6,000 SKUs, One Shop Floor

It is a fair question, and it comes up in almost every review. Which of our products actually make money?

Everyone in the room can name the big runners and the obvious duds. Those are easy. But somewhere between the star performers and the clear failures sits the vast middle, hundreds or thousands of variants, and for that middle the honest per-product answer is not on the table. People know the averages. They know the totals. Ask which specific products earn their keep and which quietly do not, and the room goes thoughtful in a way that answers the question by not answering it.

You can list all 6,000. You can price all 6,000. Ask which of the 6,000 are worth making, and suddenly it is complicated.

The mix outgrew the measurement

High product-mix manufacturing is genuinely, structurally hard to see. This is not a discipline failure or a lazy team. It is the nature of running thousands of variants across shared lines.

Costs blur, because most SKUs are not made on their own dedicated line. They share machines, people, and time with dozens of others. Every switch from one product to the next costs changeover time and often some yield, but that cost rarely gets attributed back to the specific product that caused it. It disappears into a general overhead and gets smeared evenly across everything, including the simple products that never caused a changeover in their lives.

Quality behaves the same way. A defect pattern repeats across a family of variants, but each instance gets handled as its own small fire, so nobody sees the pattern that connects them. The plant-level averages look perfectly healthy. The per-SKU truth sits underneath, unread. Nobody decided to lose track. The catalogue simply grew faster than the way the plant measures itself.

What you cannot see, you keep making

The cost of this is not dramatic. It is quiet and continuous, which is worse.

You keep producing SKUs that lose money, because nothing in your reporting flags that they do. They hide inside a healthy-looking average, subsidised by the products that genuinely earn. You under-price complexity, because the true cost of a fiddly, low-volume, high-changeover product looks the same on paper as a simple long-run one. Your scheduling optimises for the wrong things, because the changeover tax it is quietly paying is not a number anyone can see. Quality effort gets spread evenly, when it should be aimed at the few variants causing most of the trouble.

Most of all, leadership cannot confidently do the two things a wide catalogue most needs. It cannot rationalise the portfolio, because cutting SKUs on gut is dangerous when you cannot see which ones carry hidden value or hidden cost. And it cannot price for complexity, because the complexity is invisible in the numbers. So the big decisions about the mix get made on instinct and history, which is exactly how the mix got unwieldy in the first place.

Tools built for a simpler catalogue

The familiar tools are not wrong here either. They were built for a world with fewer and more stable products.

The ERP holds a standard cost for each SKU, and standard costs are useful, but they smooth over precisely the things that make high-mix expensive: the real, variable cost of short runs, frequent changeovers, and rework. BI dashboards show aggregate throughput and margin beautifully, which is the catch, because aggregates are where the per-SKU story goes to hide. The cost sheet in Excel does try to get specific, but it goes stale the moment the mix shifts, and it cannot realistically stay current across thousands of live SKUs.

Each of these answers an aggregate question well. How much did we make, how much did we ship, what was the overall margin. None of them was built to answer the per-SKU question leadership actually has, which is this particular product, across all the mess of shared lines and changeovers, is it worth it.

From a big catalogue to a clear one

The shift is not to slash the catalogue. It is to be able to see it.

Capture production reality at the SKU level, the yield, the reject, the changeover, the run length, as it actually happens on the floor. Curate it into one consistent view, so the cost and quality of each product become visible across the whole mix, including the shared costs that standard costing smears away. Then, for the first time, leadership can see which products genuinely earn, which ones carry a hidden complexity cost, and which quality problems repeat across a family rather than appearing at random.

To be clear about what this is: it is manufacturing intelligence, which is decision support. It gives leaders the per-SKU truth so they can price better, aim quality effort where it pays, and decide which products belong in the catalogue. It informs those decisions. It does not make them automatically, and it is not a scheduler that runs the floor. The judgement stays with the people. The blindfold comes off.

Questions for your next leadership meeting

• If I asked which fifty of our SKUs make the most money and which fifty lose money, could we answer, or would we estimate?
• Does the cost of a changeover ever get attributed to the product that caused it, or does it vanish into overhead?
• Are we pricing our complex, low-volume products for the real cost they carry, or the same way as our simple runners?
• When a defect repeats across a family of variants, do we see the pattern, or fight each instance separately?
• The last time we talked about trimming the range, did we decide on data or on gut and history?

A catalogue you cannot see is where margin hides

A big catalogue can be a real strength. Range wins customers, fills channels, and keeps you on the shelf. None of that is the problem.

The problem is a catalogue you cannot see clearly, because that is simply a large, comfortable place for margin to hide. You do not have to cut 6,000 SKUs to fix this. You have to be able to see them. The plants that master their mix are not the ones with the fewest products. They are the ones that can tell you, without guessing, which of their products are actually worth making.

If your catalogue has grown faster than your view of it, it may be worth looking at how manufacturers are getting cost and quality clarity at the SKU level across a wide mix. Explore our executive guide.

FAQ