How to Reduce Product Complexity in Your First Product
When developing a first product, it is easy to make everything custom.
A bespoke housing. A unique hinge. A purpose-built mechanism. Custom fasteners. Several manufacturing processes.
Sometimes that is exactly what the product needs.
But every unique part adds another layer of development, sourcing, testing and manufacturing risk. For a first product, that complexity should have a clear reason for being there.
Getting from an idea to reliable production is difficult even when there is already evidence of demand. A study of 144 successfully funded physical technology products on Kickstarter found that only 32% delivered on time. The causes varied, but product development challenges were identified as a significant contributor.
The question is not simply:
How can we use fewer unique parts?
A better question is:
Where does customisation genuinely improve the product, and where are we adding complexity without enough benefit?
Every custom part adds more than unit cost
A custom component may eventually cost very little per unit.
Getting it to that point is different.
A bespoke injection-moulded part, for example, may require:
CAD development and engineering
prototyping and design revisions
tooling investment
tolerance development
supplier communication
testing and validation
quality control during production
If the design changes, the tooling may need to change with it.
If the part fails, the cause needs to be investigated.
If it interfaces with several other custom components, those relationships also need to be controlled.
This is why the true cost of a unique part is rarely just the price quoted by the manufacturer.
In product development, every additional part can create more work across design, engineering, sourcing, assembly and production.
That does not mean bespoke components should be avoided. It means they should justify the extra commitment.
Sometimes the custom part is essential
There are many products where using standard components throughout would weaken the product rather than improve it.
Consider a compact wearable electronic device.
The internal electronics may use existing batteries, connectors, switches and circuit boards, but the physical product around them may need to be highly specific.
The enclosure could need to fit tightly around the hardware. The attachment method may determine how the product is worn. Button positions, charging access and overall dimensions may all directly affect usability.
Trying to force that product into a generic enclosure simply to avoid custom tooling could compromise the entire experience.
In this situation, the goal is not to remove everything bespoke.
It is to remove unnecessary manufacturing complexity.
That could mean asking:
Can several housing components become one or two?
Can fixing features be built into the moulded enclosure?
Can standard fasteners be used?
Can unnecessary undercuts or complex tooling actions be removed?
Can existing connectors, batteries or seals be specified instead of creating new ones?
Can one component perform several useful functions?
Can the product be assembled in fewer steps?
These decisions can have a direct effect on tooling. In injection moulding, for example, an undercut may prevent a part from being released from a simple two-part mould. Additional mechanisms such as side-actions can solve the problem, but they also increase mould complexity and cost. Sometimes that complexity is necessary - but removing an unnecessary undercut at the design stage could remove the need for it altogether.
The final product may still be highly customised.
The difference is that the complexity has been used deliberately.
Standard parts can remove a lot of risk
Now consider a relatively simple mechanical product such as a pedal bin.
Many of its functional components could potentially use existing solutions:
springs
screws and fasteners
pins and pivots
bushes
sheet-metal components
standard mechanical interfaces
There is little value in developing a proprietary screw if a standard fastener already performs the job.
Using proven components can reduce development time, tooling costs and sourcing risk. It can also make repairs, replacements and future production easier.
But there is a limit.
If almost every part of the product already exists and the final design offers very little advantage over something already on the market, another question becomes important:
Why develop a new product at all?
If a business simply needs another generic item in its range, sourcing an existing product may make more commercial sense.
A new product should usually have a reason to exist.
Perhaps it:
solves a problem existing products do not
is easier to use or maintain
fits a specific environment better
reduces packaging or storage space
improves a mechanical interaction
offers a meaningfully different user experience
Those improvements may require custom parts.
The aim is not to eliminate uniqueness. It is to use it where it creates real value.
Fewer parts can matter more than fewer unique parts
Part count itself is also important.
Every additional component needs to be sourced, inspected, stored, handled and assembled.
It also introduces another interface where something can go wrong.
Reducing the number of parts can therefore improve both manufacturing efficiency and product reliability.
There are measurable examples of this in established products. During a Design for Assembly exercise at Whirlpool Sweden, a microwave oven design was simplified to use 29% fewer parts, while its assembly time fell by 26%. The improvements came from integrating components and simplifying or removing assembly operations.
For a new product, similar thinking might mean:
five components being redesigned into two
a separate bracket becoming part of the main enclosure
two covers becoming one moulded component
a separate fixing being replaced by an integrated feature
several assembly operations being removed altogether
This is where a custom component can actually make the product simpler.
One of the more extreme examples comes from GE Aviation. A fuel nozzle that had previously been manufactured from 20 separate parts was redesigned as a single additively manufactured component. According to GE, the resulting nozzle was also 25% lighter and up to five times more durable.
It is a very different product and manufacturing environment to most consumer products, but the principle is useful: bespoke does not automatically mean more complex.
A purpose-designed component that replaces several brackets, screws and assemblies may introduce manufacturing investment, but reduce labour, part count and long-term production complexity.
So "avoid custom parts" is not a particularly useful rule on its own.
A better goal is:
Use the lowest sensible level of complexity for the product you are trying to create.
To reduce product complexity, go beyond the CAD model
A product can have very few parts and still be difficult to manufacture.
Complexity also comes from the supply chain, production processes and commercial decisions behind it.
When reviewing a first product, it is worth asking:
How many suppliers are involved?
How many manufacturing processes are required?
How much tooling is needed?
What minimum order quantities apply?
Which parts require secondary finishing?
How difficult is assembly?
How much inventory needs to be purchased upfront?
How many components depend on one another?
How easy will the product be to manufacture consistently?
These questions matter because every additional process, supplier and dependency can increase cost and risk before the product has generated any revenue.
Good product design should consider all of them.
Do not scale commitment faster than the evidence
This is also why low-volume manufacturing is not automatically the correct route for every new product.
A small production run can be useful for validating demand, assembly, manufacturing or customer response.
But the size of that run should depend on what the business actually needs to prove.
If the business model only works at several thousand units, selling 20 products may demonstrate some interest without proving that a viable market exists.
Equally, committing immediately to expensive production tooling and thousands of units based on an untested assumption creates much greater risk.
A useful principle is:
Do not commit to more complexity, tooling or inventory than your current evidence can justify.
That evidence can support increasing levels of commitment.
For example:
initial interest may justify a prototype
prototype testing may justify further engineering
confirmed demand may justify production tooling
repeatable sales may justify higher-volume manufacturing
proven volume may justify further optimisation or additional custom components
Evidence should not simply answer whether a product should continue.
It should determine how far you commit at each stage.
Complexity should earn its place
There is nothing inherently wrong with a complex product. Some products need bespoke mechanisms, specialist materials and multiple manufacturing processes to perform properly. The problem is complexity without enough value behind it.
A custom hinge that creates an important product interaction may be worth developing.
A custom hinge that behaves exactly like an existing catalogue component probably is not.
A bespoke moulding that removes five other parts may simplify production.
Another custom component added purely for a minor cosmetic difference may create unnecessary tooling and sourcing work.
For a first product, the aim should be to keep development, manufacturing and supply chain complexity proportional to what has actually been proven.
Use standard components where they already solve the problem.
Create bespoke parts where they provide meaningful product value.
Combine components where doing so simplifies manufacture or assembly.
And avoid committing to expensive processes before there is enough evidence to justify them.
Your first product does not need to be simple for the sake of simplicity.
It needs to be only as complex as it needs to be.
