From idea to IPO · Stage 03

From Breakthrough to Working Prototype: Engineering for What Comes Next

The decisions made during development lock in most of the product’s future cost, quality and schedule. The best development teams design for production long before production begins.

By Eyal Yavets · 3 min read · לקריאה בעברית

Development is where a breakthrough becomes a product. It is also where many companies make choices, often without noticing, that will cost them heavily in industrialization: materials that cannot be sourced at volume, tolerances that cannot be held, and architectures that cannot be tested efficiently.

Prototypes have different jobs

Not all prototypes answer the same question. Early prototypes prove that the concept works. Later builds prove that the design meets its requirements, and then that it can be produced consistently. Mixing these purposes leads to prototypes that impress in demonstrations and fail in validation.

• Proof-of-concept builds: does the core principle work?
• Engineering validation builds: does the design meet its requirements across conditions?
• Design validation builds: does it pass the tests the customer and the regulator require?
• Production validation builds: can it be made consistently, on the intended process, at the intended rate?

Systems engineering as a discipline

In multidisciplinary products, the hardest problems sit at the interfaces: between optics and mechanics, between electronics and software, between the product and its test equipment. Systems engineering keeps requirements, interfaces and trade-offs visible, and prevents each discipline from optimizing its part at the expense of the whole.

Designing for manufacturing and test

Design for manufacturing (DFM) and design for test are not reviews at the end of development. They are design criteria from the start. Every tolerance, material and assembly step has a cost, a yield and a lead time attached to it.

• Involve manufacturing and quality engineers, and key suppliers, in design reviews early.
• Track the product cost against a target, as a design requirement, not as an outcome.
• Decide early what to make and what to buy, and qualify critical suppliers in parallel with the design.

Hardware, software and cyber: what changes at this stage

The stage is the same for every technology company. What it demands is not.

• Hardware: prototype builds with clear purposes, design for manufacturing and test, and early qualification of critical suppliers.
• Software: an architecture that matches the business model, automated testing and continuous delivery, and deliberate management of technical debt.
• Cyber: a secure development lifecycle, threat modeling, a software bill of materials and vulnerability management, especially for connected products.

Common traps at this stage

• Building prototypes by hand in ways that can never be repeated in production.
• Freezing the design before the requirements are stable, or never freezing it at all.
• Treating test strategy as an afterthought, and discovering that the product cannot be tested at volume.
• Depending on a single expert who holds critical knowledge that is not documented.

Metrics that matter

• Requirements coverage: the share of requirements verified by test.
• Product cost versus target, updated with every design iteration.
• Open design risks and their trend over time.
• Readiness of critical suppliers and long-lead components.

Questions for the leadership team

• Which design decisions taken this quarter will be hardest to reverse in production?
• Do we know how we will test every unit, and how long it will take?
• Is our product cost converging on the target, or drifting away from it?

This guide reflects the author’s professional opinion only. It is general information, not professional advice or a recommendation to take any action, and any use of it is the reader’s sole responsibility.