This page is for a brand owner, merchandiser, or quality controller who must decide what to do once a test result is a genuine fail and the delivery date is close. Lessons 3.1 to 3.3 taught how tests are named, how results are rated, and how a report is read against a buyer's standard. This lesson teaches what happens after all of that, once a real fail is confirmed and there is little time left before the goods are due. This is chiefly a decision skill for the money and time flow of an order: a late failed test does not remove the need to decide; it narrows the options and raises the cost of each one. Without a clear view of the real options, a business under time pressure can default to shipping a failed result anyway, simply because stopping to fix it looks slower in the moment.
You need: Lesson 3.2's pass/fail comparison and Lesson 3.3's report-reading method. If you have not read those lessons, the key terms below give you enough to continue.
By the end of this lesson you can:
Not covered here: the main test categories, colour fastness and shrinkage results, and reading a test lab report (Lessons 3.1 to 3.3); and building a complete testing tracker for an order (Lesson 3.5, next).
| Term | Meaning |
|---|---|
| Shrinkage test | A test measuring how much a fabric shrinks after washing or wet processing. |
| Retest | Running a test again, usually after a change to the fabric, process, or sample, to check whether the result now meets the required result. |
| Rework | Reprocessing already-made fabric or garments, such as an extra finishing step, to bring a failed property back within the required result. |
| Sort and cull | Separating a batch into pieces that pass a required result and pieces that do not, rather than treating the whole batch the same way. |
| Required result | The specific rating or percentage a standard states a test must meet or better. |
| Concession | An agreement, recorded in writing, where a buyer or brand knowingly accepts goods that failed a stated required result, usually at a reduced price. |
A genuine test fail is a fact about the fabric or garment, not something a closer delivery date changes. Once a result is a confirmed fail, a business has four real options. Retest: fix the cause, then run the test again to confirm the required result is now met. Rework: reprocess the already-made fabric or garments, such as an extra finishing step, to bring the failed property back within the required result, then retest to confirm. Sort and cull: where only part of a batch failed, separate the batch so passing pieces move forward and failing pieces are held or reworked separately. Documented concession: ask the buyer or brand to knowingly accept the goods as they are, in writing, usually at a reduced price, when the fail is minor and the buyer agrees. Shipping or delivering a failed result without doing one of these four things, hoping it goes unnoticed, is not a fifth option; it is the same fail, undisclosed.
| What can differ | The shapes you meet | Why it matters |
|---|---|---|
| How much time is left before the delivery date | An export order with weeks left before its port cut-off, such as Order A, can usually absorb a retest after a quick fix; a domestic Make to Stock brand replenishing shops on a tight weekly cycle may have far less slack. | Less time left narrows which of the four options is realistic; a retest that needs two weeks does not fit a delivery due in three days. |
| How much of the batch actually failed | A whole fabric lot failing shrinkage usually calls for rework or retest on the whole lot; only some pieces failing, for example from an uneven finishing step, can call for sort and cull instead. | Sorting and culling only makes sense where the fail is uneven; applying it to a lot that failed uniformly wastes time sorting pieces that are all, in fact, the same result. |
| Export order vs. domestic Make to Stock order | An export buyer such as Northline Apparel usually has a written concession process in its own contract; a domestic Make to Stock brand such as Cedar & Co., selling to its own shops, sets its own internal rule for what it will or will not accept from its own supplier. | A domestic brand without a written buyer contract still needs its own clear internal rule; "we usually let it through" is not a documented concession. |
| Physical fail vs. chemical fail | A physical fail, such as excess shrinkage, is often reworkable through reprocessing; a chemical fail, such as a restricted substances result, usually cannot be reworked out of already-made fabric and typically forces retest on a new fabric lot instead. | Treating a chemical fail as reworkable, when the substance is already in the fibre, wastes time on an option that cannot actually fix the problem. |
| Socks, terry, and home-textile product families | A hosiery, terry, or bedding order facing a late test fail meets the same four options; which one fits still depends on time left, batch size, and whether the fail is physical or chemical, exactly as for apparel. | The four-option method taught here is not apparel-specific; only the sample facts change by product family. |
Cedar & Co., the domestic brand with 12 shops, buys a standing-stock T-shirt style from Delta Garments on a 60-day open account, replenishing its own shops from forecast rather than one buyer's confirmed order. A routine shrinkage test on the latest production batch of this style comes back at 5% shrinkage against Cedar & Co.'s own required maximum of 4%, a genuine fail, with the next scheduled shop delivery only five days away. Omar, the owner of Cedar & Co., must choose among retest, rework, sort and cull, or a documented concession, rather than letting the batch ship to shops on the schedule as if the test had passed.
The worked numberOnly options 2, 3, and 4 fit within the five days actually available; option 1 does not, given this mill's typical retest lead time. Cedar & Co. must choose among the options that fit the real time available, not the option that would be preferred with more time.
The methodA common mistake is treating time pressure itself as a decision, shipping or delivering a genuinely failed result quietly because none of the four real options fit the preferred timeline. Running out of time to retest, rework, sort and cull, or arrange a concession does not make the fail disappear; it only means the remaining option, most often a documented concession or a delayed delivery, is the one actually left. A business that ships a known fail without choosing and recording one of the four options has not solved the problem; it has hidden it from whoever receives the goods next, who then discovers it later at a worse point in the chain.
Sometimes a test comes back as a real fail, and the delivery date is close. This does not remove the decision; it just narrows the choices. There are four real things you can do. You can fix the cause and test again. You can reprocess the goods you already have to bring the failed property back within range, then test again. You can sort the batch into the parts that actually pass and the parts that do not, if only some of it failed. Or you can ask the buyer or brand to knowingly accept the goods as they are, in writing, usually at a lower price. What you cannot do is ship the goods as if the test had passed when it has not. Running out of time does not remove the fail; it only limits which of the four real choices still fits.
Cedar & Co.'s latest production batch of its standing-stock T-shirt style has failed its shrinkage test at 5% against Cedar & Co.'s own required maximum of 4%, with five days left before the batch is due at its own shops.
The testing tracker's own note on this batch, exactly as recorded:
Should Cedar & Co. simply ship this batch to its shops on the existing five-day schedule? Decide before you open the decision.
No. Shipping the batch unchanged would mean delivering a known failed result without choosing any of the four real options, which is not a decision but a hidden fail.
The two facts that decide this are the tracker's own recorded fail, actual 5% against a required maximum of 4%, and the fact that two of the four options, rework then retest, or sort and cull, both fit within the five days available. Omar should choose rework then retest if the whole batch shrank uniformly, or sort and cull if only part of the batch is affected, and record which option was chosen and why before any stock reaches Cedar & Co.'s own shops.
This lesson's cost is framed by what an undisclosed fail risks rather than one fixed figure, since Cedar & Co. sets its own internal standard rather than a buyer's contract. If Cedar & Co. ships the failed batch to its own 12 shops unchanged, garments that shrink out of size after a customer's first wash come back as returns, each one costing Cedar & Co. the goods' full retail value plus the shop staff time to process the return, and repeating across every shop the batch reached. Reworking the batch, or sorting and culling it, within the five days available costs only the rework step's own processing time or the labour to measure each piece, a cost concentrated once at Cedar & Co.'s own warehouse rather than spread across returns from 12 separate shops later.
This course's testing tracker records an "action if failed" column for every failed test line, naming the chosen option, retest, rework, sort and cull, or documented concession, so the decision and its reason are visible alongside the failed result itself. [Download the tracker](templates/T09-C03-testing-tracker.xlsx).
| Test name | Actual result | Required result | Result | Action if failed |
|---|---|---|---|---|
| Shrinkage test | 5% | 4% maximum | Fail | Rework (extra finishing step) then retest |
1. What are the four real options once a test has genuinely failed?
2. A batch fails a restricted substances test. Is rework a realistic option for this batch?
3. A different order has 20 days left before delivery, and a retest after a fix typically takes 10 days. Does the same time pressure that ruled out retest for Cedar & Co. also rule it out here?
| EN | AR | FR | TR | Meaning |
|---|---|---|---|---|
| Shrinkage test | اختبار الانكماش | Test de rétrécissement | Çekme testi | A test measuring how much a fabric shrinks after washing or wet processing. |
| Retest | إعادة الاختبار | Nouvel essai | Yeniden test | Running a test again, usually after a change to the fabric, process, or sample, to check whether the required result is now met. |
| Rework | إعادة التشغيل (تصليح العيب) | Retouche | Yeniden işleme (tamir) | Reprocessing already-made fabric or garments, such as an extra finishing step, to bring a failed property back within the required result. |
| Sort and cull | الفرز والاستبعاد | Tri et élimination | Ayıklama | Separating a batch into pieces that pass a required result and pieces that do not, rather than treating the whole batch the same way. |
| Required result | النتيجة المطلوبة | Résultat requis | İstenen sonuç | The specific rating or percentage a standard states a test must meet or better. |
| Concession | تنازل موثق | Concession documentée | Belgelenmiş taviz | An agreement, recorded in writing, where a buyer or brand knowingly accepts goods that failed a stated required result, usually at a reduced price. |
This lesson taught the options that follow a genuine test failure close to shipment. The next lesson brings every test, standard, and result from this course together into one complete testing tracker for an order. Back to the course map