How to Write a Quality Control Plan

A quality control plan is a written document that maps every control point in making a product or delivering a project to a measurable outcome: what gets checked, how it is measured, what counts as a pass or fail, who is responsible, what record is kept, and what happens when something falls outside limits. It turns "we check quality" into a set of decisions anyone can follow.

Most quality problems on a floor are not mysteries. They are the predictable result of a check that was skipped, a limit nobody wrote down, or a failure with no agreed response. A quality control plan closes those gaps before production starts. This guide covers what a quality control plan is, how it differs from the QMS above it and the control plan inside it, exactly what belongs in one, a seven-step way to write one, and the mistakes that turn a good plan into a binder nobody opens.

What is a quality control plan?

A quality control plan is the operational document that defines the inspections, tests, acceptance criteria, and corrective actions used to keep a product or service meeting its requirements. It sits between two things people often confuse it with. Above it is the quality management system the policies, procedures, and standard that govern quality across the whole organization. Below it, at the characteristic level, is the control plan in the automotive sense: a row-per-characteristic table of specs, methods, and reactions. The quality control plan is the bridge. It scopes a specific product, project, or contract and lays out how quality will be controlled across it end to end.

ISO gives it a formal home. ISO 10005:2018, Guidelines for quality plans defines a quality plan as a specification of the procedures and resources that will be applied to a specific output, a product, service, project, or contract, and by whom and when (ISO 10005:2018). The American Society for Quality frames it the same way: a quality plan translates the customer's requirements, the what into the actions needed to produce the result, the how and names who does it (ASQ, Quality Plans).

The reason to write one is transfer and repeatability. A quality control plan lets the person who understood what mattered on the day the product was developed hand that understanding to every shift, every inspector, and every auditor who comes after, without being in the room. It is how a one-time decision about what to check becomes a durable control.

What goes in a quality control plan?

A quality control plan is built from a fixed set of sections, and the value is that each one forces a decision vague quality talk skips. A complete plan covers scope and objectives, the standards it answers to, the control points, the methods and acceptance criteria, responsibilities, records, the reaction and corrective-action path, and a review schedule. The table lays out each section and the decision it captures.

Two sections carry more weight than the rest. Acceptance criteria have to be measurable limits tied to a drawing, a customer spec, or a standard, a number and a range, not an adjective. And the reaction path is what separates a plan from a checklist: it names exactly what happens when a check fails, so the response is a procedure decided in a quiet room, not an improvisation during a bad shift.

How do you write a quality control plan?

You write a quality control plan by deciding what to control before you decide how to control it, and you choose using risk, not habit. The sequence that produces a plan people actually run:

1. Define the scope and pull the requirements. Name the product, project, or line the plan covers, then gather every source of requirements: customer specifications, engineering drawings, applicable standards, and regulatory rules. These are where your acceptance limits will come from.
2. Map the control points across the flow. Walk the process and mark where quality is verified, incoming material, key in-process steps, and final inspection. A control point belongs anywhere a defect could be introduced or should be caught before the next step adds cost. In batch production that usually means a check at the start of each lot, plus in-process checks tied to the steps that drift.
3. Select the characteristics by risk. At each control point, list the characteristics that carry risk, the outputs your process FMEA flagged as high severity or likely to fail. Do not try to check everything; control what matters.
4. Write measurable acceptance criteria. Give every characteristic a target and a tolerance tied to its source. If you cannot state a limit as a number and a range, you cannot control it, you can only argue about it.
5. Choose the method, sample, and frequency. Pick the gauge or test, confirm it is capable for the tolerance, and set sample size and frequency by risk and stability. Higher-risk, less-stable characteristics get larger samples and more frequent checks; a proven characteristic can be checked less often.
6. Assign responsibilities and records. Name who performs each check, who dispositions a failure, and who signs off. State exactly what gets logged and where, so a check becomes evidence, not a memory.
7. Write the reaction path and set a review trigger. Spell out what happens on a fail, contain suspect product, notify, adjust, handle everything made since the last good check, and link it to your corrective-action process. Then name an owner and the events that trigger a review: any change to the product, process, equipment, or requirement.

Wherever a characteristic is charted rather than pass/fail, statistical process control is the engine and the quality control plan is the instruction sheet, it names which characteristic, which chart, which sample, and which reaction. One tells you the process is drifting; the other tells you what to do about it.

How thin is the average plan, and why does it matter?

The cost of a weak plan shows up as the cost of quality. ASQ reports that for many organizations, the total cost of poor quality, scrap, rework, inspection, warranty, and returns, runs as high as 15% to 20% of sales revenue and that a large share of it traces to defects that a defined check would have caught earlier (ASQ, Cost of Quality). ISO 10005 makes the counterpoint: a quality plan increases confidence that requirements will be met, provides assurance that processes are in control, and gives the people running them a clear reason for how the work is done (ISO 10005:2018). The 1-10-100 rule sits underneath both: a defect caught at its control point costs a fraction of one that escapes to final inspection, which costs a fraction of one that reaches a customer.

How is a quality control plan different from a control plan?

They are the same idea at different altitudes. A control plan in the automotive sense is a row-per-characteristic table, the tightly structured core that lives at the station. A quality control plan is broader: it wraps that table in scope, reference standards, responsibilities, records, and a review schedule, and it can cover a whole project or contract rather than a single part. In many operations the AIAG-style control plan is literally a section inside the larger quality control plan.

Why do quality control plans go stale?

Quality control plans die the same way lean programs stall, the document and the floor drift apart until the plan describes a process that no longer exists. Three failure modes account for most of it.

The plan and the process change independently. Someone moves a check, swaps a gauge, or tweaks a fixture, and the plan is never updated. Months later an auditor finds the floor doing one thing and the plan saying another. Every process change has to trigger a review, that is the whole job of the owner and the review schedule.

Reaction paths that only live on paper. The plan says "hold the lot and notify quality," but at 2 a.m. the operator adjusts the machine and keeps running. A reaction path nobody follows is worse than none, because it creates a false record of control.

Records nobody reviews. If check data lands on a clipboard that gets filed unread, the plan is theater. The value of a control is fast feedback when something drifts, and paper logs totaled at month-end are history, not feedback. When checks are captured digitally at the station and flow into one operational layer, a failed check can trigger the reaction path in the moment, flag it, prompt the operator, hold the lot, instead of surfacing days later. Harmony digitizes the checks operators already run and connects them to the machines and systems around the line in one live layer with no rip-and-replace of existing gauges or software. When CLS moved its production and quality logging off paper the checks stopped being a record filed after the fact and started being a signal the team could act on during the shift.

Start with the handful of control points that carry the most risk, write real criteria and real reactions for them, and make sure the data comes back fast enough to act on. A quality control plan is not a compliance artifact. It is the memory that keeps quality from depending on who happens to be on shift.