Why Repeat Towel Orders Don’t Match: Lab Dips and Dye Lots

Where the colour actually goes wrong

When a hotel chain opens cartons from a reorder and the towels look a touch lighter or warmer than the previous delivery, the instinctive conclusion is that the supplier “changed the colour”. The supplier, meanwhile, insists the recipe never moved. Both sides are usually telling the truth, because colour in textiles is not a setting — it is an outcome. The same recipe run twice produces two shades that are close, not identical, and the drift creeps in at several points:

• The substrate: a new cotton crop, a different yarn spin (ring-spun versus open-end), or a changed blend absorbs the same dye slightly differently.
• The dye lot itself: dyes are weighed and dissolved in batches, and machine load, liquor ratio, water hardness and temperature curves vary a little from batch to batch.
• The finishing line: softeners, optical brighteners, and the heat and tension of drying and stentering each shift the final shade a visible amount.
• The light you judge under: a match made under one light source can visibly break apart under another — metamerism, and the reason two “matching” towels can disagree in the hotel lobby.

None of these are defects; they are physics and chemistry doing what they do at industrial scale. The defect is when nobody measured the drift, so nobody knows whether this lot is inside an agreed band or outside it. Colour control is not the art of preventing variation. It is the discipline of measuring it and writing down how much of it you will accept.

What a lab dip actually is — and what it does not promise

A lab dip is a small swatch, typically a few centimetres of the actual yarn or fabric, dyed to a proposed recipe and submitted for your approval before bulk production. Approving it locks down the recipe on that substrate, judged under the agreed light source, at that moment. It is the cheapest place in the whole chain to fix a colour, which is why skipping it to “save time” is a false economy.

But read the scope of that approval honestly. A lab dip does not promise that the production machine will reproduce it exactly: a laboratory beaker and a production dyeing machine differ in liquor flow, heat-up rate and dye liquor ratio, and scaled-up dyeing routinely shifts the shade by a small but visible amount. It does not promise anything about future lots either. It approves a recipe, not a factory-wide colour guarantee. Mature buyers treat the lab dip as step one of a chain — recipe, then bulk standard, then lot-by-lot checks — not as a certificate that the conversation is over.

The documents that settle colour arguments

Because colour drifts, the industry has evolved a short chain of documents, each answering a different question. Know what each one covers and you will never again argue about colour from memory:

Control step	When it happens	What it locks down	What it does not cover
Lab dip	Before production	The dye recipe on the actual substrate	Machine-scale shade of the bulk lot
Bulk shade standard	At bulk start	A sealed physical reference both sides keep	Nothing — it is the arbiter in any later dispute
Grey scale rating	Bulk versus standard	A 1–5 visual grade of the difference	The direction of the difference (lighter, redder, duller)
Instrumental read (ΔE)	Bulk versus standard	A number both sides can compare by email	Which light source the reading was taken under
Top of production (TOP)	First metres or pieces of bulk	Confirmation the running lot matches the standard	Later lots — keep one per dye lot

The number in the middle of that table deserves a paragraph of its own. Colour difference can be expressed instrumentally as ΔE — a single figure for how far two samples sit apart in colour space, calculated from spectrophotometer readings. As a rough working map: below about 1.0, most observers under controlled light struggle to see a difference; 1.0 to 2.0 is visible side by side but usually acceptable as a commercial shade match; much above 2.5 and the two samples start to look like different colours. Many hospitality programmes work with a ΔE limit around 1.0–1.5 against the sealed standard for face-to-face comparisons, paired with a grey scale rating of 4 or better. The exact numbers are negotiable; having numbers at all is not. Instrumental colour-difference calculation is standardised in ASTM D2244, and the visual grey scale method most mills and labs use comes from the AATCC evaluation procedures at aatcc.org — citing one or both in the purchase order removes the “whose eyes do we believe” argument before it starts.

Metamerism: the match that breaks under hotel lighting

Metamerism is the classic late-stage surprise: the sample and the bulk matched perfectly in the office under daylight, then the housekeeping manager says the new lot looks pinker under the corridor lights. A metameric pair is two fabrics whose dyes reflect light differently across the spectrum, so they agree under one illuminant and disagree under another. It happens because different dye combinations are used to hit the same target, and it is invisible in a single-light review.

The fix is procedural, not technical: approve colour under more than one light source, and make one of them the light the product will actually live under. A towel program for a hotel chain should be checked under D65 (daylight) and a warm interior source; a retail program should add the store’s own lighting. Textile testing labs keep standard light booths for exactly this reason, and a mill that volunteers a two-light check is telling you something useful about how they work.

How to write colour control into the order

Everything above compresses into six lines of procurement discipline. None of them cost the supplier extra if they were doing the work anyway; all of them are expensive to retrofit after a mismatched container has landed:

1. Name the colour by a shared reference: a Pantone TCX code for textiles as the starting point, or better, a physical swatch the mill can measure with a spectrophotometer.
2. Require lab dips on the actual yarn or fabric of the order — a dip on a substitute base proves nothing about your batch.
3. Seal the bulk standard: two or three signed, dated swatches, one kept by the buyer, one by the mill, one archived for the reorder argument that will eventually come.
4. Set the tolerance in numbers: a ΔE limit against the standard and a minimum grey scale rating (4 is a common commercial floor), with the test method named.
5. Name the light sources for approval — D65 plus the light the end user will see — and require the match to hold under both.
6. Require a top of production for every dye lot, and keep your own shade library keyed by lot number so the next reorder starts from a reference, not from memory.

The last line matters more than it looks. Most colour disputes between buyers and mills are not about chemistry; they are about the absence of a reference. When both sides can put a sealed swatch and a ΔE reading on the table, the conversation moves from opinions to measurements — and measurements have a way of ending conversations quickly.

Repeat orders: keeping month six on the same shade

The first order is the easy case, because everything is fresh: dips, standards, approvals. The reorder is where shade consistency actually gets won or lost, months later, often with a different dye lot, sometimes a different cotton crop. The mills that keep hotel programmes on shade for years run a simple loop: archive the sealed standard from every production run; re-verify the recipe against it before each new run; re-dip if the substrate has changed (new yarn lot, new blend, new finishing chemical); and where order size forces multiple dye lots, plan the lot blending deliberately — packing so that lots are not mixed within a carton, and documenting which lot went to which delivery.

Buyers can hold up their half of the loop with almost no effort: keep the shade library, quote the lot number when something looks off, and send a swatch back rather than a description. “It looks a bit lighter” is not actionable; a returned swatch next to a sealed standard under two light sources is a fifteen-minute diagnosis. This is the same verification habit that protects every other claim in a linen programme — the private label checklist approach of putting each promise in writing applies to colour more than to anything else, because colour is the promise most easily misremembered.

One boundary worth knowing: some drift is not fixable by process control at all. Wash after wash, every towel lightens — the colour bleeding question is about dye leaving the towel, while this article is about dye never quite landing in the same place twice. A shade standard governs the day the carton opens, not year three of the laundry cycle. Specifying colourfastness alongside the shade tolerance is how both ends of that lifespan get covered.

Nobody has ever won an argument about “a bit lighter”. Sealed swatches and a ΔE number end in minutes what memories would argue about for weeks.

Frequently asked questions