How to develop a physical product: a founder's guide from idea to market

If you have an idea for a physical product and no engineering background, the hardest part is not any single technical problem. It is knowing the shape of the whole journey: what the stages are, what order to do them in, and how to avoid spending a lot of money before you know whether the thing will work. This guide lays that out end to end, in plain terms, from the experience of taking consumer products from a first sketch all the way to volume manufacture.

The one rule that saves the most money

Develop a product as a sequence of stages, each of which reduces a risk before the next, more expensive, stage begins. Spend a little to learn whether the idea holds up, more to prove the concept works, more again to verify it and prepare it for manufacture. The single most expensive mistake in hardware is to skip the cheap early thinking, commit to tooling, and discover the problem on the factory floor, when fixing it means recutting a steel tool rather than moving a line on a drawing.

So the journey below is deliberately front-loaded with cheap certainty.

Stage 1: Prove the idea is worth building

Before any design work, it is worth finding out whether your idea can actually be built, roughly what it would cost, and where the technical risks are. This is what a feasibility study delivers: a written assessment of the routes to manufacture and their costs, the design directions and their trade-offs, and a straight go, no-go, or go-with-changes recommendation. It is usually the smartest first spend on a new product, because it replaces optimism with evidence for a few hundred pounds.

Read next: what a feasibility study is and when you need one, and how much developing a product actually costs.

Stage 2: Turn the idea into a working concept and MVP

Once the idea is proven worth pursuing, the next job is to turn it from a description into an evaluated, buildable concept, and then into a working prototype. For a physical product, an MVP is not a rough app build; it is a prototype at a defined level of fidelity that proves the core function and is good enough to put in front of users, investors or a crowdfunding audience.

This is where concept and MVP development sits. The important discipline is deciding, up front, what the prototype has to prove, so you build the right fidelity and no more: a proof of principle to answer “does the core idea work”, a looks-like model for form and feel, or a works-like prototype that functions as a tooled part would.

Stage 3: Verify it before you commit to tooling

Tooling is where a product gets expensive to change. Before you cut a mould, it is worth proving, with analysis and testing, that the design performs, survives real use, and meets its safety and compliance targets while fixes are still cheap. That is design verification: engineering analysis, simulation and structured testing that catch the problems that would otherwise fail certification or trigger a recall, at the point where they cost hours rather than a new tool.

Stage 4: Prepare it for manufacture

A design that works as a prototype is not yet a design a factory can build a thousand times a week. Manufacturing preparation closes that gap: design for manufacture and assembly, tolerances, detail production CAD, a full drawing pack with GD&T, and technical liaison with your chosen factory in the UK or overseas. Getting this right is where the real, repeated savings live, in cycle time, assembly labour, tooling maintenance and scrap, the costs that compound at volume.

How the stages fit together

Each stage is a service you can buy on its own, and together they form the full path. Some founders take one stage, learn what they need, and carry on themselves. Others want one experienced engineer to own the whole journey; that is product development, delivered in scoped, gated phases so you commit one step at a time and always know the cost and the next milestone before you reach it.

Whichever way you do it, the order is the same, and so is the principle: never commit the big money until the big unknowns are gone.

The questions this raises

Most founders, at this point, have a handful of specific questions. Each of these has its own guide:

• How much does it cost to develop a hardware product?
• What is a feasibility study, and when do you need one?
• Design engineer or design agency: who should build your product?

Where to start

If you are at the very beginning, the honest first step is almost always the cheapest one: a short study that tells you whether the idea holds up and what it would take. If you would rather just talk it through first, a quick call is the easiest way to work out where you actually are and what a sensible next step looks like.