current information, not a lagging picture.
See what the numbers could look like for your AP function. Use Dost's ROI calculator to estimate your cost per invoice and potential savings based on your current volume.
For UK mid-market businesses processing between 500 and 5,000 invoices a month, the manual processing cost typically sits between £8 and £20 per invoice. Automated platforms bring this below £2. The precise figure varies based on how much of the process is genuinely automated versus how much still requires manual intervention. The most important number is your own baseline, tracked consistently over time, so you can measure actual improvement rather than comparing against a benchmark that may not reflect your process complexity.
Straight-through processing rate is the number of invoices that moved from receipt to payment approval without any manual action, divided by the total invoices processed in the same period. Most AP automation platforms track this automatically. If yours does not, you can approximate it by counting the invoices that generated no exceptions and required no manual matching step, then dividing by total invoices. The key is to be consistent in how you define "manual action" so the metric is comparable period over period.
Core metrics like exception rate and processing time should be visible in real time so the AP team can identify emerging issues before they affect payment runs. Strategic review of the full KPI set is worth doing monthly, aligned with the close cycle, to identify trends and adjust priorities. Annual benchmarking against industry data is useful context but should not replace the internal trend analysis, which is more actionable.
Accounts payable KPIs are the instrument panel for a finance function. Without them, decisions about process improvement, staffing, and technology investment are based on anecdote rather than evidence. With them, finance leaders can identify exactly where the process is costing money, where supplier relationships are at risk, and where the return on automation investment is being realised.
The metrics that matter most are cost per invoice, processing time, straight-through processing rate, on-time payment rate, and early payment discount capture rate. These five numbers, tracked consistently over time, tell most of the story. The others fill in the detail.
Calculate your current cost per invoice and potential savings with Dost's ROI calculator.
.png)
