From sketch to tooling: what a lure programme commits to

Most of what a lure programme will cost is decided before anybody sees a sample. The drawing fixes the material, the mould fixes the action, and the sample only reports back on decisions already made. Buyers who treat sample rounds as the negotiation get the worst of both: tooling already cut, changes still open.

The chain, and what each link has to leave behind

A product development chain that can be audited has six steps. Each produces a document, and the document is the point.

Step	What it fixes	Record
Requirement	What the buyer needs the product to do, and for whom	Requirement version with a date
Specification	Dimensions, material, weight, colour reference, packaging	Controlled specification with a version number
Sample	The first physical attempt at the specification	Sample ID tied to the specification version
Test	Whether the sample does what the specification says	Method, conditions and result
Change	What is being altered, and why	Change note stating what it invalidates
Approval	Permission to move to the next gate	Named approver, date, open items

The fifth row is the one that gets skipped. A change without a note leaves the previous test results floating, and nobody can say afterwards whether the approved sample still describes the product. Every change after sample approval invalidates the tests run before it. That is not a bureaucratic rule, it is the reason the sample-to-production gap exists.

The tooling gate is a commitment, not a step

Cutting a mould converts a conversation into capital. After that point, a change to wall thickness, weight distribution or the line tie is not an adjustment, it is a new mould or a modification that has to be welded and re-cut. Three consequences follow, and all three are worth stating plainly to whoever signs the approval.

• Every change after the gate costs more than it would have before. Not by a small margin. This is why the colour, the logo position and the hook choice should be settled at the drawing.
• The tooling belongs to somebody. Ownership and possession are separate questions. A mould that you paid for but that sits in a factory you have stopped using is a mould you may never see again. Write down who owns it, where it lives and who may move it.
• The first mould is a prototype as well as a production tool. Expect the first parts to need tuning, and budget the sample rounds accordingly rather than treating the first pull as final.

If you want the tooling cost broken down before you commit, the glossary entry sets out the terms factories use, which helps when two quotations describe the same thing with different words.

Patent claims: what a database hit does not prove

Design claims appear on quotations and catalogues, and they are worth handling carefully because the wording is often looser than the underlying right.

A patent application is not a granted patent. A granted patent in one jurisdiction does not apply in another. A search result that shows a similar design does not establish who owns it, whether it is in force, or whether it covers the product in front of you. Public registers such as WIPO's PATENTSCOPE will tell you what has been filed and published; they will not tell you that your supplier has the right to use it.

So when a factory describes a product as patented, the usable questions are: which jurisdiction, what number, application or grant, what is the current status, and who is the named holder. If the answer is “our designer registered something”, the claim is not a commercial asset, and you should not let it carry weight in the price.

The same restraint applies in the other direction. Do not assume that a design you commissioned is yours by default. Ownership of commissioned design work depends on the agreement, and it is far cheaper to settle that at the start than after the product is selling.

Where the programme actually goes wrong

Three patterns account for most of the trouble, and none of them is a manufacturing failure.

The specification is a catalogue page. A photograph and a weight do not define a product. When two parties hold different ideas of what was approved, the sample round discovers it, and the argument is about memory rather than evidence. The cure is the controlled specification, not a longer email.

Changes arrive one at a time. A colour tweak in week three, a packaging change in week six. Each is small, each resets something, and together they push the programme past the season it was meant to catch. Collect them before the tooling gate; if a change genuinely cannot be anticipated, accept that it is a new timeline rather than a delay.

Nobody wrote down who approves. Development stalls when a decision needs a signature that no one has agreed to give, or worse, moves forward on a verbal yes that is later denied. Naming the approver for each gate takes one line in the agreement and removes the most common cause of a stalled programme.

Related reading: the sourcing route in order, what to do when a sample fails, and the evidence behind performance claims.