Nov 11, 2026

Quality Control Plan: Example, Template and 6 Steps

A quality control plan is a document that states what gets checked during a process, how it gets checked, how often, and what happens when a check fails.

It turns "we inspect our products" into something specific enough that a new operator could follow it on their first shift.

This guide shows a filled example, explains every column a control plan needs, and covers the three different documents that go by this name.

What Is a Quality Control Plan?

At its simplest, it answers four questions for each step in a process.

What are we measuring? How are we measuring it? How many and how often? What do we do if it is out of spec?

Everything else on the form supports those four answers.

A plan that says "inspect dimensions regularly" is not a control plan. One that says "measure bore diameter with a bore gauge, 5 pieces per hour, quarantine the batch and notify the shift lead if outside 25.00 to 25.05 mm" is.

How It Differs From a Quality Assurance Plan

Quality control is about detecting problems in the output. Quality assurance is about designing a process that does not produce them.

A control plan sits on the QC side. It specifies the checks.

The full distinction has its own guide covering quality assurance and quality control, which is worth reading if the terms blur together for your team. They do for most people.

Three Different Documents Share This Name

This causes genuine confusion, and search results make it worse by mixing all three together.

Three Different Documents Share This Name

The Control Plan (Automotive and IATF 16949)

This is the most formally defined version.

It comes out of APQP, the advanced product quality planning process used across automotive supply chains. IATF 16949 requires it, and it is one of the elements submitted in a PPAP package.

Characteristics on the control plan are supposed to flow from the PFMEA. If a failure mode was rated as high risk, the control plan is where you show what detects it.

Structured format, defined columns, customer approval required. If someone in automotive or aerospace asks for a control plan, this is what they mean.

The Contractor Quality Control Plan

Common in construction and government contracting, often abbreviated CQCP.

It covers who does the inspections on a project, at which stages, what records get produced, and how non-conforming work gets handled. Federal and state highway contracts routinely require one before work starts.

Much broader than the automotive version. It covers organisation, personnel qualifications, and submittals as well as inspection points.

Teams in construction are almost always talking about this one.

The Project Quality Plan (ISO 10005)

ISO 10005 gives guidelines for quality plans - documents that set out how a quality management system applies to a specific project, product, or contract.

Used when a one-off project needs quality arrangements that differ from your standard system. A new customer with unusual requirements, or a contract with its own acceptance criteria.

Broader than a control plan, narrower than a quality manual.

Which do you need? Automotive or a customer asking for PPAP, use the control plan format. Construction or government contract, use the contractor version. A project needing its own quality arrangements, write a quality plan to ISO 10005.

The rest of this guide focuses on the control plan format, because it is the most transferable. The structure works in manufacturing regardless of sector.

What Goes Into a Control Plan

The standard automotive format has thirteen columns in the body. They group into four questions.

What Goes Into a Control Plan

Column	What it holds
Part or process number	Links the row to the process flow
Process name	The operation being controlled
Machine, device, jig, tools	What performs the operation
Characteristic number	A reference for tracing back to the PFMEA
Product characteristic	A feature of the part. Bore diameter, surface finish
Process characteristic	A setting of the process. Temperature, pressure, speed
Special characteristic class	Flags safety or regulatory significance
Specification and tolerance	The acceptable range
Evaluation or measurement technique	The instrument or method used
Sample size	How many units are checked
Sample frequency	How often the check happens
Control method	How the process is controlled. SPC chart, poka-yoke, visual check
Reaction plan	What happens when the result is out of spec

Two columns carry more weight than the rest.

Product versus process characteristic. A product characteristic is on the part. A process characteristic is on the machine. Confusing the two is the most common mistake on a first control plan.

The reaction plan. Most incomplete control plans have every other column filled and this one blank or vague. It is the column an auditor reads first.

A Filled Quality Control Plan Example

A machined aluminium bracket, moving through five control points.

Each one sets out the characteristic, the specification, how it is measured, how much and how often, and what happens when it fails.

Operation 1. Incoming Material

Characteristic: alloy grade. A product characteristic, verified before anything is cut.

Specification: 6061-T6, confirmed against the supplier certificate.

Method: certificate review against the purchase order.

Sample and frequency: every lot, on each delivery.

Reaction plan: quarantine the lot. Notify purchasing. Request a replacement certificate before releasing the material to production.

Operation 2. CNC Milling - Bore Diameter

Characteristic: bore diameter. This is a product characteristic - a feature of the finished part.

Specification: 25.00 to 25.05 mm.

Method: bore gauge, calibration ID BG-04.

Sample and frequency: 5 pieces, hourly.

Control method: X-bar and R chart. The process is stable enough for statistical control rather than inspection.

Reaction plan: stop the machine. Quarantine everything produced since the last good check. Adjust the tool offset, then verify 5 pieces before restarting.

Operation 3. CNC Milling - Spindle Speed

Characteristic: spindle speed. Same operation as above, but this is a process characteristic - a setting on the machine rather than a feature of the part.

Specification: 8,000 rpm, plus or minus 200.

Method: machine readout, checked against the setup sheet.

Sample and frequency: once, at each setup.

Reaction plan: halt the setup. Recheck the program. First-off approval required before the run begins.

Operation 4. Deburring

Characteristic: edge condition.

Specification: no burr exceeding 0.1 mm.

Method: visual comparison against a boundary sample held at the station.

Sample and frequency: 100 percent, every piece. Checking is quick and cheap here, so there is no reason to sample.

Reaction plan: rework at the station. Log the occurrence. If three occur in a single shift, escalate for review.

Operation 5. Final Inspection

Characteristic: overall length.

Specification: 120.0 mm, plus or minus 0.2.

Method: vernier caliper.

Sample and frequency: 8 pieces per batch.

Reaction plan: hold the batch. Sort 100 percent. Raise a non-conformance.

Reading the Example

Three things in there are worth pulling out.

Operations 2 and 3 are the same machining step. One controls a feature of the part, the other controls a setting on the machine. Both belong on the plan, and they get controlled in completely different ways.

That product-versus-process distinction is the one most first control plans get wrong.

Sample sizes track risk and cost, not habit. Deburring is checked on every piece because a visual check takes two seconds. Bore diameter is 5 per hour because measuring takes longer and the process is already under statistical control.

Nothing here is sampled at a frequency somebody picked because it sounded thorough.

Every reaction plan names an action. Not one says "investigate" or "inform supervisor".

Each specifies what happens to the product already made, what happens to the process, and who gets involved. That is the test of a reaction plan worth writing down.

How to Write a Quality Control Plan in Six Steps

Step 1. Define the Scope

Name the part, the process, or the project the plan covers. Give it a number and a revision.

Scope creep on control plans is real. A plan trying to cover a whole product family usually controls none of them properly.

Step 2. Map the Process

List the operations in sequence, from incoming material to despatch.

Walk the line rather than working from the drawing. The documented process and the actual process diverge more often than anyone expects, and the control plan has to match what people actually do.

Step 3. Identify the Characteristics That Matter

For each operation, ask what could go wrong and what would detect it.

If you have a PFMEA, pull the characteristics from there. The high-RPN failure modes are the ones needing controls, and the traceability between the two documents is something auditors check.

No PFMEA? A risk assessment matrix scoring likelihood against impact gets you most of the way for a first version.

Resist listing every dimension on the drawing. A control plan with ninety characteristics gets followed for a week.

Step 4. Set Methods, Sample Size and Frequency

Name the actual instrument. "Measure" is not a method. "Bore gauge, calibration ID BG-04" is.

Sample size and frequency should scale with risk and with how stable the process is. A capable process under SPC needs less checking than a new one.

Be honest about what is achievable. A frequency nobody can hit gets quietly ignored, and then the plan is fiction.

Step 5. Write the Reaction Plan

For every characteristic, state what happens when it fails.

Three things minimum. What happens to the product already made. What happens to the process. Who gets told.

This is where the control plan connects to the rest of your system. A failed check should trigger a non-conformance, and a repeating failure should trigger root cause analysis and corrective action.

Step 6. Approve, Issue and Review

Get it approved by the people who own the process, not just by quality.

Issue it as controlled documented information with a version number. An uncontrolled control plan is a contradiction that still turns up regularly.

Then set a review trigger. Process change, customer complaint, repeated non-conformance, or a defined interval.

The Three Control Plan Phases

The automotive format recognises that a plan changes as a product matures.

Prototype. Heavy measurement, often 100% inspection of key characteristics. You are still learning what the process does.

Pre-launch. Controls tightened and sampling increased compared to production. The process is running but not yet proven.

Production. Sampling reduced to what the demonstrated capability justifies. SPC replaces inspection where the data supports it.

The useful idea here applies outside automotive. Inspection frequency should fall as a process proves itself, and rise again when something changes.

Plenty of operations are still inspecting at prototype intensity on a process that has run clean for four years. That is cost without information.

Where Control Plans Go Wrong

Blank or vague reaction plans. Everything else filled in, and "inform supervisor" in the last column. It tells nobody what happens to the parts already made.

Written by quality, in isolation. If the operators who run the process were not involved, the frequencies are usually unrealistic and the plan gets worked around rather than followed.

Never revised. Control plans go stale faster than most quality documents because processes change constantly. A plan referencing a machine that was replaced two years ago is a finding.

Too many characteristics. Trying to control everything controls nothing. Twenty well-chosen characteristics beat ninety that nobody checks.

Disconnected from the PFMEA. The characteristics on the plan should trace back to identified failure modes. When they do not, it usually means one document or the other was written to satisfy a requirement rather than to control a process.

Quality Control Plans in Effivity

A control plan in a spreadsheet works until the process changes and four versions are in circulation.

Effivity's quality module holds inspection and control criteria as live records. Characteristics, specifications, methods, sample sizes and frequencies sit against the process step rather than in a separate file.

When a check fails, the reaction is not a line of text. It raises a non-conformance with an owner and a due date, and repeated failures surface as a pattern rather than as separate incidents.

Control plans are issued as controlled documents with version history, approval records and e-signatures. Internal audits link to the plans they test, so an auditor asking which version was live last March gets an answer rather than a search.

Because it sits inside the wider quality management system software, the plan connects to operations and to the rest of the quality management system rather than running alongside it.

Try Effivity for Free and build a control plan against your own process.

Frequently Asked Questions

Can you provide an example of a quality control plan?

Yes. The example in this guide covers a machined aluminium bracket across five control points - incoming material, two checks on the milling operation, deburring and final inspection. Each gives the characteristic, specification, measurement method, sample size, frequency and reaction plan. It also shows the difference between a product characteristic such as bore diameter and a process characteristic such as spindle speed.

What is a quality plan example?

A quality plan sets out how quality will be managed for a specific project, product or contract. A typical example names the quality objectives, the responsible people, the processes and standards that apply, the inspection and test points, the records to be produced, and how non-conforming work will be handled. ISO 10005 gives guidelines for writing one.

What are the 7 steps of quality control?

There is no single formal list, and different sources phrase it differently. A common sequence is: set the standard, measure performance against it, compare the result to the standard, identify deviations, find the cause, take corrective action, and review whether the action worked.

What are the four types of quality control?

Most commonly: process control, which monitors the process as it runs. Acceptance sampling, which inspects a sample to judge a batch. Control charts, which track variation statistically over time. And product inspection, which checks finished items against specification.

What is the difference between a control plan and a quality control plan?

In automotive and IATF 16949, a control plan is a specific structured document with defined columns, produced through APQP and submitted as part of PPAP. "Quality control plan" is used more loosely, often meaning a contractor quality control plan in construction or a general inspection plan. The automotive control plan is the most formally defined of the three.

Is a control plan required by ISO 9001?

Not by name. ISO 9001 requires control of production and service provision, including monitoring and measurement activities at appropriate stages, which a control plan satisfies. IATF 16949 does require control plans explicitly.

How often should a control plan be reviewed?

Whenever the process changes, when a customer complaint or repeated non-conformance points at it, when measurement equipment changes, and on a defined interval. Many organisations review annually as a minimum, though a process under active change needs more.

What is a reaction plan in a control plan?

The reaction plan states what happens when a characteristic falls outside specification. A complete one covers three things: what happens to product already made, what happens to the process, and who is notified. It is usually the weakest column on an incomplete control plan.

Kaushal Sutaria

Managing Director at Effivity Technologies

Kaushal Sutaria is an expert in strategic business management and an entrepreneur behind three global companies. His latest venture, Effivity Technologies, simplifies ISO standard compliance with innovative automation. Kaushal's dedication to best practices and mentorship has earned him clients in over 50 countries.

View All Posts

Improve Workflow & Profitability with Superior Integrated Management System Software free-demo

Schedule a Free Demo