When I was running a manufacturing operation scaling from zero to $75 million in revenue, I had a finance team, operations managers, and a warehouse crew. We had systems. We had tracking. And we still had a COGS line that, when I finally broke it apart, contained about $400,000 a year in costs that had no business being there.
Not fraud. Not negligence. Just the accumulated cost of small inefficiencies — material waste that got normalized, rework that got counted as "part of the process," over-ordering that turned into shrinkage that turned into write-offs. Each individual problem was small enough to ignore. Together, they were hemorrhaging margin.
That's The Bleed — one of The 8 Profit Leaks I diagnose in every business I walk into.
Why COGS Looks Fixed When It Isn't
Here's the problem with how most P&Ls are structured: Cost of Goods Sold is one line. One number. It tells you what you spent to produce what you sold — but it doesn't tell you how much of that spending was necessary.
Your accounting system doesn't distinguish between:
- Materials that went into finished product
- Materials that went into the trash
- Labor that built something you sold
- Labor that rebuilt something you built wrong the first time
All of it lands in the same bucket. And when revenue is growing — when the business feels like it's working — nobody digs into the bucket. The COGS percentage gets normalized. "That's just what it costs to make our product." And the bleed continues.
Where The Bleed Actually Hides
I've diagnosed enough operations to know the patterns. These are the five places I look first:
1. Material Yield Loss
What percentage of your raw materials actually make it into finished product? If you're buying 1,000 units of material and 850 end up in product, your theoretical yield is 85%. Every point below your target yield is money you paid for something you threw away. In food, agriculture, CPG, and manufacturing, this number is frequently 5–15% worse than operators believe. This is why The Bleed is the second leak in the diagnostic framework — it's where the fastest money hides.
2. Rework and Quality Failures
Every unit that fails QC and gets reworked costs you twice — once to build it wrong, once to fix it. Rework doesn't show up as a line item in most systems. It shows up as elevated labor cost and elevated materials consumption. Track rework rate separately. If you haven't, you don't know what it's costing you.
3. Vendor Pricing Drift
You negotiated a price with your supplier 18 months ago. Since then, they've had three quiet price adjustments — a fuel surcharge here, a materials surcharge there. Each one was small enough that nobody caught it. Your contracted price and your actual price have quietly diverged. I've seen this account for 1–3% of COGS in businesses that think they're on top of their vendor relationships.
4. Over-Ordering and Inventory Shrinkage
Buying more than you need to get the volume discount feels like good procurement. But if the excess sits in inventory, degrades, becomes obsolete, or walks out the door — that "discount" cost you more than it saved. Shrinkage rarely gets attributed to its real source. It gets buried in COGS adjustments at the end of the period.
5. Inefficient Labor Deployment on Direct Costs
Direct labor is a COGS component. Labor doing production-adjacent work that isn't directly contributing to output — waiting on materials, managing avoidable bottlenecks, covering for poor scheduling — is still coded to COGS. It's real cost for output that never happened.
Unlike new revenue — where only your net margin % hits profit — every dollar you recover from COGS goes straight to the bottom line. A $2M business recovering 3% of COGS from The Bleed adds $60,000 in annual profit without acquiring a single new customer. (Run your own numbers with the Profit Leak Calculator →) That's not a rounding error. For most operators, that's meaningful cash.
How to Measure Your Bleed
The diagnostic starts with one question: What should your COGS be? If you don't have the answer, the Profit Pressure Test™ will give it to you in 45 minutes.
Most operators know their actual COGS percentage. Almost none of them know their theoretical COGS — what it would be if materials yielded at standard, labor ran at standard efficiency, rework was zero, and vendor pricing matched contracts.
Build that theoretical number. The gap between theoretical and actual is The Bleed. If you don't know how to build the theoretical number, that's the first thing I'd work through with you.
The businesses that manage COGS tightly don't have dramatically different operations — they just measure what others ignore. They know their yield rate. They track rework. They audit vendor invoices against contracts. The information is there. The question is whether anyone is looking at it.
What To Do With It
Once you know the size of The Bleed, you prioritize by impact-to-effort ratio. Some fixes are cheap and immediate — a vendor invoice audit can take a day and recover tens of thousands of dollars. Others require process redesign and take months. Start with the cheap wins. Build the data. Then go deeper.
The pattern I see repeatedly: operators who go through this exercise for the first time are surprised not by the size of The Bleed, but by how long it's been there. In most cases, the waste has been running for years — normalized, invisible, and never formally measured.
Finding it is usually the hardest part. After that, fixing it is mostly a matter of deciding to.
