Days sales outstanding
How long your money takes to arrive, expressed as a single number that no individual client matches.
Also called DSO, Debtor days, Average collection period.
What it is
Days sales outstanding is the average time between invoicing and being paid. It is the difference between what your profit and loss says you earned and what your bank account contains, which is the difference most businesses that fail were profitable throughout.
(accounts receivable ÷ credit sales) × days in period
- accounts receivable
- Invoiced and not yet paid at the end of the period
- credit sales
- Invoiced sales in the period, excluding anything paid immediately
- days in period
- 365 for a year, or the actual day count for a shorter window
The part the formula leaves out
One average describes nobody. Payment behaviour is not normally distributed around a mean; it is a cluster of clients who pay on time and a small group who do not, and the average sits in a gap between them where no client lives. In the worked cash-flow example, two clients on 30-day terms pay at 74 and 61 days, they are 38% of revenue, and they account for every day the account spent overdrawn. The portfolio average would have hidden all of that.
Terms are not behaviour. An invoice that says 30 days records what you asked for. The forecast needs what happens, and the gap between the two is where the entire problem lives: the example business has invoices that all say 30 days and money that arrives, on average, on day 46.
The denominator also has a trap. Including cash sales, which are paid instantly, drags DSO down without anything changing about collections. A business shifting mix toward card payment will show collections improving while its invoice clients get slower.
How it is usually computed wrongly
01
Computing one DSO for the whole book
It averages the clients who pay on time with the ones who do not, producing a figure that is a true statement about nobody and actionable for no one.
Compute a distribution per client and rank by revenue at risk. The output that matters is which clients, not what average.
02
Including cash and card sales in credit sales
Sales paid at the point of purchase have no collection period. Putting them in the denominator makes DSO fall whenever the payment mix shifts, which reads as a collections improvement that has not happened.
Restrict the denominator to invoiced sales, and state the restriction.
03
Measuring from invoice date when the delay is in invoicing
A business that does the work in March and invoices in May has a collection problem that DSO cannot see, because the clock starts at the invoice.
Measure from delivery to cash as well as from invoice to cash. The difference between the two is entirely within your own control, which makes it the more useful number.
04
Reading a falling DSO as good news during a downturn
DSO uses current sales as the denominator. If sales fall sharply, the ratio falls even when collections have not improved at all, which is the classic false signal in a bad quarter.
Use a countback method that ages the receivable balance against the months it came from, rather than dividing by a denominator that is moving.
What your file needs
- Invoices with an issue date and an amount
- Payments with a date, and a reference that ties them to invoices
- A way to tell credit sales from immediate payment
Anything missing is reported as unavailable rather than substituted with something weaker computed on worse evidence.
Compute it on your own file
No account needed to start. You only pay when you like what you see.