Handing a design to a manufacturer and taking their feedback seriously

Send intent, not only geometry

A manufacturer given models and drawings can tell you what is hard to make. A manufacturer told what the product has to do can tell you how to make it better. Include the function of the assembly, the loads and environments it faces, which surfaces are seen by a customer, which features exist for regulatory or safety reasons, and where previous versions failed. Engineers who understand purpose propose useful alternatives; engineers holding only a solid model propose whatever suits their machines. The extra context takes an afternoon to write and changes the quality of everything you get back.

Separate the review into things that block, cost and irritate

Feedback arrives as an undifferentiated list, and reading it that way leads teams either to accept everything or to dismiss the lot. Sort it. Some items make the part unmakeable as drawn and must be resolved before tooling. Some are cost drivers, where a draft angle, a radius, a tolerance or a finish is achievable but expensive, and the decision is a trade you should price. Some are preferences that suit the shop's habits without benefiting your product. Insist that each proposal states which category it falls into and what it saves, because a change presented without a quantified benefit is usually a preference wearing a technical argument.

Somebody has to keep design authority, and it should be written down

Trouble follows when the manufacturer starts making small adjustments during process development and nobody records them: a wall thickened for filling, a radius added to protect a tool, a component substituted for availability, a process temperature nudged. Each may be sensible; together they mean the product in production is not the product you validated. Agree explicitly that changes affecting form, fit, function, material, safety-related features or regulatory status require your written approval before implementation, and that the manufacturer holds latitude over the rest. Then keep your documentation current, so the approved design and the built design do not quietly separate.

Getting the feedback early enough for it to matter

Design reviews held after tooling has been ordered are a formality that generates awkward findings nobody can act on. The useful review happens while the geometry is still soft, which usually means involving the manufacturer before the design feels finished, and living with the discomfort of showing incomplete work. Where the choice of manufacturer has not been made, several candidates can be asked the same design questions during assessment, which produces free engineering insight and reveals which shop actually thinks. The cost is disclosure, so stage what you send and cover it with confidentiality terms first.

Close the loop so the next product starts smarter

Every handover produces knowledge that ordinarily evaporates once the programme ships. Capture it as design rules of your own: the tolerances this process route holds comfortably, the features that always attract comment, the material grades your manufacturer stocks, the fastener and connector families already qualified. Feed them into the next design at concept stage rather than rediscovering them during review. Companies that do this find their second and third products reach production noticeably faster, not because the manufacturer got better, but because the design stopped presenting the same problems in a new shape.

Frequently asked questions

Should we accept every manufacturability change the factory proposes?

No, but decline with reasons rather than reflex. Ask what each change saves, in tooling, cycle time, scrap or capability margin, and what it costs in function, appearance or validation rework. Some proposals are excellent and reflect experience you do not have. Others exist because the shop prefers a familiar route. A pattern of proposals that all move work off their equipment and onto your specification is worth naming aloud, politely, early in the relationship.

Who owns a design improvement suggested by the manufacturer?

Whatever the parties agreed in writing, which is why this belongs in the commercial terms before the collaboration produces anything valuable. Improvements generated during a joint development can be genuinely inventive, and ownership, rights to use and rights to file are separable questions that intellectual property offices treat differently. Settle it early with legal advice appropriate to the jurisdictions involved, because negotiating ownership after a valuable idea exists is far harder than agreeing the principle while the idea is still hypothetical.

How do we keep our documentation in step once production changes start?

Insist that the controlled definition lives with you and that no change reaches the floor without appearing in it. Give every revision an identifier both sides use in orders, drawings and inspection records, and require the manufacturer to state which revision each shipment was built to. Audit it occasionally by comparing a delivered unit against the released documents. Where drift is found, correct the records rather than tolerating a shared understanding that exists only in the heads of two engineers.