Containment & Response Clocks
What Is Containment in Quality? A Supplier's Guide to the First Hours
Containment is the work that protects the customer while the cause is still unknown. This guide covers where suspect stock sits, how to bound it, and what to record.
Rovaryn Digital · · 10 min read

A rejection notice arrives and the clock is already running
When containment is slow or vague, suspect parts keep moving. Consider a burr found on a bracket at a customer's receiving inspection. Some of those brackets are still in your plant, some are on a truck and some are on the customer's dock, and every hour without a count makes the eventual sort larger. The customer's request asks for a containment response within the window stated in its manual, and that window is already running.
The first hours of a request like this are rarely about root cause. They are about a plainer question: where is the suspect material, and what is being done so that no more of it reaches the customer? The answer has to be specific enough to write down and sound enough to send.
This guide explains what is containment in quality from the supplier's side of the desk. It covers how to bound the suspect population, where to look, how containment differs from corrective action, and what to record so the response still makes sense when someone asks about it weeks later.
What is containment in quality, in practical terms?
Containment is the set of actions that protect the customer while the cause is still unknown. It isolates suspect material, checks what can be checked, and puts a temporary barrier between the problem and the next shipment. It does not explain why the defect happened, and it does not fix the process.
IATF 16949 clause 10.2.3 asks for a documented problem-solving process. That process includes containment and interim actions, root cause analysis with its methodology and results, and systemic corrective actions (IATF 16949:2016 with Sanctioned Interpretations, Nov 2025). The same clause says that where a customer has prescribed problem-solving tools or formats, you use them unless the customer approves otherwise. Aerospace suppliers answering a prime's request work with SAE AS13000A, which describes the process a supplier follows when responding to a customer request for corrective and preventive action (SAE International, 2024).
In an 8D response, containment is the interim containment discipline, usually D3. The D3 containment action is worth reading on its own if you are building the response step by step.
Three questions sit underneath every containment effort:
- What is suspect? Which parts, lots, dates and locations could carry the defect.
- What was done to it? Which checks, sorts or holds were applied, by whom, and with what result.
- What stops more from escaping? What temporary barrier now sits between the process and the customer.
If you can answer all three in writing, you have containment. If you can answer only the second, you have a sort, which is useful but incomplete.
Where suspect stock sits: plant, transit and customer
Suspect stock rarely sits in one place. A practical habit is to walk the part's path from the last point you trust to the customer's line, and ask at each stop where a unit could be. Think of three rings: your plant, the pipeline, and the customer.
Inside your plant
Start with the obvious areas and then check the ones people forget:
- Finished goods in the warehouse and at the shipping dock
- Work in process at downstream operations
- Parts at outside processors, such as plating, heat treat or coating
- Quarantine, rework and returned-goods areas
- Line-side containers, kits and packaging stations
- Upstream material or purchased components, if they are suspected
Outside processors are the most common blind spot. A lot can be split across several locations in a single week, so confirm quantities by phone or email rather than relying on a last known count.
In transit and at warehouses
Shipments already on the road count as suspect. So does consignment stock, stock at a third-party warehouse, and stock at any cross-dock or sequencing site the customer designates. Your logistics or shipping contact can match ship records to lot or date codes. Record the carrier, the ship date and the expected delivery, because the customer will need to know which loads to intercept.
At the customer
The customer may hold suspect parts in receiving, in stores, at the line, or in assemblies already built. Your supplier quality engineer (SQE) contact at the customer can tell you what has been quarantined and what the customer expects from you. Sorting at the customer's site, by your people or by a third-party sort service, is something the customer typically directs, so ask before you send anyone. Agree who counts, who records and how results are shared.
How to bound suspect material quantity containment
Bounding is the step that turns a vague worry into a number you can defend. Suspect material quantity containment works best when you fix the edges of the population first and count second.
- Find the last known good point. This is the most recent production event you can show was conforming. It might be a lot, a shift, a tool change, a material heat or a verified inspection record.
- Find the first known bad point. Use the traceability on the rejected parts: date code, lot, machine, cavity, die or fixture.
- Treat everything between as suspect. Stay with that assumption until evidence narrows it. Narrow the boundary only with data, such as gauge results or a documented change event.
- Count by location. Record each location's quantity with the date and time of the count. Quantities move, so a number without a time stamp invites questions.
- Choose a check method that can see the defect. A gauge, fixture or visual standard must be able to detect the actual condition. Run it against a known bad sample first, and keep the sample.
- Sort and tag. Segregate the parts, then label them conforming, rework or scrap, with the lot and the person who checked.
- Reconcile. Produced quantity should equal shipped plus on hand plus scrapped. If it does not, you have a gap, and the gap is suspect until it is explained.
- Widen if the defect keeps appearing. If checks keep finding bad parts near the edge of your boundary, move the boundary outward and say so in the record.
Here is a small worked example of step 7, using round numbers purely to show the method. Suppose a lot of 1,000 parts was produced. Of those, 600 were shipped, 250 are in the warehouse and 100 were scrapped at an earlier operation. That accounts for 950, so 50 parts are unaccounted for. Those 50 might be at an outside processor, in a rework bin or miscounted, but until someone finds them they stay inside the suspect population.
Interim containment action vs permanent corrective action
Readers often search for interim containment action vs permanent corrective action because the two get blended in a response. They answer different questions and have different evidence behind them.
| Question | Interim containment | Permanent corrective action |
|---|---|---|
| Purpose | Protect the customer now | Remove the cause so it does not recur |
| Depends on root cause? | No, it starts before the cause is known | Yes, it follows from the root cause |
| Targets | Suspect parts and the path to the customer | The process, system or control that allowed the defect |
| Typical evidence | Sort results, hold tags, inspection records, quantities by location | Changed procedure, tooling, control plan or training, with verification data |
| Duration | Temporary, kept in place until corrective action is verified | Intended to stay in place |
| Ends when | Verified corrective action is in place and the customer agrees | Effectiveness is verified over time |
One distinction matters here: containment deals with parts that have already escaped or might still escape, while corrective action deals with why the defect occurred and why it was not detected. A strong response usually addresses both the occurrence cause and the escape cause. For a fuller treatment, see containment versus corrective action.
A common mistake is treating the sort as the solution. If 100% inspection is catching the defect, the sort is doing its job, but it is also a cost and a dependency on the person doing it. The customer will want to know what replaces it, and when.
A containment plan template you can build in a spreadsheet
You do not need software to start. A containment plan template needs two parts: a header that identifies the request, and a log that records what was done at each location.
Header fields to capture when the request arrives:
- Customer, request reference and the date and time received
- Part number, description and revision
- Defect description in plain words, with a photo if the customer sent one
- Last known good point and first known bad point
- Containment owner and backup
- Customer contact notified, and when
- Where the evidence is stored
Containment log layout:
| Location | Suspect qty | Checked | Conforming | Rework | Scrap | Method | Checked by | Date/time | Tag or lot ID |
|---|---|---|---|---|---|---|---|---|---|
| Finished goods | |||||||||
| WIP, downstream op | |||||||||
| Outside processor | |||||||||
| In transit | |||||||||
| Customer site |
Add a row for each location you identified in the three rings. Keep the log in one file so that nobody has to reconcile five versions on a deadline.
A containment response is only as strong as the boundary behind it. If you cannot say where the suspect stock starts and stops, you cannot say the customer is protected.
Two more habits help. Photograph the tagged, segregated stock, because a picture of the hold area is often the quickest evidence to send. And log each decision when it is made, not afterward. For worked scenarios across different process types, see containment action examples in manufacturing.
Reading your customer's manual for the containment window
Containment windows, formats and expectations are set by each customer. They are written in that customer's supplier manual or customer-specific requirements (CSRs), and they change between revisions. One customer may expect a response in hours, another in days, and a third may tie the window to the severity of the issue. This guide does not state any customer's rule because the current revision of your customer's manual is the only reliable source.
When you open a manual, read the response section with these questions:
- What starts the clock? The notification, the email, a portal posting or the receipt date?
- Calendar or working days? And in which time zone, with what holiday treatment?
- What counts as a containment response? A call, a form, an initial 8D or a portal entry?
- Is extra control expected? Some customers expect a stricter shipping control while containment is open. Controlled shipping is explained separately.
- Who is the contact, and what is the after-hours route?
- What do the manual's administrative or late-response terms say? Read them as written, and note the revision.
- Which revision are you reading? Record its date, because next year's version may differ.
Windows and fee terms vary by customer and by revision, so a table of your own customers' terms is worth maintaining. Note the manual's revision beside each entry and recheck it when the customer issues an update. For more on how these expectations differ, see 8D response time and customer requirements. If a window in a manual is unclear, ask your customer's SQE or buyer to confirm it in writing.
Keeping the containment record in one place
Many suppliers keep containment records in a spreadsheet, a Word file and an email thread, then re-enter the content into each customer's format. That works, but it puts the burden on whoever is holding the file when the next request arrives.
A supplier-side record can hold this in one place. Supplier CAR Response Desk is designed to log each request as its own record, with editable response rules per customer, action items with an owner and due date, and evidence uploads attached to the relevant discipline. The starting values for those rules are editable defaults, not a statement of any customer's requirements. The Response Desk records and calculates; it is not advice or certification, and it is still pre-launch.
Whatever tool you use, the habits above carry the weight: bound the population, count by location with a time stamp, tie each check to a method, and keep the evidence next to the record. If you would like to hear when the Response Desk opens, you can join the waitlist.
Put it to work
Compare plans on the pricing page.
Get the next guide
Practical guides for supplier quality teams, delivered straight to your inbox.
Related guides

8D Response Time: Reading Your Customer's Requirements and Running the Clocks
There is no single 8D deadline. Each customer sets its own windows, and this guide shows how to find them, record them once and keep every clock running.
Rovaryn Digital · · 13 min read

8D D3 Containment Action: What to Do in the First Hours
D3 is the first thing a customer sees from you. This guide covers locating suspect stock, sorting and marking it, and drafting an interim response your team can sign off.
Rovaryn Digital · · 8 min read

The 24-Hour Containment Window: Running the Clock on a Customer Complaint
Many teams work to a 24-hour containment window, but the real number is whatever your customer's manual says. This guide covers confirming it, starting the clock and recording the interim response.
Rovaryn Digital · · 6 min read