The short answer
A hardware MVP is the smallest physical product that lets a real user test your core value claim in their real environment. At To Be Named Labs, that means an integrated unit — working electronics, stable firmware, a usable enclosure and just enough software — built in a small batch. It is not a demo, a render, or a breadboard on a desk.
Software founders borrow the term MVP and then get stuck, because hardware can't ship half a feature and patch the rest on Friday. The definition has to change. Minimum still means minimum. But viable means a person outside your team can use it without you standing next to them.
Hardware teams split early prototypes into three types. Each answers a different question.
| Prototype | Answers | Typically built with |
|---|---|---|
| Works-like | Does the core function work? | Dev boards, early custom PCB, test firmware |
| Looks-like | Is the form, size and appearance right? | 3D prints, foam, machined models, no electronics |
| Feels-like | Is it comfortable and intuitive to hold and use? | Weighted models, button and dial mock-ups |
None of these is an MVP. An MVP is what you get when they merge into one unit that works, looks right enough, and can be used without explanation.
One core function, done reliably. Not five functions done sometimes. Reliability is the feature at MVP stage. A device that works 90 percent of the time teaches you nothing, because every user complaint might be a bug.
A custom or near-final PCB. Dev boards and jumper wires fall apart in the field. You don't need the final board, but you need one designed for the job.
Firmware that recovers on its own. Watchdogs, error handling, sensible behaviour on power loss. Users won't reset your device the right way.
An enclosure the user can live with. Printed or machined is fine. It must protect the electronics, mount where it needs to, and survive handling.
Just enough software. Usually a setup flow and one screen that shows the result of the core function. Not a dashboard with twelve tabs.
A way to update it. Over-the-air updates, or at minimum a simple field update procedure. You will find bugs in your first batch.
A way to learn from it. Logs, usage data or a feedback channel. An MVP that can't tell you what happened is a gift, not a test.
Founders under budget pressure often cut reliability and keep features. That's backwards. Ten users with a reliable single-feature device give you clean signal. Ten users with a flaky five-feature device give you noise.
The other common mistake is over-investing in looks. A beautiful shell around unvalidated electronics gets you great photos and a second enclosure bill when the board changes.
Build enough to put the product in the hands of real users simultaneously, plus spares. For most startups that means 10 to 30 units. Fewer and you're testing one person's opinion. More and you're paying for pilot production before the design deserves it.
When we scoped v2.0 of ATSN, our assembly traceability scanner, the full idea included many reading modalities and form factors. The scope we committed to was two SKUs — handheld and rack-mount — on one shared PCB, covering QR, barcode, UHF RFID and NFC. Exotic sensor formats went to a separate feasibility study. A counterfeit detection module went into its own scope.
That split kept the core build focused and testable, and kept the ambitious ideas alive without letting them stall delivery. Every MVP needs that line drawn explicitly and in writing.
If all six are yes, you have an MVP.
Only if the enclosure houses working, stable electronics and a user can operate it unaided. A printed shell alone is a looks-like prototype.
Typically three to six months for a connected device, depending on battery, radio and environmental requirements.
Usually yes. Dev boards work for proofs of concept but rarely survive real users. A purpose-designed board is far more reliable.
A prototype answers a specific engineering or design question. An MVP is an integrated product real users can test in their own environment.
Send us the one-line version of your product. We'll come back with a technical read: the hard parts, what to cut for v1, and a realistic path from prototype to field.
Consumer electronics architects. End-to-end.