cmyk, colour, to dye, color fan, color picker, printing inks, printing house, print, print production, paper, magazine, flyer, folder, poster, reproduction, repro, cmyk, cmyk, cmyk, cmyk, cmyk, color fan, color picker, color picker, color picker, printing house, print, print, print, print, print production, print production, print production, magazine, magazine, magazine, flyer, flyer, folder, folder, reproduction, repro. Print shop quality control: four gates and a reprint log
Photo by markusspiske on Pixabay

Guides

Print shop quality control: four gates and a reprint log

Print shop quality control starts at file arrival and continues through finishing, with four checkpoints, a reprint log, and measures owners can act on.

What to take away

  • Print shop quality control starts at four gates, and three of them happen before a customer sees anything. A check at delivery is not quality control; it is sorting.
  • The reprint log is the most useful document in the building, and only if it records cause rather than customer.
  • Measure what the shop can act onfirst-pass yield, reprint rate by cause, and where jobs wait. Satisfaction scores tell you how a customer felt, not what to change.
  • Most dissatisfied print customers do not complain. They order less. Watch order frequency, not just the inbox.
  • Color disputes are settled by the approved proof and by a stated tolerance, so both have to exist before the argument.

Four gates, in the order they save money

Each gate catches faults that are far cheaper here than at the next one.

Four gates, in order

  1. File arrival: preflight before scheduling
  2. Proof: content, position, folding dummy
  3. Press: first good sheet signed
  4. Finishing: count, trim, fold, label

Gate one, at file arrival. Preflight before the job enters the schedule. Color space, bleed, fonts, resolution, trim and fold marks, and whether the finished size matches what was ordered. A fault caught here costs an email. The same fault caught after plates costs a makeready.

Gate two, at proof. Content, position, folding dummy where the piece folds, and an explicit statement of what the proof does and does not cover. A folding dummy takes minutes and prevents the most expensive category of reprint: a job that is correct on every page and wrong when it is folded.

Gate three, at the press. Check and sign the first good sheet against the approved proof. Before the run, set a sheet-count interval in the job ticket based on run length, stock, and press stability; pull and check sheets at that interval and after any adjustment, stoppage, or stock change. Record the count, result, and any correction so a fault can be traced to a range rather than to the whole job.

Gate four, at finishing and packing. Count, trim size, fold accuracy, that the right job is in the right box, and that the label matches the address. This is where correct printing becomes an incorrect delivery.

The reprint log

One document does more for quality than any inspection regime, and most shops either do not keep it or keep it wrong.

Log every reprint, credit, and rework with these fields:

  • Job number, date found, and customer or account.
  • Stage where the fault was found and its fixed cause code.
  • Quantity affected and whether it was reprinted, credited, or reworked.
  • Direct cost to the shop.
  • Corrective action, resolution, and the person who approved it.

What the reprint log records

  • Job and job type
  • Stage where fault originated
  • Cause from fixed list
  • Cost in materials and minutes
  • Who paid, under which term

A short fixed cause list is what makes the log usable. Free text cannot be counted, and a log that cannot be counted is a diary. Six or eight causes covering supplied file, prepress, plate or setup, press, finishing, packing, stock, and specification error will cover almost everything.

Read it monthly, sorted by cost rather than by count. The pattern is rarely spread evenly, and the top two causes usually account for most of the money.

What to measure, and what not to bother with

What to measure

MeasureWhat it tells youWhy it is worth the effort
First-pass yield by job typeShare of jobs delivered without reworkThe single clearest signal, and it points at a job type rather than a person
Reprint cost by causeWhere the money actually goesDirects the fix instead of spreading a general warning
Promised against actual deliveryWhether the schedule is honestLate delivery is the complaint customers act on silently
Time a job waited, by stageWhere the constraint really isTurns a quality complaint into a capacity answer
Preflight failure rate by customerWhich accounts need a file conversationPrevents the same argument every month

Two things not worth much: a general satisfaction score with no attached action, and inspection headcount. Adding an inspector to a process that produces faults produces the same faults, found later.

Worth measuring vs not

Worth the effort

First-pass yield
Clearest signal
Reprint cost by cause
Directs the fix
Promised vs actual
Honest schedule
Satisfaction score
—
Inspection headcount
—

Not worth much

First-pass yield
—
Reprint cost by cause
—
Promised vs actual
—
Satisfaction score
No attached action
Inspection headcount
Finds faults later

Use the stage-by-stage picture of how a job moves through the shop to identify where jobs wait; put an owner, a pass/fail result, and a time on each gate in the job ticket. That makes the measures actionable without relying on unnamed procedures.

Color, proofs and the argument you can win

The hardest quality dispute in print is not a defect. It is a job that matched its proof and did not match the customer's expectation.

Three things that settle color disputes

  • Approved proofwho, what artifact, when
  • Stated tolerance, measured as you measure
  • Written term for out-of-tolerance runs

The approved proof, recorded with who approved it, what artifact they saw, and when. A screen view on an uncalibrated monitor checks content and position; it does not check color, and that has to be said before the run rather than after.

Name the color reference and the tolerance the shop and customer agreed to, and state how the shop measures it. For example, if your process follows ISO 12647, identify the applicable part rather than citing the series generally; check the current text from ISO before specifying it. A stated tolerance only holds up if the shop's own written standard is checked against the current compliance checklist.

And a written term covering what happens when a run falls outside it. That is a contract question rather than a legal fact, so have a qualified adviser look at your wording, particularly on reprints, overruns and rights in customer-supplied artwork.

Stock lots are the usual physical cause. Two runs months apart come from different lots, and shade varies between them. Telling a customer that in advance is accepted; telling them afterward sounds like an excuse.

The complaints that never arrive

Most unhappy print customers do not call. They place the next job somewhere else, and the shop records a good month.

Signals of silent complaints

  • Order frequency by account
  • Broken ordering rhythm
  • Ask about a specific job
  • Ask operators about difficult job types

So watch order frequency by account, alongside the complaint log. An account whose ordering rhythm has broken is telling you something the inbox is not. Ask them directly, and ask about a specific job rather than in general, because general questions get polite answers.

Ask the floor as well. Operators know which job types are difficult and which customers send unusable files long before the numbers show it, and they will say so if asked about a job rather than about performance.

Where quality and safety overlap

A rushed setup and an unsafe setup are frequently the same setup.

Machine guarding and energy isolation before maintenance are governed requirements, not negotiable, and they erode first under schedule pressure. For applicable machines and tasks, consult OSHA's machine-guarding rule (29 CFR 1910.212) and its control-of-hazardous-energy, or lockout/tagout, rule (29 CFR 1910.147), alongside OSHA's printing industry pages.

Chemical information reaching the people who handle it is covered by OSHA's Hazard Communication Standard (29 CFR 1910.1200). Solvent handling and waste routes are questions for your state environmental agency about your own site.

Where schedule pressure is the problem, plan the week against due dates and the documented capacity of each press and finishing line; scheduling that does not slip can guide that planning. Do not recover time by skipping a gate; record each line's limits in the shop's capacity notes so the schedule reflects what the equipment can actually do.

The records behind all of this, job tickets, reprint logs and approvals, are ordinary business records, kept in the form set out in the IRS guidance on record keeping.

Common questions

How often should press sheets be pulled during a run?

There is no universal interval. Before the run, set a sheet-count interval in the job ticket based on run length, stock, and press stability; shorten it when the press or stock is less stable. Pull at each planned count and after any adjustment, stoppage, or stock change. Record the count, check result, and correction on the job ticket; the last passing pull and first failing pull bound the affected range.

Should I show customers the reprint log?

No, but you should show them the change it produced. A customer who hears that a recurring fault has been traced and fixed is being given something more useful than an apology.

What is a reasonable reprint rate?

Nobody can give you a number that means anything about your shop. Track your own, by cause, and judge it by whether it is falling.

How do I handle a reprint that is genuinely nobody's fault?

Decide in advance, in your terms, and apply it consistently. What destroys goodwill is not the answer; it is discovering that the answer depends on who is asking.

More in Guides

Latest from Guides Desk