An expensive prototype is a predictable result, not a failure

Your prototype does exactly what you promised. People who try it get it immediately. Then the costing comes back and the unit cost is well above what the market will pay. This is one of the most demoralizing moments in hardware development, and also one of the most normal.

A prototype is expensive because of how it was built, not because the product is unviable. Parts were printed or machined one at a time with no tooling amortization. Off-the-shelf components were chosen for availability rather than cost. Assembly took hours of skilled hand work. Nobody optimized anything for a process. The real question is not why it costs so much — it is how you run the transition from this build to a version that can be manufactured.

Measure before you touch the design

Cost reduction without a number is just opinion. Start from the top: pick a target retail price, subtract the retailer or distributor margin, subtract the margin your business needs, and you have a permitted landed unit cost at a stated volume. That number is the goal, and everything else is measured against it.

Then break your current cost apart — parts, electronic components, assembly labor, test time, packaging, freight, scrap. In nearly every project, three to five line items carry most of the total. Those are the only places worth engineering effort. A structured bill of materials is the tool that makes this visible instead of anecdotal.

The gap between permitted cost and current cost tells you the size of the job. A gap of tens of percent means targeted adjustments. A gap of three times or more means a genuine redesign, and it is better to say that out loud early. The individual levers are covered in our guide to value engineering to cut product cost.

The decision that changes everything: how many units, really

The same product has wildly different unit costs at different volumes, so you cannot choose a manufacturing method before you make a business decision about scale.

  • Low volume. Stay with tool-free processes — machining, printing, cast urethane, sheet metal. High per-unit cost, near-zero upfront investment, and freedom to change the design between batches.
  • Mid volume. Simple or soft tooling, aluminum molds, and semi-automated assembly start to pay back. Moderate upfront cost, meaningful per-unit savings.
  • High volume. Hardened steel injection molds, die casting, automated assembly and test. Large upfront investment that only makes sense above a real threshold.

Where those thresholds fall for your specific geometry is the subject of our breakdown on choosing a manufacturing process by volume, and the upfront tooling side is quantified in our guide to injection molding costs. This is a founder decision as much as an engineering one, and it should be made against a sales forecast you would defend to an investor, not against hope.

What a manufacturing-readiness project actually looks like

  • A short diagnostic phase. A few weeks of cost teardown, identification of the cost-carrying parts, and two or three target scenarios showing expected savings against required investment. This phase ends in a decision, not a folder of files.
  • Focused redesign. Only the parts carrying the cost get reopened, and each is redesigned for the specific process it will be made by. Everything that already works and is not expensive stays untouched, so you do not put proven function at risk.
  • Re-validation. Every change of material, wall thickness, or joining method can change behavior. You need a fresh set of parts made by the intended production process, put through the same tests the original prototype passed.
  • Locking the documentation. Dimensioned drawings with tolerances, the full bill of materials, material specifications, and an inspection plan. Without a complete manufacturing data package, the attractive quote you received will not survive the first production run.

Timeline depends almost entirely on whether tooling is involved. Localized changes with no new tools usually land within a few months. A redesign that ends in injection molding adds tool build time plus the debug rounds that follow. Plan that schedule against your selling season deliberately — pushing a cost-reduction project into a launch window is the reliable way to pay for it twice.

Expect the product to change

The manufactured version will not be identical to the prototype, and it helps to accept that in advance. It may be slightly larger because of draft angles and structural ribs. Two parts may merge into one. A secondary feature may disappear because it cost more than it earned. The typical differences and how to plan for them are described in our article on going from prototype to production.

The single highest-leverage direction is almost always the same: fewer parts, fewer fasteners, shorter assembly time. Consolidating components removes part cost, inventory cost, assembly labor, and failure modes in one move. That discipline is covered in our guide to design for assembly.

And when the honest answer is to stop

Sometimes the conclusion is that this product, in this configuration, will not reach the required price at a realistic volume. That is still a useful result, and it has several respectable exits: split into a stripped-down affordable version and a loaded premium one, redirect to a business or professional market that tolerates a higher price, license the design instead of manufacturing it, or delay launch until forecast volume justifies tooling.

The one thing not to do is go to production on negative margin and hope volume fixes it. It does not. More material on building and de-risking physical products is collected on our prototyping hub.

Sitting on a working prototype with a unit cost that does not close? Get in touch through our contact form and the Projects House team will run a cost teardown, show you where the money actually is, and lay out what it would take to reach your target price.