The Working Capital Lever No Customer Gets to Veto

The last two articles in this series stayed on the receivables side of working capital. One showed why your headline DSO can read seven days better after a single large order, with no customer paying any faster, and how a countback read fixes it ("Your DSO Is Lying to You"). The next showed why sending more reminders to everyone wastes the collection effort that a Pareto read would aim at the few accounts that actually move the number ("You're Chasing the Wrong Customers"). If you are new here, the series opens with the net working capital operating model that frames all three levers together.

Every receivables lever shares one catch. You can invoice the day you ship, set tight terms, keep billing clean, and chase hard, and the last move still belongs to a customer who pays when they decide to. So here is the question this article answers: which working capital lever has its final move sitting inside your own house, and why does the one lever you control most directly get a fraction of the attention?

The answer is days payable outstanding. The day your own money leaves the building is decided by your own team, on your own system. And most finance teams cannot tell you what their DPO really is.

The payables side you can act on

DSO and the aging report were about cash you influence. You shape the odds, the customer settles the bet. Days payable outstanding is the mirror image with one difference that changes everything. Once an invoice is valid and the term is agreed, the date the money leaves sits with you. Procurement sets the terms, treasury sets the policy, the ERP holds the settings, supplier health and the law set limits, so the lever is not unconstrained. But the final action, the day the cash moves, is yours.

So why does the lever you act on last get so little attention? Because effort follows visibility, not control. An overdue receivable shouts. It ages into a red bucket, lands on a report, gets raised in the monthly meeting. A payment that clears twenty days early says nothing. The supplier is content, the run cleared, no flag is raised. Finance ends up working hardest on the lever that complains and ignoring the one that stays quiet.

A reliable improvement habit fixes that. Separate what you control from what you only influence, and aim your effort at what you control. In Six Sigma language, you are separating controllable inputs from noise factors. On working capital, most teams have it backwards.

The data backs the reframe. The Hackett Group's 2025 US Working Capital Survey, covering the 1,000 largest US public nonfinancial companies, reported the cash conversion cycle improving by 4% to 37 days after a year of broad deterioration, and the rebound was led by payables, a 3% gain in DPO, rather than by collections. Receivables tells the opposite story. It is the single largest pile of trapped cash, an 18-day gap between top and median performers worth roughly $600bn. It is also the harder lever to move operationally, because the timing depends on other companies changing how they behave. The lever that delivered this year's improvement is the one that needs no customer's permission.

The lever you control most directly is the one nobody is watching.

What DPO actually measures, and what most people get wrong

Before you pull a lever you read the gauge. Here the gauge is usually misread, in a way nobody checks. Days payable outstanding measures how long, on average, your supplier invoices sit unpaid before you settle them. The standard formula divides average accounts payable by cost of goods sold, then multiplies by the days in the period. A $120K average payable on $1.46M of annual COGS reads 30 days. Quick, familiar, and good enough for benchmarking against peers. For managing payables inside the business, it is too blunt, in two ways.

Start with the numerator, because most people skip it. "Accounts payable" on the balance sheet can fold in trade payables to suppliers, intercompany balances, capex payables, taxes, GR/IR clearing, accrued liabilities, and supplier-finance balances. DPO only means something when the payable balance and the spend base describe the same population. Decide what you are measuring before you compute it: trade payables to suppliers in scope, or every payable on the ledger.

Then the denominator. Accounts payable comes from what you bought on supplier credit. COGS is the cost of what you sold, the materials, labour, and overhead that flowed into the units that left this period. Related flows, not the same one.

COGS leaves out the materials you bought and parked in inventory, and it folds in costs like labour and depreciation that never pass through accounts payable at all. The cleaner denominator is purchases, but which purchases depends on what you are measuring. For inventory-related trade payables, a rough fallback is COGS plus the change in inventory over the period. Use that one carefully in a manufacturer: inventory movements also carry direct labour, overhead absorption, production variances, and valuation effects that were never supplier-credit purchases, so the base stays partly polluted. The cleaner managerial base is the AP or procurement subledger: posted supplier-credit purchases from the suppliers in scope. For total supplier payables, which also carry freight, subcontracting, and services that never flow through inventory, the subledger is the only base that truly matches. The gap matters most exactly when you would notice least. In a quarter where you build inventory ahead of a season, COGS understates what you bought, so your DPO looks shorter and tighter than the terms your suppliers are actually carrying.

Payables come from purchasing, so the denominator should too.

The payables figures in this article come from a working capital model I rebuilt on GoFast.Finance, computing DPO three ways on the same set of books: COGS-based, purchases-based, and invoice-level. The three answers were more than ten days apart.

Read it the honest way: countback DPO

The denominator is the first problem. The average is the second. One number for a whole year quietly assumes your purchasing was flat across it. Few manufacturers buy evenly. You stock up before a launch, draw down in a slow quarter, ramp for a large order. A single average smears all of that into one figure that matches no actual month. Receivables has the same trap, which is why simple DSO can mislead, and the fix is the same on both sides.

It is called the countback, or exhaustion, method. Start with your ending accounts payable balance. Subtract last month's purchases. If a balance remains, subtract the month before, and keep going, counting the days as you walk back, until the payable is used up. The number of days you traversed is a truer operating read of DPO when purchasing is seasonal or growing. This is the same countback logic finance uses for DSO, turned onto payables as a management view. It is not an external benchmark formula. It is the quick internal read when purchases are uneven and invoice-level data is not ready yet.

The truest version is invoice-level, and the clock you pick depends on the question. For supplier behaviour, compare payment date to the contractual due date. For AP process health, measure invoice receipt to approval and scheduling. For working-capital timing, measure payment against the baseline date your terms run from. Run a due-date compliance split alongside whichever clock you choose.

Walk a countback through. Your ending payable is $4M. Last month you purchased $2.5M, the month before $2.2M. The full first month consumes $2.5M and 30 days, leaving $1.5M. Against the prior month's $2.2M, that $1.5M is about twenty more days. Countback DPO is roughly 50 days. Run the simple formula across a flatter annual average and you might read closer to 40. The ten-day gap is the seasonality the average hid. When the two methods diverge by more than five days, the average is hiding something. Dig into the gap, and manage by the countback view until you can read payment behaviour invoice by invoice.

Walk the balance back month by month and the seasonality stops hiding.

What the lever is worth, measured right

Once you read it honestly, size the prize, and size it on the right base. Take a manufacturer that buys $84M a year in goods and services, about $230K a day. Suppose its average payment lands on day 30 while the terms it already agreed run to day 45. On average, it releases its own cash fifteen days early for nothing in return. Move the average payment date out to the date already negotiated and you add roughly $3.4M to average cash. At a 10% cost of capital, preserving that $3.4M carries an economic value of about $345K a year, found without sending a single dunning letter. (These figures come from a sample case I modelled, not a single named company.)

Read that $345K correctly. It turns into visible P&L only if the freed cash lets you borrow less, avoid financing, or fund something you would otherwise have paid for. Until then it is opportunity cost recovered, not a line in the income statement. Two more cautions on the arithmetic, because anyone who checks it will catch them. Value the lever on supplier spend, never on revenue, since payables live on the purchasing base. And do not dollarize the gap between DSO and DPO directly, because the two sit on different denominators. Read that gap as a signal of timing pressure, not a figure you can bank. For the largest US companies DPO now runs near 59 days on average, and the strongest run higher. What you can reach depends on your industry, your supplier mix, and your terms.

Where does the early cash actually go? Some of it leaves early on purpose: to capture an early-payment discount, to protect a key supplier, to earn a return on a supply-chain-finance arrangement. Far more of it leaves early by habit. The payment run clears everything that has been approved, nobody schedules to the due date, and "on time" quietly gets read as "before the due date." A meaningful share of supplier invoices clear ahead of when they are owed, and in many manufacturing AP functions that share is higher still, because approval-to-pay cycles are short and runs go out weekly. The cash that leaves by habit is the cash you can recover first.

Most early payments are a default setting, not a decision.

A second read: when a high DPO is the warning, not the win

The first example showed a number that was too low for no good reason. Now take a number that looks too good. A different set of books, a mid-size components maker, reports DPO of 61 days, up from 52 two years earlier. On the dashboard that reads as progress, and the team gets credit for improving payables by nine days. Read it by supplier instead of in aggregate, and the story splits.

Most of the nine-day gain came from two places, neither of them discipline. The first is a reverse-factoring program a bank set up the year before. A slice of what now sits in trade payables is a supplier-finance balance, debt-like for working-capital analysis even when the accounting keeps it inside payables. Strip it out of operating DPO and the underlying number barely moved. The second is a cluster of small subcontractors now being paid twenty to thirty days past their agreed terms, because no one flagged them. Their invoices age quietly while the headline average climbs. The number went up. The risk went up with it: liquidity-disclosure exposure on one side, supplier goodwill draining on the other.

This is the failure mode the simple gauge cannot show you. In the first example, a rising DPO would have been pure win, recovered cash from closing a self-inflicted gap. Here a rising DPO is a flag. The direction of movement is identical. Only an invoice-level and supplier-level read tells the two apart, which is the same lesson the receivables side taught: the aggregate hides the composition, and the composition is where the truth lives.

The same DPO increase can be earned cash or hidden risk. The aggregate won't tell you which.

A higher number is not always a better one

That second case generalises. DPO can rise for good reasons: fewer accidental early payments, cleaner terms, better payment scheduling. It can also rise for bad ones: late payments, squeezed suppliers, disputed invoices left to age, or supplier finance arrangements parked inside ordinary payables.

The number on its own cannot tell those apart. Separating disciplined timing from cosmetic improvement is what next week is for. For now one rule holds: read what created the number before you treat it as a win.

A bigger DPO is only good news once you know how you got it.

Don't manage DPO alone

DPO tells you how long cash stayed. It cannot tell you whether you paid early, on time, or late. That is what a due-date compliance split is for: early, on-time, and late, measured by value, not by invoice count. APQC tracks the percentage of supplier invoices paid on time as its own AP process measure for exactly this reason.

A payables dashboard that shows only DPO will quietly let you miss discounts and stall approvals while the headline looks fine. Put four numbers next to it: invoice receipt-to-approval cycle time, the share of available early-payment discounts captured, the early/on-time/late split by value, and a supplier-finance-adjusted DPO. Together they tell you whether a higher number came from discipline or from drift.

One guardrail before you act on any of it. Do not aim the same DPO target at every supplier. Segment by criticality, financial fragility, the discount on offer, bargaining power, and supply risk. A blunt push costs you the suppliers you can least afford to lose.

View Use it for Basis Main risk External DPO Peer benchmarking Average AP / COGS x days Blunt, scope mismatch Purchase-based DPO Internal payables management Trade AP / supplier-credit purchases x days Needs a clean spend base Countback DPO Seasonal or growing purchasing Ending AP walked back against recent purchases Management view, not a benchmark Invoice-level lag Actual payment behaviour Payment date minus invoice, due, or baseline date Needs subledger data Due-date compliance Discipline vs abuse Early / on-time / late, by value Reveals whether DPO is healthy

Four objections you'll hear

1. "Stretching DPO just pushes the problem onto your suppliers." It does, if stretching is what you are doing. The move in the first example was paying to the terms already agreed, not a day past them, without breaching anything. Paying past terms to inflate the number is exactly the cosmetic rise this article warns against. The discipline and the abuse produce the same metric and opposite consequences, which is why you read the composition before you celebrate.

2. "Our ERP pays on the run date. We can't schedule to the due date." That is a payment-run configuration and cadence problem, not a constraint of nature. Payment terms held per vendor, a payment proposal filtered to invoices reaching their due date in the next run window, and a run cadence aligned to those dates are standard configuration in SAP and most major ERPs. The early payments are a default nobody turned off, not a rule you have to live with.

3. "DPO is a balance-sheet ratio for benchmarking. Invoice-level analysis is overkill." For comparing yourself to peers, the COGS-based ratio is fine and everyone uses it. The moment you try to manage payables rather than report them, the COGS denominator and the annual average feed you noise, and you end up steering by a gauge that swings on inventory builds and seasonality. Benchmark with the simple version. Manage with the honest one.

4. "Higher DPO always helps cash, so just push it up." Only when the rise is disciplined. Supplier finance and late payments both raise the number while raising risk, and a stressed key supplier or a disclosure problem costs far more than the cash you parked. The number is a means. The cash, and the supplier relationship that keeps the cash flowing, is the end.

Why this matters: read the gauge before you pull the lever

This is the payables side, where the last decision is yours, and it is the side most teams measure with a borrowed denominator and a flat average. Fix the gauge first. Scope the payable, compute DPO on purchases, run it back with the countback, and read it inside the full cash conversion cycle rather than on its own.

DPO is the only stretch of the cash cycle whose end date you set.

A number built the standard way can flatter you into thinking payables are already where they should be. Once the figure is honest, you are ready to use it well.

FAQ

• How do I get the AP or ERP team to schedule payments to the due date instead of the run date? Start with the cost, not the request. Pull a month of paid invoices, compare payment date to due date, and total the days of cash that left early. A single number, "we released roughly X of cash a median of Y days ahead of terms," moves a treasurer faster than a process argument. Then ask for one change: a payment proposal that selects invoices reaching their due date before the next run, with held vendors flagged. It is a filter on the run you already do, not a new system.

• What if procurement won't renegotiate supplier terms? Can I still improve DPO? Yes, and it is the better place to start, because it needs nobody's agreement. Closing the gap between when you pay and the terms you already hold is pure recovery, no negotiation involved. Renegotiating terms is a second, slower lever that touches supplier relationships and belongs to procurement. Earn the easy days first by paying to the terms you have.

• Does the countback method work for services and freight, or only for inventory purchases? It works for both, as long as the base matches the payables in scope. For inventory-related trade payables, the rough fallback is COGS plus the change in inventory, used with the manufacturing caveats above. For total supplier payables, which carry freight, subcontracting, and services that never touch inventory, use the AP subledger: posted credit purchases from the suppliers in scope. Mixing an inventory denominator with a total-payables numerator is the most common way the number comes out wrong.

• How do I tell a healthy DPO increase from a cosmetic one? Decompose it before you report it. Split the change into three buckets: fewer early payments to agreed terms (healthy), any supplier-finance or reverse-factoring balance sitting inside payables (treat it as debt-like and show an adjusted operating DPO, whatever the accounting presentation), and invoices paid past their due date (a risk flag, not a gain). If the increase survives once you strip out financing and late payments, it is real. If it does not, you have a dashboard improvement and a growing problem.

For two weeks the receivables side, DSO and the aging report, shared one catch: the customer finishes the job. This week we crossed to the payables side, where the final move is yours, and learned to read it honestly first.

Next week, how to use it well without managing the number instead of the cash, starting with the one moment when paying a supplier early beats every other use of that cash.

Ask This Next Month

Is our reported DPO built on COGS, on purchases, or on actual AP invoice flow, and would the answer change if we read it back month by month instead of as one yearly average?

Controller Move

First, scope the payable: confirm you are measuring trade payables to suppliers, not intercompany, capex, taxes, GR/IR, or supplier-finance balances mixed in. Then pull ending accounts payable and the last several months of purchases, not COGS. If the subledger total is not at hand and you are focusing on inventory-related trade payables, approximate purchases as COGS plus the change in inventory, knowing it still carries labour and overhead. Calculate DPO the simple way, then again with the countback: subtract each month's purchases from the ending payable, counting days until the balance is exhausted. If the two answers differ by more than five days, your headline DPO is hiding seasonality, and the countback is the one to manage by. Strip any supplier-finance balance out before you call an improvement real. Then read all three levers together as the cash conversion cycle, DIO plus DSO minus DPO, so payables sits next to inventory and receivables instead of being managed in isolation.

The Net Working Capital Operating Model course (CC-AA02) on GoFast.Finance turns this into a refreshable DPO, DSO, and DIO workbook, each computed on the right base, so you can read your own cash conversion cycle the honest way the month you finish it.

This article originally appeared in the Practical Lean Finance newsletter on LinkedIn. Read it here, or subscribe to the newsletter for the full weekly piece in your inbox.

Sources

1. The Hackett Group, 2025 US Working Capital Survey. Press release, "2025 Working Capital Survey: Payables Rebound, but Receivables and Inventory Lag," 18 August 2025. Analysis of the 1,000 largest US publicly traded nonfinancial companies. Confirmed figures used here: cash conversion cycle improved 4% to 37 days after a year of broad deterioration; the rebound was led by a 3% improvement in DPO, which stood at 59 days; $1.7 trillion remained in excess working capital; receivables was the largest excess bucket, about $600bn, tied to an 18-day DSO gap between top and median performers. The reading that DSO is the harder lever to move operationally, because collection timing depends on customer behaviour, is this article's interpretation, not a Hackett statement. URL: https://www.thehackettgroup.com/2025-working-capital-survey-payables-rebound-receivables-inventory-lag/

2. Days payable outstanding, standard formula. The conventional calculation is average accounts payable divided by cost of goods sold, multiplied by days in the period (Wall Street Prep, https://www.wallstreetprep.com/knowledge/days-payable-outstanding-dpo/ ; Allianz Trade, https://www.allianz-trade.com/en_GB/insights/protect-revenues/dpo-days-payable-outstanding-definition-formula-and-calculation.html ). The COGS denominator is the standard external presentation. The shift to a purchases or AP-subledger base is the managerial correction made here, since payables arise from purchasing rather than sales.

3. Cost of goods sold, composition. COGS is the direct cost of goods sold in the period and, for manufacturers, includes direct materials, direct labour, and manufacturing overhead (Investopedia, https://www.investopedia.com/terms/c/cogs.asp ). This is why COGS does not equal supplier-credit purchases and why the denominator correction matters.

4. Countback (exhaustion) method. Established on the receivables side as countback DSO: start from the ending balance and count back through recent flows until the balance is covered (Salesforce, https://www.salesforce.com/sales/revenue-lifecycle-management/days-sales-outstanding-dso/ ). Adapted here to payables as an internal management view when purchases are seasonal or growing. It is not a standardised external benchmark formula. Invoice-level payment lag remains the most precise reading.

5. Supplier finance arrangements, accounting treatment. Under IFRS, the IASB's May 2023 amendments to IAS 7 and IFRS 7 require disclosure of supplier finance arrangements (terms, carrying amounts, balance-sheet line items, ranges of due dates, liquidity-risk information) but do not mandate reclassification of every balance as debt (IFRS, https://www.ifrs.org/news-and-events/news/2023/05/iasb-increases-transparency-of-companies-supplier-finance/ ). Under US GAAP, FASB ASU 2022-04 adds disclosure requirements without changing recognition, measurement, or presentation on the face of the balance sheet or cash-flow statement (Deloitte DART, https://dart.deloitte.com/USDART/home/publications/deloitte/heads-up/2022/fasb-asu-supplier-finance-programs ). This supports the article's treatment: debt-like for working-capital analysis, presentation unchanged.

6. Accounts payable process metrics. APQC open-standards measures used as companions to DPO: percentage of supplier invoices paid on time ( https://www.apqc.org/what-we-do/benchmarking/open-standards-benchmarking/measures/percentage-supplier-invoices-paid-time ), cycle time from invoice receipt to approval and scheduling ( https://www.apqc.org/resources/benchmarking/open-standards-benchmarking/measures/cycle-time-days-receipt-invoice-until-0 ), and percentage of invoices paid within the discount period.

7. James P. Womack and Daniel T. Jones, Lean Thinking: Banish Waste and Create Wealth in Your Corporation. Background reading for the broader Lean habit behind this series: aim effort at what creates value rather than at activity that consumes effort without improving the outcome. View on Amazon.

As an Amazon Associate I earn from qualifying purchases. Book links above are affiliate links; using them costs you nothing extra and supports this newsletter.

#finance #controlling #FPA #workingcapital #accountspayable