The Operator’s Guide to Working Capital

• What Working Capital Actually Is, in Operating Terms
• The Cash Conversion Cycle, and How to Read Yours
• Why It Is One System and Not Three Initiatives
• Diagnosis: Where Is Your Cash Actually Sitting?
• Inventory: The Largest Pool and the Hardest Conversation
• Receivables: A Discipline Problem Before It Is a Credit Problem
• Payables: The Easiest Lever and the Most Expensive One
• What Changes When the Cash Is Yours
• The Sequence That Releases Cash Without Breaking Service
• How to Keep the Cash From Going Back In
• What to Do With the Cash Once It Is Out
• Frequently Asked Questions
• About the Stagnation Assassin

Working capital is the money your business has already spent but has not yet got back. That is the operator’s definition, and it is more useful than the accounting one. The accounting definition, current assets minus current liabilities, describes a position on a balance sheet at a moment in time. The operator’s definition describes a queue: cash left the building to buy raw material, it is now sitting in a rack or a work order or an unpaid invoice, and at some point it will come back as cash again. The only question that matters is how long that round trip takes and how much cash is stuck inside it while you wait.

Most manufacturers are financing a substantial amount of that queue with debt, and paying for the privilege. The money is already theirs. They are renting it back.

This guide covers the whole of it: inventory, receivables, payables, and the cycle that connects them, treated as one system rather than as three departmental projects. It is written for the person who has to release the cash without taking down the service level, which is the constraint that makes this hard and the reason most working capital programs stall in month four.

What Working Capital Actually Is, in Operating Terms

Working capital in a manufacturing business is cash held hostage by the operating cycle. It sits in three places: inventory you have bought but not sold, invoices you have issued but not collected, and, offsetting both, invoices you have received but not yet paid. What is left over is what you are funding yourself.

The three accounts, and what each one really represents

Inventory is a decision about uncertainty. Every unit of stock above what is being consumed today exists because somebody, somewhere, decided that the risk of not having it was worse than the cost of holding it. Usually that decision was never made explicitly. It accumulated.

Receivables are a decision about terms, and then a decision about whether anyone enforces them. The first decision is commercial. The second is administrative, and it is where most of the money is.

Payables are a decision about whose balance sheet carries the float. They are also the only one of the three where the cash you release comes directly out of somebody else’s business, which is why they need to be handled last and handled carefully.

Why the balance sheet version misleads operators

The balance sheet shows you a number. It does not show you a duration, and duration is the thing you can actually manage. Two businesses with identical working capital balances can have wildly different problems: one is holding six months of slow-moving finished goods, the other is collecting from good customers forty days late. The balance is the same. The fix is nothing alike.

This is the single most common reason working capital initiatives get assigned to the wrong function. Finance sees a balance and owns the balance sheet, so finance gets the project. But inventory duration is set by the schedule, the mix, and the changeover profile, all of which live in operations, and receivable duration is set by invoice accuracy and collection cadence, which usually live in three functions at once. Give the whole thing to the controller and you will get a report, not a release.

The Cash Conversion Cycle, and How to Read Yours

The cash conversion cycle is the number of days between paying your suppliers and collecting from your customers. It is calculated as days inventory outstanding plus days sales outstanding, minus days payables outstanding. It is the only working capital number that tells you a duration rather than a balance, which makes it the one worth putting on a scorecard.

What each number is telling you

Days inventory outstanding is a statement about how much uncertainty you are carrying and how much of your production is running to forecast rather than to order. In a manufacturing business it splits three ways: raw material days, work in progress days, and finished goods days. Look at the split before you look at the total, because the three have completely different causes. High raw material days is a purchasing and supplier lead time problem. High work in progress days is a flow problem, usually batch sizes or a bottleneck. High finished goods days is a forecasting and mix problem, and it is where obsolescence eventually lands.

Days sales outstanding is a statement about your commercial terms and, more often, about your own back office. Before you conclude that customers are slow, find out what percentage of your invoices go out with an error on them. That number is usually higher than anyone in the room believes and it is the cheapest thing on this entire list to fix.

Days payables outstanding is the one to read with suspicion. A rising number can mean you have negotiated better terms. It can also mean you are quietly late, your suppliers have noticed, and your input prices are about to reflect that.

The balance sheet shows you a number. It does not show you a duration, and duration is the only part of working capital you can actually manage.

Why It Is One System and Not Three Initiatives

The standard approach is to launch three workstreams: an inventory reduction project in operations, a collections push in finance, and a terms renegotiation in procurement. Each gets a target. Each reports separately. And the cash released is reliably less than the sum of the three targets, because the three accounts are connected and the projects are not.

Consider what actually happens. The inventory project cuts safety stock. Service levels wobble. Customers start disputing short shipments, disputed invoices do not get paid, and days sales outstanding rises. Finance responds by pushing harder on collections, which lands on the same customers whose orders are now arriving incomplete. Meanwhile procurement has stretched supplier terms, so suppliers have shortened lead times commitments or gone to allocation, which forces operations to carry more raw material to stay safe. Net release: close to nothing, and three functions now have evidence that the other two are the problem.

This is not a failure of execution. It is what happens when you run three optimizations against a system that has one constraint. The constraint in almost every manufacturer is the same: the business is carrying inventory to compensate for variability it has not removed. Attack the variability and inventory falls on its own, service improves, disputes drop, and receivables shorten as a side effect. Attack the inventory directly and you are removing the buffer while leaving the variability fully intact.

Which is why the diagnosis has to come first, and why it has to be one diagnosis.

Diagnosis: Where Is Your Cash Actually Sitting?

Before deciding what to do, establish which of three profiles you are. Pull twenty four months of monthly data for the three components of the cycle, and look at the trend and the split rather than the current balance. Most businesses are recognizably one of these.

The inventory-heavy profile

Days inventory outstanding dominates the cycle, finished goods is the largest of the three inventory buckets, and the slow-moving tail is substantial. This is the most common profile in mid-market manufacturing and it is almost always a mix problem wearing an inventory costume. The business has accumulated products, configurations, and part numbers faster than it has retired them, and each one carries stock.

The tell: ask how many active part numbers had zero movement in the last twelve months. If nobody can answer that within a day, you have found both the profile and the reason it persists.

The receivables-heavy profile

Inventory is reasonable, payables are normal, and days sales outstanding sits well above the stated terms. This profile is usually administrative rather than commercial. Somewhere between the shipment and the payment there is a document that is wrong, late, or going to the wrong person, and nobody owns the handoff.

The tell: a large share of the overdue balance is concentrated in a small number of invoices, and each of those has a dispute attached to it. Aged debt in a healthy business is rarely about customers who cannot pay. It is about invoices that cannot be approved.

The payables-financed profile

The cycle looks respectable, but only because payables days are stretched well beyond terms. This is working capital that has been borrowed from suppliers without asking, and it is the most fragile position of the three, because it can be withdrawn by other people on short notice. Suppliers respond to it with price, with lead time, or with credit holds, all of which arrive at the worst possible moment.

The tell: the cycle improved significantly last year and nobody can point to an operational change that caused it.

Inventory: The Largest Pool and the Hardest Conversation

Inventory is usually the biggest recoverable pool and the one with the most organizational resistance, because every unit of stock is protecting somebody from a conversation they do not want to have. The plant manager’s stock protects against a schedule that changes weekly. The sales team’s finished goods protect against a promise made to a customer. Nobody is being unreasonable. They are all buffering a real problem that lives somewhere else.

Raw material, work in progress, finished goods

Split the pool and treat each part separately. Raw material responds to supplier lead time, order frequency, and the number of distinct inputs you have chosen to carry. The fastest wins are consolidation of duplicate items across plants and more frequent ordering on the highest value inputs, which costs a little in freight and returns a lot in cash.

Work in progress responds almost entirely to batch size and flow. If work in progress days are high, you have large batches, long changeovers, or a bottleneck that everything queues behind. Cutting work in progress without addressing the changeover profile just moves the queue.

Finished goods is where the mix problem shows up as cash. The top of the range turns quickly and is rarely the issue. The tail holds stock indefinitely against demand that arrives unpredictably, and it is subsidized by the products that are working.

The service level question, answered honestly

Every inventory reduction meets the same objection: we will miss shipments. Sometimes that objection is right. The way to settle it is to stop discussing inventory in aggregate and start setting a service level per item class. Your top items get high availability, deliberately and expensively. The tail gets made to order with a stated lead time, or gets retired.

What you cannot do is run a uniform service target across the whole range. That is the decision that creates the inventory problem in the first place: it requires you to hold stock against your least predictable demand at the same standard as your most predictable, and the cost of that lands in cash rather than in the P&L, which is exactly why it survives scrutiny for years.

Receivables: A Discipline Problem Before It Is a Credit Problem

Receivables are where the quickest cash sits, because fixing them requires no capital, no supplier negotiation, and no change to the production schedule. It requires somebody to care every single week, which turns out to be the hard part.

Start with your own paperwork

Before any collections effort, audit the invoice itself. Wrong purchase order number, wrong entity, wrong address, quantity mismatched against the delivery note, missing supporting documentation the customer’s accounts payable system requires. Each of those puts an invoice into a queue where it will sit silently until somebody chases it, and the clock on your days sales outstanding is running the entire time.

An invoice that cannot be approved is not a receivable. It is a task nobody has been assigned.

Cadence beats escalation

The second fix is rhythm. A weekly review of the aged debt with named owners and a specific next action per account will do more than any escalation policy, because most late payment is drift rather than refusal. The account that gets a call on day thirty two pays. The account that gets a stern letter on day ninety negotiates.

Where a genuine credit problem exists, deal with it as a credit problem: limits, holds, and, if necessary, a commercial decision about whether that revenue is worth the risk. But diagnose in that order. Treating a paperwork failure as a credit failure damages a relationship with a customer who was always going to pay.

The billing structure nobody revisits

On longer lead time or engineered products, the largest receivable improvement available is usually not collection speed at all. It is when you bill. Deposits at order, progress billing at defined milestones, and payment on shipment rather than on installation are all commercial terms that were set once, often by a predecessor, against a market that has since changed. They are rarely revisited because nobody owns them, and because raising them feels like a price conversation.

It is worth having anyway. Moving from invoice-on-completion to a deposit plus a milestone on a product with a long build cycle removes weeks of funding from your side of the cycle in a single decision, and on engineered work it is normal enough that customers rarely resist it. Check what your competitors ask for before you assume yours will.

Payables: The Easiest Lever and the Most Expensive One

Stretching payables releases cash immediately, requires no operational change, and shows up in the next month’s reporting. It is also the only one of the three levers where the cash comes out of a partner’s business, and industrial supply relationships have long memories.

The distinction worth holding: negotiating terms is legitimate, and paying late without telling anyone is not. The first is a commercial conversation in which the supplier gets something, usually volume, forecast visibility, or a longer commitment. The second is a unilateral transfer that gets repriced into your input costs within two quarters, usually with a lead time penalty attached that will then force you to carry more raw material. You will have funded a payables improvement with an inventory increase and called it a win.

Use this lever last, use it openly, and use it hardest where you have real leverage and no dependency. On a single-source supplier of something critical, leave it alone entirely. The cash you free is not worth the position it puts you in.

Paying late without telling anyone is not a working capital strategy. It is a loan taken from a supplier who did not agree to make it, at a rate they set later.

What Changes When the Cash Is Yours

I ran this playbook inside large organizations for years before I bought a small industrial manufacturer and ran it myself. The mechanics did not change. The speed of learning did, sharply, because the cash in the queue was my cash, and every day of the cycle was a day I was funding personally.

Three things looked different from that seat.

The first is that inventory stops being a number and starts being a physical fact you walk past. In a large business, slow-moving stock is a line in a report reviewed by somebody who has never seen it. When you own the building, you see the same pallets in the same corner every morning, and the question of what they are for gets answered much faster.

The second is that receivables become a daily matter rather than a monthly one. There is no treasury function absorbing the timing difference. If a large invoice slips two weeks, you feel it in the decisions you can make that month, which produces a level of collection discipline that no policy document has ever produced in a larger company.

The third is the one that took longest to learn: most of what looked like a cash problem was a complexity problem. The stock existed because the range was too wide for the volume behind it. Narrowing the range released more cash than every collections and terms initiative combined, and it also made the business easier to run, easier to schedule, and eventually easier to sell.

Todd Hagopian has spent a career on cost and cash inside Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, but the sharpest lesson on working capital came from a much smaller business where the consequences were immediate and personal.

The Sequence That Releases Cash Without Breaking Service

Order matters more than effort here. Run these in sequence, not in parallel, and do not start the next one until the previous one is holding.

1. Fix the invoice. Audit a sample of recent invoices against what the customer’s system requires. Correct the template, the data, and the handoff. This is days of work and it starts shortening days sales outstanding within one billing cycle, with no risk attached to it.
2. Install the weekly cash review. One meeting, thirty minutes, three numbers and an aged debt list with named owners. The meeting is the mechanism. Without it, everything below decays in about two months.
3. Segment the range and set service levels by class. Decide, explicitly, which items you hold stock for and which are made to order. This is the decision that unlocks the inventory pool without touching the service level that matters.
4. Attack the variability behind the buffer. Changeover time, schedule stability, supplier lead time reliability. Every hour removed here is stock you no longer need to hold, and it does not come back.
5. Retire the tail. Obsolete and zero-movement stock, sold, scrapped, or written down, and the part numbers behind it closed so they cannot regenerate. Put a gate on new part numbers at the same time or the tail regrows within eighteen months.
6. Renegotiate terms last, and openly. With the operational work done, you are negotiating from a position of predictability, which is exactly what a supplier wants in exchange for terms.

The reason receivables come first is not that they are the biggest pool. They are usually not. It is that they are the fastest, the safest, and the release buys you the time and credibility to do the slower inventory work properly instead of hacking at it under deadline pressure.

How to Keep the Cash From Going Back In

Working capital is unusual among cost and cash programs in how reliably it reverses. Inventory regrows quietly, one reasonable decision at a time, and days sales outstanding drifts back the month after the weekly meeting gets cancelled. Three things hold it.

Put duration on the scorecard, not balance. A balance target can be hit at year end and released in January. Days inventory outstanding, days sales outstanding, and the resulting cycle, reviewed monthly, cannot be gamed the same way.

Give the cycle a single owner. Not finance, not operations, but one named executive accountable for the whole cycle, because every genuine improvement crosses at least two functions and anything that crosses two functions and belongs to neither will not survive a busy quarter.

Gate the things that create the need. New part numbers, new customer terms outside standard, and new supplier lead times all add cash to the queue. If those three gates exist and are enforced, the cycle stays where you put it. If they do not, you will run this program again in three years.

The INSEAD work on this is worth reading in full, particularly on the way managing to the income statement quietly encourages executives to tie up cash in stock and receivables: Need Cash? Look Inside Your Company. It is the clearest short statement of why the incentive structure, not the effort level, is what usually keeps the cash trapped.

What to Do With the Cash Once It Is Out

This is the part of working capital work that almost never gets planned, and it is the reason a successful program can still leave the business exactly where it started. Cash released from the operating cycle is the cheapest capital a manufacturer will ever raise. It carries no interest, no covenant, no dilution, and no approval process. It is also, for precisely those reasons, the easiest capital in the business to spend badly.

The default destination is the revolver, and in a leveraged business that is often correct. Paying down the most expensive money you have is a defensible use of the cheapest money you just found, and it is the only option that is impossible to get wrong. But it should be a decision rather than a reflex, because a business that has just shortened its cycle by a meaningful margin has also permanently lowered the amount of funding its operations require, and that structural change is worth more than the one-time release.

The trap is the second pattern: the cash gets released, nobody decides what it is for, and it is absorbed over the following two quarters by decisions that would not have survived a funding request. A piece of equipment nobody built a case for. An inventory build for a launch that was not going to happen at that volume. Headcount that was declined in the budget cycle and then approved informally because the money was suddenly there. None of these individually looks like a failure. Collectively they are how a genuine cash release turns into no change in the balance sheet twelve months later.

Decide the destination before the release

The fix is administrative and takes one meeting. Before the program starts, agree what the released cash is for and write it down: debt reduction to a stated level, a specific capital project with a stated return, or a stated cash balance the business is going to hold and not touch. Then report the release against that destination rather than against the working capital target.

This also solves a political problem that otherwise kills the second round. Working capital programs ask operations to accept risk, commercial to accept constraint, and procurement to accept a harder conversation with suppliers. If the reward for all of that is an unexplained improvement in a balance sheet nobody in those functions is measured on, you will not get a second round out of them. If the reward is a piece of equipment the plant has wanted for three years, funded by cash the plant released, you will.

Cash released from the operating cycle is the cheapest capital a manufacturer will ever raise, which is exactly why it is the easiest to spend without a decision.

Frequently Asked Questions

What is a good cash conversion cycle for a manufacturer?

There is no universal number, and anyone who gives you one is selling something. Manufacturing cycles are structurally longer than retail or service because production lead times sit inside them. The useful comparison is against your own trend and against direct sector peers with similar business models, not against a cross-industry average.

Should I reduce inventory or improve collections first?

Collections first, almost always. It is faster, it carries no service risk, and a large share of the delay is usually your own invoicing rather than customer behavior. Inventory is the bigger pool but it needs the range and service level decisions made first, which takes a quarter or more.

Does reducing working capital hurt customer service?

It does if you cut buffers without removing the variability the buffers exist for. It does not if you set service levels by item class and attack changeover time, schedule stability, and supplier reliability first. The order is what determines the answer.

Is stretching supplier payment terms a good way to free up cash?

Negotiated terms are legitimate and often underused. Paying late without agreement is not, and it gets repriced into your input costs and lead times, frequently forcing you to carry more raw material than the payables improvement released. Use it last and use it openly.

Who should own working capital in a manufacturing business?

One named executive with authority across operations, commercial, and finance. The most common failure is assigning it to the controller, who can measure the balance but does not control the schedule, the mix, or the terms that actually determine it.

So, to return to the question this guide set out to answer: working capital reduction in manufacturing works when you treat inventory, receivables, and payables as one cash system with one owner and one sequence. It fails when you treat them as three targets handed to three functions, because the three will release cash into each other’s problems and the business will net close to nothing.

If you want the range and mix side of this examined before you touch inventory, that is exactly what the 15-minute 80/20 Portfolio Performance Audit is for. It is free and it is the fastest way to find out whether your cash problem is actually a complexity problem. Request one here.

About the Stagnation Assassin

Todd Hagopian is a Fortune 500 transformation executive who has generated $3B in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as Global VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he is the author of two published books: The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.

See also: Why Across-the-Board Cuts Always Fail, Where Should You Cut Costs First?