Scratchy Collar After a Week: Isolate the Cause and Repair

A collar that is comfortable in the first week and scratchy by the second is not the same fault as a collar that is scratchy on the first wear. The first week rules out the cloth’s initial hand and points at something that changed after the coat was worn: fibre ends lifted by abrasion, a finish that softened and then moved, or a construction detail that only touches skin once the collar has taken the shape of a neck. That distinction decides both what you should have tested and what you can still repair, and it is the reason the order of the checks matters more than the number of them. It also decides who pays. A cloth that was too coarse from the loom is a mill conversation, a stand that presses a hard fold into a neck is a pattern conversation, and a barrier that solves the complaint on the same wearer tells you which of the two you are actually in before anyone is blamed.

Collect the Complaint Before You Touch the Cloth

Ask three questions of every customer who reports it, and log the answers: where exactly on the neck the irritation sits, whether the coat was worn against skin or over a shirt, and how many wears and what cleaning happened before it started. Those three answers split the population of complaints into groups that have different causes. A complaint from the side of the neck where the collar stand sits, on skin, after six or seven wears, is a different problem from a complaint at the centre back of the neck, over a shirt, from the first wear. Without the log you will diagnose the loudest complaint rather than the most common one.

Map Where the Collar Touches the Neck

The stand, not the leaf

The part of a tailored collar that touches skin is almost never the visible collar leaf. It is the inside face of the collar stand, the narrow band that holds the collar up, and the fold at its top edge. That band is a folded double layer of the cloth, usually with an interlining inside, and its top edge is where the cloth is most compressed and the seam allowance is thickest. Ask the customer to mark the irritated area with a fingertip on a photo of the coat, then compare that mark with the height of the stand. If the two line up, the cloth is not necessarily at fault; the band is.

What sits between cloth and skin

Inspect the inside of the stand and list everything in contact with skin, in order from the surface inward: the cloth face, the fold edge, any exposed seam allowance or overlock tail, the interlining or fusible behind the face, the lining attachment, and the brand or care label. Any one of those can produce the complaint on its own, and two of them are usually present at the same time. This mapping separates a cloth claim from a construction claim, and it is the step most brands skip on their way to asking for a softer fabric.

Run the Checks in This Order

Each check costs less than the one after it, so run them in sequence and stop at the first positive.

1. Finger and thumbnail pass along the inside of the stand. A hard edge, a ridge, a proud overlock tail or a stiffened fold will catch a thumbnail. This takes a minute and rules out half of the construction causes.
2. Tape lift on the inner face. Press a strip of clear tape onto the stand, lift it, and hold it against a window. Loose fibre ends that come away in quantity mean the surface is shedding, which is what a wearer feels as prickle.
3. Compare a worn collar with an unworn one from the same batch. Cut a 10 cm square from the stand of a returned coat and one from a stock coat, then press both flat and compare the surface under a raking light. A raised nap on the worn piece confirms abrasion rather than a bad lot.
4. Wash or dry clean one sample to the care label, then repeat the tape lift. If the surface becomes harsher after the care cycle, the finish is the variable, not the fibre.
5. Swap a smooth barrier into one collar and have the same wearer test it. This is the decisive test, because it isolates contact from cloth. Pin or baste a strip of smooth lining into the stand and ask the same person to wear it for the same number of days.
6. Only then send cloth to a lab. By this point you know what question to ask, and a specific question is far cheaper than a general comfort report.

The tape lift

Keep the tape strip, label it with the date and the piece number, and photograph it next to a ruler. It costs nothing to keep, and a complaint log with six dated strips is a stronger document than any single lab report.

The swap test

Do not run the swap test on a new coat only. Use a returned coat, so the only variable is the barrier. If the same wearer reports relief, the cause was contact and the fix is construction. If the wearer reports no change, the cause is the cloth surface and the fix is either the finish or the spec on the next order. This one test decides where you spend money.

What the Week-Long Timeline Tells You

Comfort that degrades over a week, rather than arriving that way, narrows the causes sharply. A finish that softened the surface has had time to move or wear at the contact band. Fibre ends have had time to lift through abrasion against skin, shirt collars and seat backs. The collar has taken the curve of the neck, so the fold edge now presses on a smaller area with more pressure. None of those four is a fibre diameter problem, and all four are addressable without re-weaving cloth.

What the timeline cannot tell you is whether the cloth is inherently coarse. For that you need the fibre diameter distribution, and specifically the share of fibres above roughly 30 microns, which is the working threshold used in wool comfort testing because coarser fibres are stiff enough to press into skin rather than bend against it. A mean diameter alone hides this, because a cloth can have an acceptable mean and still carry enough coarse fibres to provoke a complaint. The composition and diameter profile of the wool that went into the yarn is the number to ask the spinner for, and it should have been on the specification from the first order.

What to Send to the Lab

Send three things: a cutting from the contact band of a returned coat, a cutting from an unworn coat of the same piece, and a cutting of the interlining from inside the stand if it can be removed without destroying the collar. Ask for the fibre diameter distribution with the coarse share stated separately, a surface friction and roughness comparison between the worn and unworn cuttings, and a stiffness measurement on the stand assembly. Stiffness matters here because a stiff band presses fibres into skin, and the cantilever method in ASTM D1388 is the standard way to put a number on it. A lab asked for a softness test returns an opinion; a lab asked those three returns a number you can put in a purchase order.

The Repair Ladder

Repairs are ranked by what they change and what they risk. Work down the ladder until the complaint stops, and stop before the fix becomes more visible than the fault.

1. Remove the hard items from the contact band. Trim and grade the seam allowance inside the stand, enclose the overlock tail, remove any stiffener that reaches the fold, and relocate the brand label away from the neck. This is sewing work, not cloth work, and on a small run it can be done by the factory that made the coats.
2. Add a smooth barrier at the contact band only. A strip of smooth lining, cupro, fine cotton sateen or a soft tricot, basted over the inside of the stand and finished with enclosed edges. Do it on the band, not the whole collar, and make sure the barrier’s own seam does not create a new ridge.
3. Soften the surface, with the label in mind. Steam and press with a press cloth, brush the nap in one direction, and remove loose fibre. Any treatment that changes the cloth’s performance has to be checked against the care label first, and the division between what the mill declares and what only you can test is the question to settle before a softener touches a finished coat.
4. Offer a collar guard on pieces already sold. A detachable smooth band is a repair that does not require the coat to come back, and it is the cheapest way to stop a complaint wave while the next batch is fixed.
5. Re-make the collar. Justified only when the first four have failed on a meaningful share of returns, because the fault is then in the stand pattern rather than in one coat.

What Not to Do

Do not shave, sand or shear a tweed collar to soften it. It changes the surface permanently and exposes fresh fibre ends. Do not apply a softener and then keep the original care label, because the label now describes a garment you no longer sell. Do not judge the fix by hand feel at the bench, since only a worn test can answer a complaint that arrived after a week. And do not open a claim against the mill before the contact mapping is finished, because a claim that names the wrong cause costs you the relationship and leaves the fault in place.

The Line to Write on the Next Order

The next order is where this stops being a repair. Add one specification block to the fabric order and one to the garment order, and the complaint becomes a measurable failure rather than a matter of opinion.

The fusing question belongs in the same conversation, because a fusible that stiffens the stand is indistinguishable from a coarse cloth when a customer touches it. The interlining trial that has to happen before cutting is where that variable gets controlled, and it costs one collar rather than one season.

If the cloth itself is the answer, the blend is the next lever rather than the finishing line. A tweed with a share of mohair, alpaca or angora in the blend changes the surface the neck actually meets, and it changes it in the yarn rather than in the press room. Whatever route you choose, write the number into the order and test against it on a finished collar, because the only test a customer performs is the one you cannot run at a bench: wearing it for a week.