A customer payment reliability score is a single number that sums up how dependably a customer pays you, based on their own track record with your invoices. It usually sits on a simple scale, such as 0 to 100, where a high score means the customer pays on time and in full, and a low score flags someone who pays late, partially or not at all. It turns a gut feel about who pays well into a figure you can actually use.
For a finance team, this score is a quiet workhorse. It tells you who you can extend generous terms to, which customers to watch, and where to focus collection effort before an invoice ever falls due. Unlike a one-off credit check, it is built from how the customer behaves with you, so it stays current and improves the more you trade together.
One number for trust.It distils a customer's payment behaviour into a single, comparable figure.
Built from your own data.It reflects how they pay you, so it stays current as the relationship evolves.
It guides terms and effort.High scores earn better terms; low scores get a closer eye and earlier contact.
A reliability score is built by scoring a few payment behaviours and blending them, usually as a weighted average, into one figure. There is no single official formula, so the weights are yours to set, but these are the inputs that matter most.
On-time payment rateThe share of invoices paid by the due date, the single strongest signal.
Average days lateWhen they do pay late, by how much, since two days differs from sixty.
Payment completenessWhether they pay in full or routinely short-pay and leave balances.
Broken promisesHow often agreed payment dates or plans are missed.
Trend over timeWhether their behaviour is improving or quietly getting worse.
Relationship lengthHow much history you have, since a long record is more reliable than a short one.
The calculator below turns those signals into a score so you can see how it works. Adjust the inputs for a real customer and watch the score and risk band move.
An illustrative model with equal-ish weights. Set your own weights in practice. General information, not financial advice.
The exact weights matter less than consistency: apply the same model to every customer and the scores become comparable, which is the whole point. Good AR reporting is what makes these inputs visible, and credit control software can compute and refresh the score automatically as new payments land.
Take a customer who pays 80% of invoices on time, runs about six days late on the rest, pays 95% in full, and has broken one promise in the last year. Drop those into the model and the score lands around 85, comfortably in the reliable band. The read is simple: this is a good customer who occasionally runs a little behind, so standard terms are fine and a gentle reminder is all the chasing they need.
Now imagine a second customer at 45% on time, twenty days late on average, short-paying often, with four broken promises. Their score falls into the at-risk or high-risk band, which is a signal to act before the next big invoice: tighten their terms, ask for part-payment up front, or contact them the moment something is due rather than weeks later. Same business, two very different customers, and the score tells you how to treat each without anyone having to remember the detail. That ranking of who to chase, and when, is where the reliability score feeds straight into collections scoring.
A credit rating predicts whether a customer can pay, using broad financial and bureau data; a payment reliability score predicts whether they will pay you, using their actual behaviour on your invoices. The two answer different questions. A credit rating, often bought from an external agency, is useful before you have any history, for example when onboarding a new customer or setting an initial credit limit. A payment reliability score only exists once a customer has started paying you, but from then on it is sharper, because it reflects the one relationship you actually care about: theirs with you. A customer can hold a strong external credit rating and still earn a poor reliability score by consistently paying you late, and that gap is exactly what the score is built to catch. In practice you use both: the credit rating to decide whether to extend credit, and the reliability score to manage it once you do.
A reliability score is only worth having if it changes what you do. Tie score bands to concrete actions: reliable customers keep or earn better terms, while at-risk ones get shorter terms, lower limits and earlier contact. Refresh it continuously, because a score that is recalculated as payments arrive catches a customer sliding from good to shaky while you can still respond, which is the same logic behind predictive collections. And keep judgment in the loop: the score is a guide, not a verdict, so a long-standing customer with one rough quarter still deserves a conversation rather than an automatic clampdown. Used that way, the score quietly does the watching for you, surfacing the handful of accounts that need attention so the rest can run on autopilot.
Don't let these critical mistakes hurt your
collections - See how to fix them, today!
Request an AI summary of Paidnice
