We looked at 53,095 paid invoices from 180 businesses. Roughly four in ten are paid by the due date, and that barely changes whether the terms are due-on-receipt or Net 30. Giving a client thirty days does not make them more likely to pay on time.
Key figures
Across 53,095 paid Billbooks invoices, the average is 25 days from issue to payment. It varies by terms: invoices due on receipt are paid in 15.9 days, Net 15 in 26.9 days and Net 30 in 47.2 days. Around four in ten are paid by the due date.
No. In this study the share of invoices paid by the due date was 39.1% on due-on-receipt terms and 42.3% on Net 30, a difference of three percentage points across a thirty-day gap in terms. Longer terms move the deadline rather than improving compliance.
Yes, almost one for one. Net 30 invoices were paid at 47.2 days on average and due-on-receipt invoices at 15.9 days. The terms differ by 30 days and the outcome differs by 31, so time granted is added to the wait rather than absorbed by faster behaviour.
Every row below met the reporting threshold of at least 25 distinct businesses and 100 invoices. Terms used by fewer businesses than that were excluded rather than reported.
| Payment terms | Invoices | Businesses | Mean days to pay | Paid by due date |
|---|---|---|---|---|
| Due on receipt | 31,794 | 159 | 15.9 | 39.1% |
| Net 7 | 1,418 | 48 | 18.6 | 48.5% |
| Net 15 | 5,276 | 54 | 26.9 | 47.1% |
| Net 30 | 12,620 | 68 | 47.2 | 42.3% |
Invoices opened by the client were paid in 19.3 days against 26.2 days for invoices never opened, a gap of 6.9 days. This is a correlation and not proof of cause: opening an invoice and paying it promptly are both signs of an engaged client.
Stated because a study that only reports what worked is not worth trusting.
A late fee was set on zero of 67,634 invoices. Not a small number, none. With no invoices carrying a late fee there is no comparison to make, so this study says nothing about whether late fees work.
The raw comparison suggests invoices with reminders switched on are paid more slowly. We do not believe that is cause and effect, and we are not publishing it as a finding. Reminders are not assigned at random: a business turns them on for clients it already expects to chase, so the setting marks expected lateness rather than creating it. Separating the two needs a controlled comparison, which this data cannot provide.
Billbooks does not record an industry for a business, so this cannot be measured. Estimating industry from another field would be inventing the dimension rather than measuring it.
Anyone is welcome to cite these figures with attribution to Billbooks and a link to this page.