The cost of an electronic prototype is not set by the board itself but by the decisions made before the first schematic is drawn. Two prototypes that look alike on the bench can differ in effort by a factor of several, because one was built on a clear specification with proven building blocks, and the other was redesigned three times along the way.

That is why serious engineering partners rarely quote a price before they understand the project. This article explains where prototype costs actually come from, which technical decisions drive them up or down, and how to plan a prototype that leads straight into series production instead of into a second round of development.

You will learn:

• how prototype costs are structured (one-off engineering effort vs. hardware per unit)
• the seven technical factors that decide the effort
• how cost ranges are built without guessing
• concrete levers to reduce cost without creating risk for series production

Where prototype costs come from

Most of a prototype budget is one-off engineering effort, not hardware. Understanding this split is the first step to a realistic budget.

Non-recurring engineering (NRE) covers everything that is done once: requirements analysis, concept, schematic design, component selection, PCB layout, firmware, mechanical integration, test concepts and documentation. In prototype projects, NRE is usually the dominant share.

Recurring costs per build cover what is needed for every prototype set: bare PCBs, components, assembly, enclosure parts, cables and the bring-up of each unit. These costs are highest per unit at prototype stage, because tooling, stencils and set-up effort are spread over very few boards.

The effort is distributed across five phases:

1. Requirements and concept – functional, environmental and regulatory requirements, system architecture, make-or-buy decisions.
2. Hardware design – schematic, component selection, power and signal integrity considerations, PCB layout and design reviews.
3. Firmware and software – drivers, board support, application logic, interfaces and HMI.
4. Build and bring-up – PCB manufacturing, assembly, first power-on, debugging and rework.
5. Test and validation – functional tests, environmental checks, EMC pre-compliance and preparation for certification.

The key point: the later a change is made, the more phases it has to pass through again. A changed requirement after layout means new schematic work, a new layout, new boards and new tests. This is why the earliest phase has the biggest influence on the total cost.

The seven cost factors

1. Maturity of the requirements

The single biggest cost driver is an unclear specification. Every open question becomes either an assumption (which may be wrong) or a later change (which is expensive). A complete requirements specification should answer at least:

• operating environment: temperature range, humidity, vibration, shock, ingress protection (IP rating)
• power supply: input voltage range, transients, battery operation, power budget
• interfaces: fieldbus and industrial Ethernet, USB, CAN, RS-485, wireless, display and touch
• target standards: EMC, electrical safety, industry-specific standards such as IEC 60601 for medical devices
• expected series volume and product lifetime, because they influence component choice from day one

A prototype built on a weak specification often becomes a learning prototype. That has its place, but it should be a conscious decision, not a surprise.

2. Hardware and PCB complexity

The effort in schematic design and layout scales with the technical complexity of the circuit. The main drivers are:

• Layer count and stack-up: more layers allow denser routing and better signal integrity, but increase design and manufacturing effort.
• Package types: fine-pitch BGAs, QFNs and very small passives require tighter design rules, X-ray inspection and more careful assembly.
• High-speed and controlled-impedance signals: DDR memory, USB, Ethernet or LVDS display links need impedance control, length matching and careful return-path design.
• Power electronics: switching regulators, motor drivers or high currents require thermal design, creepage and clearance distances and careful layout.
• Mixed-signal and RF: precise analog measurement or wireless functions demand separation, shielding and often several layout iterations.
• Quality class: the IPC-A-610 workmanship class (Class 2 for dedicated service products, Class 3 for high-reliability applications such as medical or defence) affects inspection and assembly effort.

3. Firmware and software

Firmware is frequently underestimated because it is invisible on the bench. Effort depends on:

• choice of platform: bare-metal microcontroller, RTOS or embedded Linux
• number and complexity of interfaces and communication protocols
• graphical user interfaces and touch operation on HMIs
• safety and security requirements, including secure boot, update mechanisms and the growing regulatory focus on cybersecurity
• test depth: unit tests, hardware-in-the-loop tests and production test software

Reusing proven software modules and drivers is one of the most effective ways to reduce this share.

4. Components and supply chain

Component decisions made for the prototype usually carry into series production. Effort and risk rise with:

• components with long lead times or limited availability, which force redesigns or expensive spot-market purchases
• parts with an announced or likely end of life, which shorten the product lifecycle before it starts
• single-source components without a pin-compatible alternative
• industrial or extended temperature grades, which are not always stocked in small quantities

A lifecycle check during component selection costs little and prevents an early redesign. For products expected to run for ten years or more, it is essential.

5. Mechanics, enclosure and HMI

Electronics rarely work without a housing, mounting concept or operator interface. Cost drivers here include:

• custom enclosures and front panels versus adapted standard housings
• sealing for IP65 or higher, stainless-steel designs for food and hygiene applications
• displays, touch sensors and cover glass, including optical bonding for readability and robustness
• thermal management in closed, fanless housings

For prototypes, 3D-printed or machined parts are often enough. The important thing is that the mechanical concept already fits the planned series process.

6. Testing, EMC and certification

A prototype that works on the bench is not yet a product. Test and validation effort depends on the target market and application:

• Functional and environmental tests: temperature cycles, vibration and long-term operation.
• EMC pre-compliance: reduced, faster tests during development show early whether the design is likely to pass. They do not replace the accredited compliance test, but they reduce the risk of a failed test and a costly redesign.
• Formal conformity: for the EU market, CE marking under directives such as the EMC Directive 2014/30/EU. Medical, railway or defence applications add their own standards and documentation.
• Production test concept: test points and test fixtures planned into the prototype make later series testing faster and more reliable.

Planning EMC and safety requirements into the first layout is far cheaper than fixing them after a failed test.

7. Iterations and number of builds

Almost every electronics project needs more than one hardware revision. A typical path is a first functional sample (often called A-sample), a revised sample close to the final design (B-sample) and a pre-series build made with series processes (C-sample or pilot series). Each loop means new boards, assembly, bring-up and tests.

The number of prototype units also matters. A single unit for the lab is cheap per build but leaves no units for field tests, customer demos, certification and long-term tests. Planning the right quantity per build avoids an extra build later.

How cost ranges are built

Instead of a fixed price, a reliable estimate starts with a classification of the project. The table shows how complexity shapes the effort. It is a guide for planning, not a price list.

Project type	Typical characteristics	Expected hardware iterations	Where most of the effort goes
Adaptation of an existing design	Proven platform, new interface or form factor, few new components	1	Layout adaptation, firmware changes, tests
New development of medium complexity	Microcontroller-based, several industrial interfaces, standard enclosure, simple display	2	Schematic, layout, firmware, EMC pre-compliance
Complex system	Embedded processor or Linux, high-speed interfaces, custom HMI, regulated market such as medical	2–3	Firmware and software, validation, documentation and certification

 

A serious estimate is then built bottom-up: hours per phase and discipline, materials per build, number of builds, external test and certification costs, plus a buffer that reflects the open points in the specification. The clearer the specification, the smaller that buffer.

Eight ways to reduce prototype costs without creating risk

1. Invest in the specification first. A day spent clarifying requirements saves weeks of redesign. Separate must-haves from nice-to-haves and define the series volume early.
2. Reuse proven building blocks. Validated power supplies, processor cores, interface circuits and software modules shorten design and test time and reduce the risk of errors.
3. Select components for the whole lifecycle. Check availability, lifecycle status and second sources before layout, not after the first purchasing problem.
4. Design for manufacturing and test from the start. DFM and DFT rules, test points and a panel concept mean the prototype can move into series production without a redesign.
5. Run EMC pre-compliance early. Measuring on the first functional sample shows weaknesses while changes are still cheap.
6. Use standard mechanics where possible. An adapted standard enclosure or display often meets the requirement at a fraction of the effort of a fully custom design.
7. Plan the build quantities. Build enough units per revision for lab, field and certification tests, so no extra build is needed for a missing unit.
8. Keep one partner from concept to series. Every hand-over between a design house, a layout service and a manufacturer costs time, knowledge and correction loops.

How T&O Electronic Solutions keeps prototype costs under control

T&O Electronic Solutions has developed and built electronic systems since 1990, for more than 800 customers in industrial and medical technology. Our engineering team works on the levers described above in every project:

• Free initial consultation and support with your specification. We help you turn an idea into a requirements specification that can be estimated reliably.
• Custom hardware and software from one team, based on proven, reusable building blocks that shorten development and reduce risk.
• Component selection for long-term availability, backed by obsolescence management, repairs and spare-parts supply after launch.
• From prototype to series: prototypes, small and pilot series and series production, in-house or with qualified manufacturing partners, with one contact person for the whole project.
• Displays and HMIs from our own Display Solutions unit, including touch monitors, operator panels and stainless-steel terminals.
• Quality management certified to ISO 9001, with sites in Hausham (Germany) and Ramsen (Switzerland).

An example from practice: for LUDWIG SYSTEM we modernised a historically grown crane safety system, the LudwigHook, using a concept-based approach and have supported it over several years.

Planning a prototype? Send us your idea or your existing specification. In a free initial consultation, our engineers will discuss the technical options and tell you which cost factors matter most for your project. Contact our engineering team.

Frequently asked questions

Why can't I get a fixed price for a prototype straight away?

Because the effort depends on details that are often still open at the start: interfaces, environment, standards, volume and lifecycle. A fixed price without these details either contains a large risk buffer or leads to change requests later. A short requirements phase makes a reliable estimate possible.

What should I prepare before asking for an estimate?

A short description of the function, the operating environment, the required interfaces, the target market and its standards, the expected series volume and the planned product lifetime. Existing sketches, competitor products or a previous design help as well. If you do not have a specification yet, we can create it with you.

How many prototype iterations should I plan for?

For a new development, plan for at least two hardware revisions: a functional sample and a revised, near-series sample. Adaptations of a proven design often need only one. Complex or regulated products may need a third build before series production.

What is the difference between a prototype and a pilot series?

A prototype proves that the design works. A pilot series proves that it can be manufactured repeatably with series processes, test equipment and documentation. Skipping the pilot series often moves problems into the first series batch.

Does EMC testing have to be part of the prototype phase?

The accredited compliance test usually follows a near-series design. EMC pre-compliance measurements on early prototypes are strongly recommended, because they show weaknesses while a layout change is still inexpensive.

Can an existing or obsolete design be used as a starting point?

Yes. Redesigning an existing product around available components is often faster and cheaper than a completely new development, especially when the mechanics and software can be reused.

Who owns the design after development?

Ownership of schematics, layout data and source code is defined in the development contract. Clarify this before the project starts, especially if you plan to manufacture elsewhere later.

How long does prototype development take?

It depends on the same factors as the cost: specification maturity, complexity, firmware scope, component availability and the number of iterations. A realistic schedule is part of every serious estimate.