Key takeaways
- Crowdfunding a hardware product is an engineering-readiness test before it is a marketing one. Kickstarter requires a working prototype, not a rendering, before a hardware campaign can launch.
- Kickstarter and Crowd Supply fund all-or-nothing. Indiegogo moved to fixed funding only after its 2025 platform change, so a missed goal returns every pledge on any of the three platforms.
- A wireless product needs FCC equipment authorization before it can legally ship to US backers, and that process runs on its own timeline, separate from the campaign countdown.
- About 9% of funded Kickstarter projects never deliver a reward at all, per the most-cited independent study, and the FTC has taken action against a creator who spent pledge money without building the product.
Crowdfunding a hardware product means passing an engineering test before a marketing one. Kickstarter, Indiegogo and Crowd Supply all expect a working prototype, a realistic production plan and a way to actually ship what was promised, and each platform now enforces at least part of that expectation before a campaign goes live. The path runs through four engineering gates: a prototype that proves the product works, a bill of materials that supports the funding goal, a certification plan for the regions it will ship to, and a production and fulfillment plan that holds up once thousands of backers are waiting. A persuasive campaign page cannot substitute for any of the four.
Why hardware crowdfunding is an engineering problem first
A software campaign can launch on a landing page and a screenshot. A hardware campaign cannot, because the product still has to be built, tested, certified and shipped after the money arrives, not before. Kickstarter’s rules for hardware and product design projects require creators to “show working prototypes of their projects” and state plainly that the platform does “not allow photorealistic renderings” as a substitute. Crowd Supply’s own guide to when a project is ready to submit goes further: it says the platform “almost always require[s] a working prototype, preferably an EVT prototype,” before it will review a project, and creators are asked to send a physical unit the platform can photograph and film.
The contrast that matters here is not one piece of marketing advice against another. It is marketing readiness against engineering readiness, and only one of the two is enforced by a rule a creator can be rejected or suspended for breaking.
| Readiness type | What it proves | Who checks it |
|---|---|---|
| Marketing readiness | A page, a video and reward tiers that persuade a stranger to pledge | Backers, informally, as the campaign runs |
| Engineering readiness | A working prototype, a costed BOM, a certification plan and a production path | The platform at submission, then every backer at the ship date |
Choosing a platform: what each one actually commits you to
Kickstarter and Crowd Supply fund all-or-nothing: miss the goal and nobody is charged, and the creator keeps nothing. Indiegogo used to offer a flexible-funding option that let a creator keep whatever was pledged even below goal. That changed with Indiegogo’s October 2025 platform upgrade, after Gamefound took over the platform in 2025: new campaigns can only choose fixed funding, and live flexible campaigns were converted automatically. Older advice describing Indiegogo’s “choice” between fixed and flexible funding is out of date.
- Kickstarter: all-or-nothing, a 1 to 60 day campaign window, and a hardware category that requires a demonstrated working prototype before submission.
- Indiegogo: fixed funding only since the 2025 upgrade, with a built-in pledge manager and a renamed late-pledge feature for post-campaign sales.
- Crowd Supply: all-or-nothing, and the most explicit prototype gate of the three: the platform says it almost always requires a working prototype, preferably at the engineering-validation stage, before it will review a submission.
The prototype a campaign actually requires
Passing Kickstarter’s rule means the prototype has to work, not just look right in a video. That validation path starts from the same place any hardware project does: a feasibility pass before committing to full development. In our experience at Inventornest, we always prefer to begin with feasibility, whether the idea behind a crowdfunding campaign is simple or complex, because it surfaces the risks that are expensive to discover after a campaign is already live. See what product feasibility analysis actually covers for how that stage works.
In our experience at Inventornest, the proof-of-concept-to-production sequence (POC, then engineering validation testing, design validation testing and production validation testing) is useful for organizing the work, but it should not be read as assurance that a project moves through it in a straight line without returning to earlier decisions. The question that matters at each stage is what that build has to prove before moving on, not whether the project follows a perfectly linear diagram.
Engineering validation testing checks the product against its functional requirements. Design validation testing checks that the production process can build units to the required quality. Production validation testing is a faster, condensed pass through design validation on the actual production line. None of the three stages is optional just because a campaign deadline is approaching.
Costing a bill of materials the funding goal can actually support
A funding goal that does not trace back to a priced bill of materials is a guess, and backers eventually find out. Target costing for hardware works backward from a realistic selling price to the BOM budget that still leaves a margin, which is a more defensible way to set a funding goal than estimating a price first and hoping the BOM fits under it later.
Two checks worth running before a goal is published:
- Has every major line item on the BOM come from an actual supplier quote, not a catalog list price or an estimate carried over from a previous project?
- Does the goal include certification, tooling and fulfillment costs, not only parts and assembly?
Certification: the gate that runs on its own clock
A wireless product cannot legally ship to US backers until it clears FCC equipment authorization. Under 47 CFR 2.803, “no person may market a radio frequency device unless the radio frequency device is authorized,” and the FCC defines “marketing” broadly enough to include selling, offering for sale, importing, shipping or distributing a device for those purposes. The rule’s conditional-sales exception appears to allow an arrangement like a crowdfunding pledge, advertised before authorization is complete, only with a prominent disclosure that delivery depends on finishing the process, and for a device that needs full Certification rather than a self-declared Supplier’s Declaration of Conformity, delivering a unit to a backer before the certification grant is issued is not permitted. A compliance professional or the FCC’s own equipment authorization guide can confirm which path applies to a specific product; this is general information, not legal advice.
In our experience at Inventornest, external laboratory costs for the FCC and RoHS testing we have coordinated have sometimes run to approximately USD 2,000, with the laboratory’s portion of the work completing in around 20 days. Actual cost and duration depend on the specific product, the test scope and the laboratory, and can run higher or longer. That figure covers the laboratory’s invoice, not a complete client price or a universal certification package, and it should not be applied to CE, UKCA, UL or medical-device certification, which follow different requirements. For a fuller breakdown of what drives certification cost on a wireless product, see FCC certification cost for a wireless product.
Outside radio equipment, general consumer products sold in the US need a General Certificate of Conformity from the manufacturer or importer. That responsibility attaches to whoever puts the product on the market, regardless of how the product was funded; a crowdfunding pledge does not change who is accountable for product safety.
Tooling and the jump from a hand-built unit to a production run
The unit photographed for a campaign page was very likely built on a 3D printer or by hand. The units shipped to a thousand backers will not be. Who owns the tooling and what a mold actually costs both cover the decision point every funded hardware campaign eventually reaches: whether, and when, the production volume justifies committing to a mold, and what that commitment costs beyond the quoted tooling price.
Fulfillment: where funded campaigns still fail
A funded campaign has not delivered anything yet. The most-cited independent estimate of Kickstarter delivery outcomes, a 2015 survey of 47,188 backers by Wharton’s Ethan Mollick, put the share of funded projects that never deliver a reward at roughly 9%, with a plausible range of 5% to 14% depending on category and project size. That figure describes non-delivery specifically; a separate, more common failure is simply shipping late rather than not shipping at all.
In our experience at Inventornest, component procurement is a major source of schedule pressure on a hardware project, and we build schedule slack and account for sourcing risk when selecting components for that reason. The pressure is not unique to crowdfunded projects, but a campaign adds a public deadline and a few thousand people watching it slip.
The failures that draw regulatory attention tend to look like more than a missed deadline. The Federal Trade Commission’s 2019 complaint against Douglas Monahan and iBackPack of Texas alleged he raised more than USD 800,000 across four campaigns on Kickstarter and Indiegogo, missed delivery dates, made false statements about the product’s status, and spent pledge money on personal expenses and additional marketing instead of production. The resulting settlement permanently banned him from engaging in any future crowdfunding activities. This is general information about a closed case, not legal advice.
A pledge manager tool such as BackerKit does not solve any of this. It collects backer addresses, confirms reward and add-on selections, and processes payment for upgrades after the campaign closes; it does not design, manufacture or ship anything. Treat it as an administrative layer on top of a production plan that still has to be executed, not a substitute for one.
What the campaigns that actually deliver have in common
Successful hardware Kickstarter campaigns covers the delivered campaigns case by case, so this guide will not repeat that breakdown. The short version: the teams that shipped had a tested prototype and real cost math before they asked for money, not after.
Frequently asked questions
Do I need a working prototype before I can launch a hardware crowdfunding campaign?
Yes, on all three major platforms in practice. Kickstarter’s hardware and product design rules require a demonstrated working prototype and prohibit photorealistic renderings as a substitute, and Crowd Supply says it almost always requires a working prototype, preferably at the engineering-validation stage, before it will review a submission.
Can I take pledges for a wireless product before it has FCC authorization?
Pledges and advertising are allowed before authorization is complete only with a prominent disclosure that delivery depends on finishing the FCC process. Delivering a unit to a backer before that process is complete is not allowed if the device requires full Certification. Consult a qualified compliance professional or the FCC directly for a specific product and radio type; this is general information, not legal advice.
What is the difference between Kickstarter, Indiegogo and Crowd Supply for a hardware launch?
Kickstarter and Crowd Supply are both all-or-nothing. Indiegogo moved to fixed funding only after its 2025 platform upgrade. Crowd Supply applies the most explicit prototype-stage gate of the three before it will even review a submission.
How many funded hardware campaigns actually deliver?
The most-cited independent estimate, a 2015 Wharton survey of Kickstarter backers, found roughly 9% of funded projects never deliver a reward, with a range of 5% to 14%. That figure covers non-delivery only; shipping late is a separate, more common outcome.
Does crowdfunding change who is legally responsible for product safety?
No. The manufacturer or importer remains the responsible party for product safety and certification regardless of how the product was funded. This is general information, not legal advice; consult a qualified attorney or the relevant regulator for a specific product.
When should I move from a 3D-printed prototype to tooled production parts?
That depends on production quantity, enclosure complexity and unit-cost targets, not a fixed number of units. Review how tooling ownership and mold cost actually work before committing a campaign’s production budget to a decision either way.
Where Inventornest fits
Inventornest works on the engineering side of hardware crowdfunding: taking a product from a working prototype through the validation stages, a costed bill of materials and a certification plan, the technical groundwork a campaign is judged on before a backer ever reads the pitch. Campaign strategy, advertising and backer communications are outside what we do; those are better handled by a marketing or crowdfunding agency. For a technical read on whether a prototype, BOM and certification plan are ready for a launch date, book a consultation. Our portfolio has examples of prototypes taken through this process, and our crowdfunding product development page covers the service in more detail.
