Working capital is the chain, costed

Two and a half years into the hold, the deck is clean. Revenue is up. The value-creation plan is on track. Service metrics are in range, and EBITDA is at plan. Then the working-capital line. It is several million dollars heavier than at acquisition, and heavier than the plan cleanly explains. The CFO has an explanation for it, and the explanation is reasonable, and it is a different explanation than last quarter’s.

The line does not sit alone. Inventory turns have softened, and the inventory base resets a little higher each quarter; planning has added safety stock branch by branch to protect dates, a supplier minimum here that forced a larger buy, a position held there against a lead time the team learned not to trust, and nobody wants to unwind any of it without risking service. DSO has drifted by enough days to matter, and the drift is not in collections, it is in disputed lines and invoice exceptions that take longer to clear, an unmapped unit of measure on one account, a price the customer’s system rejected on another. The credit-memo line recurs every period and the backlog does not retire, each memo defensible on its own and the stack as a whole never getting shorter.

Three lines, one balance sheet, each with its own reasonable explanation. None of them explains why the same burden keeps showing up somewhere else. And each is a rational, defensible response: the planner who cannot trust the date is right to hold the stock, the customer whose invoice does not match the order is right to short-pay, the clerk who writes the memo is right to write it. The business is right to be carrying what it carries.

That is the balance sheet the partner cannot reconcile to the story, and the one the CFO has been managing as three related issues. They are not three lines to manage down. They are one thing, showing up in three places.

The balance sheet is the chain, costed

The reflex is to read the three lines as three finance problems with three owners: a turns problem on inventory, a collections problem on AR, a clean-up problem on the memo backlog. Read that way, each gets its own program and its own target, and the line resets higher next period anyway.

The three lines are the same distrust, expressed at three points in one cycle. Defensive inventory is distrust at the front of the chain: will supply arrive when the date said it would. Stretched receivables and the credit-memo backlog are distrust at the back: will the invoice match what was promised, will the commitment hold. Suppliers learn the account and the front of the cycle thickens; customers learn the account and the back of it stretches; the business finances its own distrust at both ends at once. That is why both ends thicken together, and why no single line owner can explain the whole. The inventory sits in supply chain, the receivable sits in finance, the memo sits in billing, and the thing that produced all three sits upstream of every seat that holds a piece of it.

This is also why the most common reading misses, and it is worth naming because it is usually already on the value-creation plan: that this is a treasury problem, a working-capital position to optimize back down. A treasury program can sweep cash, net terms, and factor receivables, and the base resets higher the next period, because the reason for it, the records the chain cannot trust, is untouched. Managing the line from the finance seat treats the symptom as the cause. The CFO is reading the balance sheet correctly. The line just does not originate where it sits.

One line, traced

The way to see this is not to argue it. It is to take one line and follow the money.

Take it one step at a time, the way the partner would, from the chair, with the CFO confirming the operating reason at each turn.

The line first. Turns have softened, and the inventory base is several hundred basis points heavier than the plan assumed and resets higher each quarter. The partner reads this in the monthly pack. It is the first thing diligence reconstructs, and it is the line the deck opened on.

The safety stock next. Ask why the base is heavy, and planning explains it account by account and branch by branch: stock added to protect dates, positioned defensively, nobody willing to unwind it without risking service. This is the defensive inventory the partner has heard named in a list of where the burden gets carried. Here it is, on one line, costed.

Then the two numbers that disagree. The branch holds a stock position the central system does not match. The ATP figure on the screen is close. It is not close enough to commit a customer’s date to, so the branch manager counts what is physically on the floor and answers from that. The number on the screen is approximately right and operationally insufficient, and the branch stock file exists because that is true often enough to make the file the safer place to look. The branch knows things the central record does not. The partner can see the file exists; the CFO confirms the floor number and the system number do not agree.

Then the card. The date the business quotes and plans against does not come from the floor and does not come from the supplier in real time. It comes from a standing supply card, a lead-time field the rep prices from and planning schedules against. The CFO can pull the field.

Then the decision. The field was entered once, against a supplier relationship that has since changed, and never refreshed. No one owns refreshing it. Ask when it was last reviewed and there is no clean answer, because reviewing it was never anyone’s task. That single uncorrected record is the operating decision the whole line traces back to. One un-maintained field silently sets the date the whole chain then defends, and it defends it with capital.

Read down the left column. That is the capital, sitting on the balance sheet. Read down the right. That is where it was created. The dollars are real, the line is real, and it was created upstream by an operating decision, an unmaintained record, not by a finance one. You cannot manage the line down from the finance seat, because the finance seat is not where the line was made.

This is also why leaning the inventory does not hold. Cut the safety stock and lift turns to a target, and either service breaks or the stock comes straight back, because the date it hedges against has not changed. The safety stock is not the waste a turns target treats it as. It is what a planner does with a date they cannot trust. You cannot lean out a hedge while the thing it hedges against is unchanged. Make the date trustworthy and the stock falls because the reason for it is gone, not because a target was set.

The same dynamic, the other side of the cycle

The inventory line got the worked trace because one line proves the pattern. The other two trace the same way, to the same kind of decision, on the back of the cycle.

The receivables stretch because the invoice cannot match what was quoted. An unmapped unit of measure or a price mismatch makes the customer’s AP system reject a line, the customer short-pays, and the money sits in disputed aging while a clerk hand-walks a credit memo to clear it. The dollars are in AR; the operating reason is upstream of finance; the root is a record the customer is right not to trust. This is why tightening DSO does not move it. A collections program collects faster on invoices the customer agrees with, and it does nothing to the invoice the customer is right to reject. The receivable is not aging in collections. It is aging in dispute, and the dispute was created before the invoice was ever cut.

The credit-memo backlog does not retire because each memo closes a dispute without writing the rule that produced it. The memo settles the line and leaves the exception in place, so the next order produces the next memo, and the backlog refills as fast as it clears. The commitment was made before the chain could carry it, which is why the invoice could not match the quote in the first place. That commitment is named here as the cause and left there. It is the hinge the rest of the site develops, and on this balance sheet it does one job: it is the reason the dollars sit in disputed aging and in the memo backlog instead of in collected cash.

Capital that ratchets

The line does not break. That is the part that keeps it on the balance sheet.

At a smaller scale, a defensive position was a one-time hedge. At this scale it is a standing layer that grows with the business. A supplier minimum converts a single sale into stranded stock, and the stranded stock joins the base. The position grows at the front of the cycle and at the back of it at once, and the standing layer is larger every year it goes unaddressed.

The second face is the partner’s timing question, answered structurally. The burden compounds roughly in line with the business. The remedy does not. The defensive layer becomes load-bearing, because planning now runs on it, and unwinding it later, with more branches and more accounts matured against it, asks more than unwinding it now. The cost of acting rises faster than the cost of waiting.

There is a quieter reading here that says the line will normalize as the business scales, or once the system project goes in. It does not normalize, it ratchets, because each addition is defensible and none is audited out. And a system project that migrates the same untrusted records runs the same logic faster, so the defensive positions reappear around a cleaner screen. The records do not get truer in the move; they get faster.

The reliable thing that forces the line into view is an outside party with incentive to find it. This is the working-capital line diligence reconstructs first, and the distance between the reported intensity and the operating reason is widest in roll-ups, where each add-on arrives with its own masters and its own defensive positions before the last is integrated. The EBITDA quality reading travels with it. But the line itself, the capital, is the object here.

Capital you can recover, not a line you can manage

So the decision is not the one the line invites.

The question is not how to bring working capital down. It is which of those records the business is prepared to make true, so the next order does not need the inventory, the dispute, or the memo to clear. None of this is a working-capital program. It is the work that comes before one. The line was created upstream, and it can only be released upstream. Finance does not control the cash here. The chain does.

The proof that the work happened is not a working-capital target hit and is not a new dashboard. It is the compensation retiring. The safety stock comes down because the supply card carries a date planning trusts, not because a turns target was set. The disputed aging falls because the invoice matches the quote and the dispute does not arise. The credit-memo backlog stops being refilled because the rule that produced the memo was finally written. The line falls because the reason for it is gone.