An invention is a solution that exists in someone's head and maybe on a napkin. A product can be manufactured repeatedly at a known cost, shipped without breaking, sold at a margin, and supported after the sale. The distance between those states is the whole discipline of product development, and it is longer than almost every first-time inventor expects.

The good news is that the path is well worn. Consumer hardware, industrial equipment, and medical devices move through recognizably similar stages, and fail at recognizably similar points. What follows is the sequence, what each stage produces, and what it costs.

Stage One: Pressure-Test the Idea Before Spending Anything

Most inventions that fail did not fail in engineering. They failed because nobody wanted them at the price they had to cost, or because three companies already sold something close enough. Both of those questions can be answered for a few hundred dollars and a couple of weeks of effort.

Search prior art yourself using free databases before you pay anyone. Talk to fifteen people who have the problem, and listen for whether they already improvised a workaround, which is the strongest buy signal there is. Then estimate a retail price and work backward through distribution margins to the manufactured cost you must hit. If that number is impossible for the mechanism you have in mind, you just saved a year. The structured version of this work is covered in how to validate a product idea.

Stage Two: Feasibility and Concept Engineering

Feasibility answers a narrower question than validation: can this be built, at a sane cost, with technology that exists today? An engineer sizes the motor, checks whether the battery supports the runtime, calculates whether the mechanism survives the loads, and flags what must be proven by experiment rather than calculation.

The output is a short document with an architecture, a preliminary bill of materials, a cost estimate, and technical risks ranked by how badly each could derail the project. Expect $3,000 to $15,000 for a serious feasibility phase. The full scope is described in a feasibility study for a product idea.

Stage Three: Industrial Design and CAD

Now the product gets a shape. Industrial design settles proportions, ergonomics, controls, materials, and finish. Mechanical engineering turns that shape into parametric CAD with wall thicknesses, draft angles, fastener locations, and tolerances that a factory can actually hold.

Do not treat these as one job done twice. A beautiful concept render that ignores draft and parting lines becomes a $40,000 argument at the tooling stage. Good projects run industrial design and mechanical engineering in the same room, iterating. If you are starting from a drawing rather than a model, the conversion process is described in turning a sketch into a 3D CAD model.

Stage Four: Prototypes, Plural

There is no such thing as the prototype. There is a sequence of them, each answering a different question:

  • Proof of concept proves the risky mechanism works at all. It can be ugly, oversized, and tethered to a bench supply.
  • Looks-like model proves the form, size, and feel. Usually machined or printed and painted, used for user reactions and investor meetings.
  • Works-like prototype proves the function in something close to the real envelope, with real electronics and real firmware.
  • Engineering prototype combines both, built from production-intent materials and processes, and is what goes into environmental and safety testing.

Costs range from a few hundred dollars for a printed shell to $30,000 or more for a full engineering build with custom boards. The options and tradeoffs are laid out in how to get a prototype made.

Stage Five: Design for Manufacturing and Cost

This is the stage inventors skip and manufacturers charge for. Design for manufacturing means reworking a working design so it can be produced at volume: consolidating parts, eliminating undercuts that force side actions in the mold, replacing custom fasteners with stocked ones, and matching processes to your real annual volume.

The payoff is measured in unit cost, and it compounds over every unit you will ever sell. Cutting $4 out of a bill of materials on a product that sells fifty thousand units is $200,000 of margin. The methodology is covered in design for manufacturing.

Stage Six: Tooling and the First Real Parts

Injection molds, die-cast tools, and stamping dies are where the capital goes. A single-cavity steel mold for a mid-size plastic part commonly runs $8,000 to $30,000, and a product with six molded parts can absorb a six-figure tooling budget before a single sellable unit exists. Lead times of eight to fourteen weeks are normal, and the first shots almost never pass.

Plan for at least two rounds of tool correction after the first trial samples. The economics, including when to stay with low-volume processes, are explained in injection molding costs.

Stage Seven: Certification and the Pilot Run

Anything with a radio needs FCC authorization. Anything mains-powered needs a UL or ETL listing before most retailers will touch it. Children's products face CPSIA testing at an accredited lab, and medical devices face FDA. Failing an emissions scan late costs weeks and often a board respin.

Then comes the pilot run: a few hundred units built on production tooling by the people who will build the real thing, which is where assembly problems, fixture gaps, and yield issues surface while they are still cheap to fix. The reasoning behind it is in the pilot production run.

Where the Patent Fits

File a provisional before you disclose publicly, exhibit at a trade show, or launch a campaign. It buys twelve months cheaply while the design settles, and the mechanics are in filing a provisional patent application. A full utility application on a design that will change three more times wastes money on claims you will abandon.

What the Whole Thing Costs and How Long It Takes

A simple mechanical consumer product can reach production for $40,000 to $80,000 in roughly nine to twelve months. A connected electronic product with an app is more commonly $150,000 to $400,000 over eighteen to twenty-four months. Anything regulated runs longer and costs more. Those ranges assume one competent team and no restarts, and restarts are the single largest source of overrun.

Getting the Sequence Right

Projects House takes inventions through this full path, from the first feasibility review to a manufacturing data package a factory can build from. Describe your invention and where you are stuck through our contact form and we will tell you what the next stage should be.