Track Spoilage by Root Cause in Manufacturing

When "Waste" Is One Number, You Can't Fix Anything

A converter closes out the quarter. The waste report shows spoilage running at some percentage of total footage across the plant — one line, one number, filed under "spoilage" or lumped in with "makeready." The plant manager asks the obvious question: was it the new die on the eight-color job, the operator still learning a new substrate, or a color match that ate forty extra feet before it hit tolerance? Nobody can say. Every operator logged their scrap into the same bucket, on the same form, with the same word: waste. The number is accurate and completely useless, because by the time it lands in a spreadsheet it can no longer be pulled apart by cause. Multiply that across every job run that quarter and you have a very solid figure attached to a problem you cannot act on.

This is not a data problem. It's a tagging problem, and it gets solved on the floor, in real time, or it never gets solved at all.

By the end of this piece you'll have a lightweight enough cause-tagging habit that operators will actually keep it up under real production pressure, and a simple way to turn that logged data into a ranked list of what to fix first.

How to Track Spoilage by Root Cause Without Slowing Down the Press

The reason most shops don't track spoilage by root cause isn't that nobody's thought of it. It's that every attempt to do it properly turns into a form with fifteen fields, and an operator mid-changeover is not going to fill out fifteen fields. The discipline that survives contact with a real shift is short: a cause code, a footage or count, and a job number. That's it.

A cause code is just a short, fixed vocabulary — three or four letters, printed on a card taped to the press — that maps to the handful of things that actually cause spoilage on that line. Not a free-text field where an operator writes "press issue," which tells you nothing six weeks later. A code. Operators circle or scan it, log the waste count against the job, and move on. The friction has to be close to zero, because the moment logging costs real time, it gets skipped, and skipped logging is worse than no logging — it looks like data but it's actually just gaps with a false sense of completeness.

This is also where a job-costing view earns its keep. If your shop is already comparing estimated to actual costs on a job — waste, time, and material against what was quoted — root-cause tags are what turn "we ran over" into "we ran over because of X," which is the only version of that sentence anyone can act on.

The Root Causes That Actually Recur on a Narrow-Web Floor

Every plant's exact list looks a little different, but most narrow-web spoilage collapses into a small number of recurring categories:

• Makeready and setup — the waste inherent to getting a press to color and register before good product starts, which is a known, budgeted cost, not a defect. (Worth separating cleanly from the categories below — see the deeper breakdown on makeready waste in flexo.)
• Plate mounting and registration — misalignment on mount, plate wear, or a register that drifts mid-run.
• Color match — extra footage burned getting density and hue within tolerance, especially on tight brand-color jobs.
• Die cutting — dull or damaged dies, matrix breaks, kiss-cut depth wandering as blades wear.
• Web handling — splices, breaks, tension issues, especially on thinner or harder-to-handle substrates.
• Substrate or material defect — waste that isn't the press's fault at all, but the material's, and shouldn't be scored against the operator.
• Operator or changeover error — the honest catch-all for the mistakes that happen when people are involved, which is always.

Note that only one of these — makeready — is expected and budgeted for. The rest are the ones a root-cause discipline is actually built to catch, because they're the categories that should be shrinking over time if anything is working.

Turning Cause Codes Into a Pareto You Can Act On

Once spoilage is logged with a cause code instead of a shrug, the analysis writes itself. Sum the waste by code across a month or a quarter, sort descending, and you have a Pareto ranking: the handful of causes responsible for most of the scrap, in order. This is the whole point of tagging by cause in the first place — not to produce a tidier spreadsheet, but to stop guessing about what to fix first.

A spoilage number with no cause attached is a fact about the past. A spoilage number with a cause attached is an instruction for next quarter.

The full mechanics of building that ranking — how to bucket close-call causes, how to weight by footage instead of just count, how to avoid a Pareto that's dominated by one noisy job — are covered in more depth in our walkthrough of printing waste Pareto analysis. The short version: if plate mounting and color match together account for most of your non-makeready waste, that's where a die inspection routine or a proofing step pays for itself. If it's web handling, that's a tension and splice conversation, not a color conversation.

Where Root-Cause Tracking Connects to Makeready Waste

It's worth separating makeready from the rest of the list explicitly, because conflating them is one of the most common ways a spoilage report gets muddied. Makeready waste is structural — it comes from the mechanics of getting a press to color and register, and it scales with colour count and job complexity in a fairly predictable way. It is not a defect to be rooted out; it is a cost to be estimated accurately and then watched for drift. If your makeready waste is creeping up job over job on comparable work, that's a maintenance or plate-quality signal, not a mystery — see how makeready waste is built up on a flexo job for the full mechanics. Everything else on the cause list, by contrast, is waste a shop should be actively working to shrink.

From Logging to Reduction

Cause codes and a Pareto ranking are diagnosis, not treatment. The reduction work itself — die maintenance schedules, plate mounting checks, tension audits, operator training loops tied back to the actual top causes rather than a general "be more careful" — is its own discipline, and it's covered at length in our piece on lean waste reduction in printing and in the broader playbook on how to reduce waste in label printing.

None of this requires new software to start. It requires a short cause list taped to the press, a habit of logging against a job number instead of a shrug, and someone willing to sum it up monthly instead of just at year-end. If you want a starting point rather than building the tagging sheet and the summary view from scratch, we put together a simple version of exactly this — cause codes, a logging sheet, and a ready-made Pareto rollup — as the Waste & Spoilage Tracker by Cause. It's built to be filled out by an operator between jobs, not by an analyst at month-end.