Did sales or receivables move DSO?

Part 2 decided whether a DSO movement needs explaining at all. Most don’t. This part is about the ones that do: once there is a signal, who gets the question?

DSO is receivables divided by sales. Either one can move it.

Before anyone presses the big red “AR, please explain” button, the reporting pack should show how much of the movement came from each. Interrogating the credit controller because the commercial team had a slow month is traditional, yet not very helpful. Part 1 called this the variance question, and the two numbers needed to answer it are already in the pack.

Two effects

The idea is simple. Change one number at a time and see how far DSO moves.

1. Keep receivables where they were and change sales to the new figure. The change in DSO is the sales effect.
2. Then change receivables to the new figure. The rest of the movement is the receivables effect.

The two add up to the whole movement. “Previous” can be last month, the same month last year or the forecast.

Change receivables first and the split comes out slightly different, so do it both ways and take the average. That way nobody gets to pick the order that suits them. The average can be calculated directly.

Back to the monthly review

Part 1 opened with DSO three days above forecast. Against a forecast set at the baseline average, Part 2 showed that a miss that size is a signal, so this one earns a question. Suppose the numbers were these, with DSO calculated as month-end receivables over the month’s credit sales, times 30.

	Forecast	Actual
Credit sales	12.0m	10.2m
Receivables	18.8m	17.0m
DSO (days)	47.0	50.0

	Sales first	Receivables first	Average
Sales effect	+8.3	+7.5	+7.9
Receivables effect	−5.3	−4.5	−4.9
Movement	+3.0	+3.0	+3.0

The average puts the sales effect at +7.9 days and the receivables effect at −4.9.

Sales came in 15% below forecast, and on its own that adds almost eight days. Receivables came in 1.8m below forecast, which takes about five days off.

Sales drives the movement and receivables partly offsets it. The first question goes to Sales: why did sales miss forecast? AR doesn’t get the first question. Had both effects pushed DSO up, both would get one.

When it sits in receivables

Go back to shift C in Part 2: eight months running about a day above the baseline. No single month broke the limits. Look at any one of them and the natural reaction is “inside the limits, move on”. I would have done the same. The signal is the run, so the question is not about one month. It is about why DSO has sat higher since February.

	Baseline year	Feb to Sep 2026
Average monthly credit sales	12.0m	12.0m
Average receivables	18.84m	19.20m
DSO (days)	47.1	48.0

Sales contributes nothing to the increase. The movement, 0.9 days, sits entirely in receivables, so the first question goes to AR: what has kept receivables higher since February? Their ageing analysis shows where it came from, and Part 4 follows that answer.

Where it sits is not why it happened

The split says where the movement sits in the formula, and so who gets asked first. It does not say why it happened.

The receivables effect is not a collections effect.

Effect	Could be
Sales	Volume, seasonality, price changes, a large one-off order, invoicing that slips into the next month
Receivables	Payment terms and customer mix, disputes, credit notes, unapplied cash, a large customer paying a day after month end, slower collection

Payment terms and customer mix are agreed when Sales signs the contract, yet they land in the receivables line. Disputes and credit notes often start in billing or delivery. A receivables effect can be a commercial question as easily as a collections one.

Checklist: before anyone is asked

1. Is the movement a signal under the rules in Part 2?
2. Has it been split into a sales effect and a receivables effect?
3. Which effect drives the movement, and which offsets it?
4. Who should get the first question?

Try it on your own numbers

Type in sales and receivables for two periods below. There is also a full-page version to bookmark, and an Excel version that runs the split for every month of the year.

Part 4 picks up AR’s answer on shift C, and what to measure instead of DSO.