Steps 1–2: problem, concept and requirements
A good product solves a clear problem. Write down who will use it, in what situation and what is wrong with today's solutions. It sounds obvious, but it is the most common reason projects drag on: the shape is drawn before the function has been decided.
Also look at what already exists. Search shops, online and patent databases. If you find something similar, that is not necessarily the end – but you know what your product has to do better.
The concept can be a hand sketch, a photo of a home-made solution or a few sentences. Ideally make two or three different variants. A sketch is cheap to change; a tool is expensive.
The requirements list makes the idea measurable. State outer dimensions, weight, load, temperature, environment (indoors, outdoors, water, UV), service life, preferred material and how many you think you can sell. Also state what may cost the most and what must be cheap. The requirements list then becomes the yardstick for every decision.
Step 3: the CAD model
In the CAD model the idea becomes concrete. We build it parametrically, so dimensions and variants can be changed quickly. Moving parts are checked in an assembly with collision detection, and we see early on if something does not fit.
The model is also a good basis for decisions. Photorealistic images can be shown to potential customers, used in an investor presentation or given to a patent attorney as input. Showing an image and asking what people think is cheap – far cheaper than making a prototype of the wrong product.
The intended manufacturing method is chosen at this stage, because it dictates the shape. An injection moulded part needs draft angles and uniform wall thickness. A sheet metal part needs bend radii. A 3D-printed part needs self-supporting geometry. Read more about CAD for inventors.
Steps 4–5: prototype and testing
The first prototype should answer one question: does the principle work? It does not need to look good. 3D printing is often the quickest way to get a functional prototype, sometimes on the same day the model is finished. We print prototypes ourselves and can change the model and reprint in short loops.
Test against the requirements list. Measure what can be measured, let real users try it and write down what goes wrong. Expect three to five rounds before the function works. Every round is cheaper than a fault discovered after the tool has been ordered. Read about 3D printed prototypes.
Step 6: DFM – design for manufacturing
Once the function works, the design is adapted for series production. This is called DFM, design for manufacturing. We review wall thicknesses, radii, draft angles, parting lines, part count and how the product is assembled.
Every part that can be removed saves both manufacturing and assembly. Two parts that always sit together can often become one. A screw can often be replaced by a snap fit. This is also when you get quotes from workshops, and we help you find a manufacturer for CNC, sheet metal or injection moulding. Read about design for injection moulding.
Think about assembly and packaging too. A product that can be assembled from one direction, without special tools, is cheaper to make and harder to assemble wrongly. Standard screws and bearings are cheaper and easier to source than special parts. A good aim is for every part to have a clear job – otherwise it can often be removed or combined with another.
Steps 7–8: pre-production and tooling
A pre-production run is a small batch made with the same or a similar method to real production. It shows whether tolerances, assembly and packaging work, and gives you products to show customers and retailers.
For injection moulding, the tool is then ordered. It is often the largest single investment and the most expensive to change. So the design should be frozen before the tool is ordered. Small runs can sometimes be made without a tool, by 3D printing, CNC or silicone moulding.
From idea to product – an example
Imagine an inventor who wants to sell a holder that fixes a light to a bicycle handlebar without tools. The requirements list says: fits bars from 22 to 35 mm, withstands cold down to −20 °C, stays put on gravel roads and can be fitted with one hand.
The first prototype is printed in PETG to test the clamping principle. It cracks in the cold, so the next round tries a tougher material with a thicker rib at the hinge. The third round shows the grip is sufficient, but the lever is too short for mittens. After five rounds the function works.
Before production a DFM review is carried out: two parts are combined into one, the wall thickness is evened out and draft angles are added. A pre-production run of a couple of hundred units is made before the expensive tool is ordered. Every problem found in a print would have been far more expensive to fix in a steel tool.
Regulations, time and cost
Some products are subject to requirements that affect the design. Electrical products, toys, food contact products and protective equipment have their own rules, often with CE or UKCA marking. Check which rules apply when you write the requirements list, because material and shape may need to be adapted.
We help with the design, but certification and testing are carried out by accredited bodies. Allow for that in the timeline if your product is affected.
Cost is driven mainly by the number of parts, their complexity, how many prototype rounds are needed and which manufacturing method is chosen. Clear data and a clear requirements list save the most. Late changes in the project cost the most.
It is a good idea to split the project into stages with a fixed quote or capped hours per stage. Then you always know what the next step costs and can pause if something needs rethinking. Read more about how we price our work.
Confidentiality and patents – before you go public
Do not show the idea in public before you have decided whether to apply for a patent. A patent requires absolute novelty. An invention shown online, in a magazine, at a trade fair or in a crowdfunding campaign before filing can generally no longer be patented. That applies even if you published it yourself.
Use non-disclosure agreements (NDAs) when you show the idea to suppliers and partners. We are happy to sign an NDA before you send material. In the UK, patent applications are filed with the UK Intellectual Property Office (UKIPO); a European patent through the EPO can cover most of Europe. We do not file patent applications ourselves – for that we recommend a patent attorney – but we can produce 3D models and images as input. Tell us about your idea and we will give you a free assessment.
Stages in product development
| Stage | Outcome | Typical manufacturing method |
|---|---|---|
| Concept | Sketches, requirements list | Paper, simple model |
| CAD model | Parametric 3D model, assembly | – |
| Functional prototype | Proven principle | 3D printing |
| Design prototype | Right shape, feel and appearance | 3D printing, CNC, painting |
| DFM | Manufacturing-ready files and drawings | – |
| Pre-production | Small batches for testing and showing | 3D printing, CNC, silicone moulds, sheet metal |
| Series | Product in volume | Injection moulding, sheet metal, CNC, casting |
The number of stages and prototype rounds depends on the complexity of the product.
Checklist: from idea to product
- The problem and the user described in a few sentences
- Existing solutions and patents researched
- Requirements list with dimensions, load, environment and volume
- NDA before the idea is shown to others
- No publication before any patent application
- Parametric CAD model adapted to the manufacturing method
- Functional prototype tested against the requirements list
- DFM review before tooling is ordered
- Pre-production run before full production
- Stages with a fixed price or capped hours
Frequently asked questions
How do I turn my idea into a finished product?
Start with a requirements list, produce a CAD model and test a functional prototype. Then adapt the design for manufacturing (DFM) and make a pre-production run before ordering tooling. Split the work into stages so you can control the cost.
Should I patent my idea before showing it?
If you want a patent, the application must be filed before the idea becomes public, because publication generally destroys novelty. Consult a patent attorney and use an NDA when showing the idea to suppliers.
How much does it cost to make a prototype?
It depends on the number of parts, the complexity and how many rounds are needed. 3D printing is often the cheapest way to test a function. We give you a free assessment and a fixed price or capped hours per stage.
Can you help with manufacturing?
We make prototypes and smaller batches by 3D printing. For CNC, sheet metal, photo etching and injection moulding we prepare the design data and help you find a manufacturer.