The first sample arrives and it is wrong. Not badly wrong — the leather is right, the colour is close — but the taper is off, the stitching is a shade lighter than expected, and the buckle is not the one anyone had in mind.

Three weeks go by. A second sample is made. It is closer. A third follows.

Almost none of this is a manufacturing failure. It is the predictable outcome of a brief that named a colour and a width and left everything else to interpretation. A strap has a surprising number of independent variables, and every one left unstated is a decision the factory makes on the brand’s behalf.

What follows is the specification we wish arrived with every enquiry. Defined up front, it routinely takes a strap programme from three sampling rounds to one.

Dimensions

These are the easiest to get right and the most expensive to get wrong, because a dimensional error is not correctable after production.

Lug width

The width at the case end, in millimetres. It must match the watch case exactly. This is the single most common source of a returned batch, usually because the figure came from a previous supplier’s strap rather than the case drawing.

Taper

The width at the buckle end. A 20 mm strap tapering to 18 mm and one tapering to 16 mm look and feel materially different on the wrist. State both numbers, never just the first.

Length

Both halves, separately. Standard, long and short are not universal terms — one manufacturer’s long is another’s standard. Give the measurement in millimetres for the buckle half and the tail half.

Thickness

Thickness at the lug and at the tip. A strap that is uniformly thick feels stiff and sits poorly; one that is properly skived flexes where it should. If the watch is thin and dressy, this matters more than the leather choice.

Material

Leather type and finish

Full-grain, Nappa, waxed, suede, printed. Each ages differently and each signals something different about the product. Name the specific material, not the category.

Lining

The layer against the skin. Leather, calf, or a technical lining — and its colour, which is visible at the tip and at every adjustment hole. Contrast linings are a design decision that is frequently made by accident.

Padding

Flat, lightly padded, or fully padded. This changes the strap’s profile against the case more than most brands anticipate, and it should be decided against the watch it will be paired with.

Construction

Stitch colour and density

Tonal or contrast, and how many stitches per centimetre. Tighter stitching reads as more considered and takes longer to produce. It is also the detail photographers get closest to.

Edge finish

Painted, burnished, folded or raw cut. This is the step most often reduced under cost pressure, and it is the one that determines whether the strap is still presentable after a year of cuff friction.

Hardware and finishing

Buckle

Type — pin, deployant or butterfly. Material. And finish: polished steel, brushed steel, gold PVD or black PVD. If the watch case is coated, the buckle should be specified against the same reference so the two match.

Spring bars

Standard or quick-release. Quick-release adds cost per unit and removes a reason for the customer to visit a service counter. Whether that trade is worth it depends on the channel, but it should be a decision rather than an omission.

Branding

Method — embossed, debossed, hot foil or printed. Placement, usually on the lining of the buckle half. And the artwork, supplied as vector. Branding decided after sampling generally means sampling again.

What is not on the list, and should be discussed

Two things sit outside the specification but shape it, and they are worth raising in the first conversation rather than the third.

The first is quantity and its effect on material. Some leathers are available in consistent batches at any volume; others vary between hides in ways that become visible across a large run. A manufacturer who knows the order size can steer the material choice accordingly.

The second is tolerance. Every dimension above carries one. A brand that expects nothing and a factory that assumes generously will disagree eventually. Agreeing tolerances at sampling, and keeping an approved physical sample on both sides, ends that argument before it starts.

A specification is not bureaucracy. It is the fastest route to the strap you already had in mind.

None of this makes the creative decision for you. It simply ensures that what arrives is the thing you intended, and that the second order matches the first.