When I was developing my first hardware product, we had planned for 2 hardware iterations. It took us 7.
If you’ve built hardware before, you know the roadmap always starts out clean. Until reality hits.

Things start breaking down as soon as you move from idea to development. And they continue through to production.

Here’s why that happens. And what you can do to prevent (some of) it.

1. Stabilizing a Product Takes Time

At the idea and design stage, many things are estimated: power consumption, mechanical fit, signal noise, antenna behaviour etc. etc. Once the real hardware is built, those assumptions get tested, and many don’t hold up.

Then you have firmware that works fine in unit testing but fails when you integrate it. You start hitting corner cases that weren’t obvious before. These often require redesign, and more firmware development.

It takes time to stabilize a product.

A good rule of thumb is to plan for 3 hardware iterations before hitting mass production.

2. You Built for the Lab, Not the Assembly Line

While prototyping the focus is usually on getting things to work. Function over everything. You ignore designing for manufacturability.

But what works in a lab isn’t always feasible at scale.

Maybe you used components or placements that are a nightmare for automated assembly. Unclear silkscreens, tight spacing, or non-standard footprints increase error rates.

Maybe you used expensive PCB layer stack. Maybe your enclosures are too fragile or difficult to assemble.

What to do:

Design for manufacturing (DFM).

Keep adequate component spacing and pad size. Avoid components that need hand soldering.

Have a few calls with your contract manufacturer to learn about assembly challenges. Get his feedback on your design.

3. You Didn’t Think About Testing on Scale

Testing in the lab is easy. Testing at scale is not. A few percentage points of failure can become expensive at high volume.

You need test fixtures, test cases, and consistent procedures. If you’re planning to ship thousands of units, you can’t afford to test by hand or guess what’s failing.

What to do:

Design to support the following test methods:

In-Circuit Test (ICT): Design to support bed-of-nails or flying probe fixture to make electrical contact with test points on the PCB. Keep minimum spacing between test points (usually > 50 mil).

Functional Test (FCT): Powers the device and tests interfaces like USB, UART, BLE, buttons, sensors, etc. Often uses custom test fixtures and scripts.

Automated Optical Inspection (AOI): Use standard footprints and orientations for all components. Maintain clear silkscreen for part labels and polarities.

4. You Got Hit by Supply Chain Surprises

This one hurts. You think you’re on schedule, then a part suddenly has a 40-week lead time.

Now your roadmap is toast.

What to do:

You can mitigate this risk by designing the circuit with drop-in replacements in mind.

Watch lead times. Keep in mind the risks of geopolitical issues on your supply chain. Build relationships with vendors of key components to secure their supply.

TLDR; Summary

Your roadmap falls apart because early-stage hardware is full of unknowns, and most roadmaps don’t account for what happens outside the lab.

Have an experienced team of developers who have exposure to manufacturing. Give integration and testing ample time. Plan for 3 hardware iterations.

That’s how you get from prototype to product without ending your runway.