You asked whether you have a cost problem or a demand problem. The more precise version of that question, and the one that actually decides what you do on Monday, is this: are you failing to win the work, or are you winning the work and failing to convert it profitably? Those are two different businesses with two opposite responses, and the reason this matters more than almost any other call you will make in a recovery is that one of the two responses is not reversible.
What is the actual difference?
A demand problem means the orders are not there: you are quoting and not winning, or the segment itself has contracted. A cost problem means the orders are there and you cannot convert them at an acceptable margin. Same shortfall in the P&L, opposite cause, opposite fix.
The reason the two get confused is that they produce identical financial symptoms in the month you first notice them. Revenue is below plan and margin is below plan. Everything after that point depends on which direction you read the causation, and the P&L will not tell you, because a P&L reports outcomes rather than mechanisms. You have to go get the mechanism yourself.
Which one do the signals point to?
Four signals separate them reliably: quote win rate, the size and age of the backlog, past-due orders, and whether the constraint is starved or saturated. Read all four together. Any one of them in isolation can be produced by either cause, which is exactly how operators talk themselves into the wrong answer.
| Signal | Demand problem | Cost problem |
| Quote win rate | Falling, or quote volume itself is down | Stable or rising |
| Backlog | Thin and getting thinner | Healthy or growing |
| Past-due orders | Few, because there is capacity to spare | Long and aging |
| The constraint | Starved, waiting on work | Saturated, or switching rather than running |
| Where the loss shows | At the top line first | Between revenue and gross margin |
Two mixed readings are worth naming. A thin backlog with a saturated constraint means you have a lead time problem that is destroying demand, which behaves like a cost problem and gets solved like one. A healthy backlog with a starved constraint means you have a planning or material availability problem, which is neither of the two and is usually the cheapest thing on this page to fix.
Why does the standard answer get this wrong?
The standard answer is to cut cost first because cost is controllable and demand is not. It is the counsel of every board in a bad quarter and it is correct roughly half the time. The other half it takes a recoverable business and makes it permanently smaller, which is why it deserves more scrutiny than it usually gets.
Here is the mechanism. Cost cuts applied to a demand problem remove capability faster than demand returns. You surrender the second shift, lose the operators who qualified on the constraint, defer the maintenance window, and shrink the quoting capacity. Demand recovers eight months later and you cannot serve it, so you quote longer lead times, lose the work to someone who kept their capacity, and confirm the original diagnosis by producing the evidence for it.
That loop is self-sealing, which is what makes it dangerous. The plant ends up profitable at a volume it will never grow past, and everybody involved can point to the numbers that justified each decision. This is also why the decision deserves a genuine fork rather than a default, and why speed on the wrong call is worse than a week spent getting it right. Harvard Business Review’s piece on making great decisions quickly is right that most executives are too slow. This is one of the few calls where being fast and wrong costs more than the delay.
What if it is both?
Sometimes it is both, and when it is, the sequence still matters: fix the cost problem inside the work you are already winning before you touch capacity. Margin recovery on existing volume funds the commercial work. Capacity reduction forecloses it.
The practical version is that you can pull every 30-day margin lever without reducing capability. Discount drift, premium freight, expedite fees, scrap on the two worst part numbers, unscheduled overtime. None of those removes a machine, a shift, or a qualified operator. They are the moves available to you before the fork is even resolved, which means there is no reason to wait on them and no reason to confuse them with the structural decision.
What should you do in the next ten days?
Collect four numbers yourself: quote win rate by product family for the last four quarters, backlog in weeks of capacity, past-due orders by age, and constraint utilization measured hourly for one full week. Ten days is enough. Do not commission a study.
Measure the constraint yourself rather than reading it from a report, because utilization figures in underperforming plants are frequently calculated on scheduled hours rather than available ones, which flatters a saturated line and hides a starved one. If you want the fuller diagnostic that surrounds this question, it is cause three in why manufacturing plants underperform, and the recovery sequence that follows the call is in how to turn around a manufacturing plant.
A limitation worth stating plainly: this diagnostic assumes a single plant with a coherent product set. In a multi-site business, or one with two genuinely different segments under one roof, the four signals will average into a mush that reads as neither. Run it separately by site and by segment, or it will tell you nothing.
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.
If the four signals point at cost, the next question is which products are carrying it. A 15-minute 80/20 Portfolio Performance Audit will tell you where your margin actually lives, and which part of your line is being subsidized by the rest of it.