In short: The cash conversion cycle has three components, days inventory, days receivable and days payable, and planning moves only the first of them directly. Converting a week of coverage into money finance recognises means multiplying the weekly cost of goods by the weeks removed, and that translation is what gets a planning decision into a treasury conversation. A one off release of stock and a change in the ongoing run rate are different things to a finance function, since the first is a single cash inflow while the second lowers the funding requirement permanently. Measuring the cycle where the decisions are made, by category or by branch rather than as a company total, is what makes the number usable by anyone who can move it.
Take a distributor turning over 400 million at a 22 percent gross margin. The P&L is unremarkable and the monthly treasury call is entirely about the revolver. Between paying the supplier and being paid by the customer there are eleven weeks, and eleven weeks of turnover has to be financed by somebody. That financing is the largest use of capital in the business, ahead of the fleet, the buildings and the systems put together.
Distribution is a working capital business before it is a forecasting business. The forecast matters because it sets coverage, coverage sets the inventory, and the inventory is where most of the borrowed money is sitting. Planners rarely see it framed that way, and finance rarely sees the planning decisions that produce it.
The three numbers, and who actually moves them
The cycle is days inventory outstanding plus days sales outstanding minus days payables outstanding. Three components, and planning's grip on each is very different.
Days inventory outstanding is almost entirely a planning output. Coverage targets, order frequency, whether a supplier minimum order quantity is accepted or negotiated, where stock sits in the network, seasonal pre-build, and how many items are listed all land in this number. If the cycle is eleven weeks and DIO is nine of them, planning owns the large majority of the business's working capital position whether or not anyone has said so out loud. How much coverage each item should carry is a sizing question with its own method, and it is a separate discussion from this one.
Days sales outstanding looks like a credit control number, and mostly it is. There is a planning component that almost nobody measures. Invoices in dispute do not age normally. A short shipment, a substituted pack, a delivery that arrived outside an agreed window, a proof of delivery nobody can locate: each of those stops the payment clock while two administrative teams reconcile it, and the delay runs to weeks rather than days. If four percent of invoiced value is in query at any moment and queries take an average of thirty days to clear, that is more than a day added to DSO, caused by a fulfilment defect. Worth computing against your own aged debt, because the fix belongs to operations rather than to collections.
Days payables outstanding sits outside planning altogether. Terms are negotiated commercially and a planner cannot change them. What planning does change is the relationship between the two ends. If supplier terms are 60 days and stock cover is 90, you fund 30 days of goods from your own facility on every turn. If cover is 45 days, the supplier funds you. That pairing is the object worth managing, and it goes invisible when DIO and DPO are reported to different people on different slides.
Turning a week of coverage into a number finance recognises
The translation fits on a page, and doing it changes how planning proposals get received.
Start with annual cost of goods sold and divide by 365. At 312 million of COGS that is about 855 thousand a day, so a week of coverage is roughly 6 million of inventory at cost. Take a week out across the network and 6 million of cash comes back once.
The second number is the ongoing one. Holding 6 million less costs less to hold, every year, for as long as the coverage stays out. Build the carrying rate from your own figures rather than from a textbook percentage, because the components differ enormously between businesses. The cost of capital should be your marginal rate, which for a business drawing on a revolver is the revolver rate rather than a blended weighted average cost of capital. Storage and handling should be the part that genuinely varies with volume, which is smaller than total warehouse cost because most warehouse cost is a step function. Obsolescence and markdown should come from your own write-off history at category level, then insurance and shrink on top.
Suppose that stack comes to 18 percent. The 6 million release also produces about 1.08 million a year of carrying saving, and that part repeats.
Two cautions on the arithmetic, both of which a finance reviewer will apply anyway. Value the release at cost, since cost is what sits on the balance sheet. And use marginal storage cost, because if the warehouse is half empty and no lease is being surrendered, the storage component is worth zero this year and only becomes real when you exit space or avoid signing for more.
Why the release and the run rate are different animals to a CFO
The 6 million and the 1.08 million behave differently, and confusing them is the fastest way to lose a working capital business case.
The 6 million is a one-off release. It appears in the cash flow statement as a movement in working capital, it happens once, and repeating the same action next year does not repeat it. It is genuinely useful. It can fund a capital programme without new debt, cover a covenant test, pay an acquisition deposit, or take the revolver down. Treasury cares about it a great deal in the quarter it lands.
The 1.08 million is a run rate. It appears in operating cost, it recurs, and in a valuation it attracts whatever multiple the business trades at. A CFO planning a three-year path cares about this one more.
A business case that adds the two together and presents 7.08 million as the value of the project will be sent back, correctly. Present them as two lines with two labels and say which year each one lands in.
The distinction cuts the other way as well. Growth eats the release. A business growing twelve percent a year that holds coverage flat in days will see inventory in currency rise twelve percent, so a 6 million release is absorbed inside a year and the balance sheet number returns to where it started. This is why finance asks about days rather than value, and why a planning team reporting a reduction in inventory value during a growth year is reporting something other than what it achieved. Report days of cover, with value alongside.
Measure the cycle where the decisions get made
A company-level cycle is an average over a portfolio that ranges very widely, and the average points at no action.
The components do not move together across that portfolio. A domestically sourced category with four weeks of cover on sixty day terms is being financed by the supplier. An imported category with fourteen weeks of cover paid against documents at shipment is financed entirely by you, and needs around ninety days of funding per turn before a customer pays anything. A single blended figure describes neither, and a uniform target set against it will be met by whichever of the two is easier to move.
The money is also concentrated. Rank your categories by cash tied up, which is that category's cycle in days multiplied by its daily cost of goods, and the top few will hold a majority of the working capital in the business. That ranked list is the real work list, and it usually bears little resemblance to the list of categories anyone has been asked to reduce.
Ageing matters as much as the average. Two businesses with identical DIO carry different risk if one holds a fifth of its stock older than twelve months. That portion has a cash number attached and a probability of converting that is well below one, and it belongs in a different conversation from the coverage discussion.
One measurement caution. A cycle computed from a period-end snapshot in a seasonal business is close to meaningless. Use an average of monthly balances and a trailing twelve month cost of goods figure in the denominator, so that a seasonal quarter does not manufacture an improvement.
The payables trap
The cheapest way to improve the cycle on paper is the third component. Move from 45 day terms to 75 and thirty days come out immediately, with no planning work, no service consequence, and no implementation project. It is also the move that gets proposed most often once a working capital target exists.
The cost does not disappear, it transfers to the supplier, and it returns in a form that is harder to see. Murfin and Njoroge (2015), writing in the Review of Financial Studies, examined what happens when large buyers extend payment terms to smaller suppliers and found that credit constrained suppliers cut investment in response. A supplier funding an extra month of your working capital funds it at their own cost of capital, which for a smaller firm sits well above yours. That cost comes back as price at the next negotiation, or as a quieter reduction in the attention your orders get when their capacity is short.
There is a version of this where the price is printed on the invoice. Ng, Smith and Smith (1999), in the Journal of Finance, documented how standardised two ten net thirty terms are across US industries. Forgoing a two percent settlement discount to take twenty extra days of credit costs two divided by ninety-eight over those twenty days, which annualises to well above forty percent. A business stretching payables while letting settlement discounts lapse is borrowing at a rate no treasurer would sign for if it arrived as a term sheet.
Supply chain finance is the defensible version of the same move. The buyer extends terms, a bank pays the supplier early priced off the buyer's credit rather than the supplier's, and the supplier ends up better funded than they would be carrying the extension alone. It works, and it drew enough accounting attention that both standard setters acted: the Financial Accounting Standards Board issued ASU 2022-04, and the International Accounting Standards Board amended IAS 7 and IFRS 7 in 2023, in each case to require disclosure of supplier finance arrangements. The concern behind both was that a payable financed through a bank programme can behave like borrowing while sitting in trade payables. Anyone proposing a terms extension as the working capital plan should expect that question from the auditor and have the answer ready.
Where the metric stops being useful
The cycle is computed from balance sheet snapshots, it aggregates every product and customer in the business, and it lags the decisions that produced it by a quarter or more. Mix moves it too. A shift toward a faster turning, lower margin category improves DIO without anyone having planned anything better.
Managing directly to it produces quarter-end behaviour. The patterns are recognisable once you know to look for them. Purchase orders due in the final fortnight get pushed into the next period, which flatters closing inventory and produces a shortage three weeks later. Shipments get pulled forward into the last days, moving stock from your balance sheet to a customer's, which improves DIO and damages DSO on a lag long enough that nobody connects the two. Buying stops, then over-corrects.
Baños-Caballero, García-Teruel and Martínez-Solano (2014), in the Journal of Business Research, found the relation between working capital investment and firm performance to be an inverted U rather than a straight line, meaning there is a level below which further reduction hurts performance. That result is worth having to hand when a target arrives expressed as a direction rather than as a level.
The cycle works well as a scoreboard. It works poorly as an instrument, because the instruments are the coverage decisions underneath it, taken item by item and node by node, and they respond to information a company-level percentage does not carry. For an importer there is a further component the cycle cannot see, since the order month is a currency decision as much as a demand one and has its own analysis. What an individual customer costs to serve, including the payment days they take, is a separate calculation again.
The first thing to check
Pull your last four quarter-end inventory values and compare each with the average of the two month ends either side of it. If quarter end is consistently the low point, part of the cycle you have been reporting is a calendar artefact, and that is worth establishing before anyone sets a target against it.