Hardware Patent Strategy for Early-Stage Founders: What to File, When, and Why It Matters Before Your Raise

Why Hardware Patent Timing Is Categorically Different from Software

Software founders get a structural advantage that hardware founders don't talk about enough: the source code never ships. A competitor can run your app, study your UX, and benchmark your API, but the implementation stays hidden. Hardware eliminates that protection the moment your product leaves the factory. Anyone who buys a unit, runs a teardown, and photographs the board has your complete functional architecture, legally. Under both the Uniform Trade Secrets Act and the federal Defend Trade Secrets Act (18 U.S.C. § 1839(6)(B)), reverse engineering a lawfully acquired product is explicitly excluded from the definition of improper means of trade secret acquisition. Your trade secret evaporates the moment a sophisticated competitor puts your device on a workbench.

Copyright doesn't fill the gap. Software companies routinely use copyright to protect source and object code as literary works, giving them overlapping layers of protection that operate independently of patent status. Hardware gets nothing equivalent. Under 17 U.S.C. § 102(b), copyright protection expressly does not extend to "any idea, procedure, process, system, method of operation, concept, principle, or discovery." The functional operation of your circuit design, your mechanical assembly, your sensor integration, none of it is copyrightable. The doctrine traces to Baker v. Selden, 101 U.S. 99 (1879), and its principle is codified in 17 U.S.C. § 102(b). If your hardware innovation is functional, copyright is not a fallback. Patent is your only structural protection.

The filing race is real and the clock starts earlier than most founders expect. The America Invents Act, effective March 16, 2013, converted the U.S. to a first-inventor-to-file system under 35 U.S.C. § 102. Under pre-AIA law, you could "swear behind" a competing filing by proving you invented first. That option no longer exists. A competitor who observes your demo at a conference and files before you walks away with the patent, regardless of who built it first. Hardware prototyping cycles, typically 18 to 36 months from concept to market, extend the exposure window significantly.

The manufacturing process itself creates disclosure risk that software founders never face. Before your product ships, you are sharing detailed technical specifications with ODMs, contract manufacturers, and component suppliers. NDA coverage in those relationships is frequently incomplete, and a single unprotected disclosure in a foreign jurisdiction starts a clock that runs with no grace period. Hardware founders who plan to file "when we're ready to launch" systematically lose international rights before they know they've lost them. The strategic imperative is to file before any technical disclosure, not after your product is ready to sell. The rest of this guide explains how to do that without burning your seed budget on premature filings.

Provisional Patents: The 12-Month Bridge Strategy

A provisional patent application, filed under 35 U.S.C. § 111(b), does one thing that matters above all else: it locks in your priority date today, without requiring formal claims, an abstract, or examiner review. The USPTO will not examine it. It will never become an issued patent on its own. What it does is buy you 12 months of protected runway, during which you can disclose to investors, iterate on the design, talk to manufacturers, and make every business decision you've been holding back on.

The cost argument is straightforward. Filing fees under the current USPTO fee schedule per 37 CFR § 1.16(d) run $130 for small entities (companies with fewer than 500 employees qualifying under 37 CFR § 1.27), $65 for micro entities, and $325 for large entities. Add attorney preparation time and a well-drafted provisional typically lands between $2,000 and $5,000 all-in. A non-provisional utility patent, from filing through issuance, routinely costs $10,000 to $20,000 in attorney time alone, before USPTO search and examination fees of $2,000 to $3,500. For a hardware company that will file in multiple technology classes and across multiple jurisdictions, the provisional is not a shortcut. It is the only rational sequencing.

The moment your provisional is filed, you can mark your product and materials "Patent Pending." Under 35 U.S.C. § 292, marking a product "patent pending" when no application has been filed is a civil violation. Once the provisional is on file, that marking is legally accurate and appropriate. In investor decks, on the product itself, and in conversations with distribution partners, "Patent Pending" signals that a competitor cannot simply copy and race to market without consequence. The deterrence value is real, even before a single claim has been examined.

Write the Provisional Broadly, Not Just for What You've Built Today

The most common and most damaging provisional mistake is writing to the current prototype. Under 35 U.S.C. § 119(e)(1), your non-provisional can only claim priority back to the provisional for subject matter that was actually disclosed in the provisional, a requirement reinforced by MPEP § 211.05. If your product evolves over the next 12 months and you add a feature, change a mechanism, or refine a process that wasn't described in the provisional, that element gets no priority date. It is as if you filed for it the day you filed the non-provisional, which means 12 months of competitor activity could have generated prior art against it.

Founders who plan to expand internationally have one more decision to make before the 12 months expire: whether to file a PCT application. Filing under the Patent Cooperation Treaty within 12 months of your provisional extends the deadline to enter national-phase prosecution in most countries from 12 months to 30 months from your original priority date, per PCT Art. 22 and 35 U.S.C. § 363. That additional 18 months lets you validate product-market fit, close your seed or Series A, and decide which jurisdictions actually matter for your business before committing to translation costs and national filing fees in each country. In 2024, WIPO recorded 273,900 PCT filings, a figure that reflects how central this mechanism has become to hardware IP strategy across markets. For any hardware startup with international manufacturing or distribution ambitions, a provisional-to-PCT bridge is the standard structure.

Freedom-to-Operate Searches: What Founders Get Wrong

Most founders who think about patents are thinking about the wrong question. A patentability search asks whether your invention is novel and non-obvious enough to receive a patent. A freedom-to-operate (FTO) search asks whether your product can be made, used, or sold without infringing someone else's existing patent claims. These are distinct inquiries with distinct methodologies, and conflating them is one of the most expensive mistakes early-stage hardware founders make.

The uncomfortable reality: you can have a highly patentable invention that simultaneously infringes an existing patent. You can also have an unpatentable product that you can sell without restriction. Patentability and freedom to operate are independent variables. Knowing one tells you nothing about the other.

What an FTO Search Actually Analyzes

An FTO analysis is claims-centric. Under 35 U.S.C. § 271(a), infringement occurs when you make, use, sell, offer for sale, or import a patented invention, and patent scope is defined by the claims, not the preferred embodiment. That means it is not enough to identify ways your product differs from what the patent depicts in its drawings. If an independent claim is written broadly enough to capture your product as literally built, you have an infringement exposure regardless of how different your underlying innovation is. This is the core lesson from Markman v. Westview Instruments, 517 U.S. 370 (1996), which established that claim construction, not product-to-product comparison, governs infringement analysis.

Hardware amplifies this risk. Wireless communication alone involves over 100,000 active U.S. patents. Semiconductor processing, MEMS fabrication, power management, and medical device hardware each carry patent densities that can generate genuine infringement exposure even for genuinely novel products. A single broad independent claim, filed by a company operating in a completely different market segment, can read directly on what you are shipping.

When to Run the Search

The timing of your FTO search determines what options remain available to you. A design-around, where you modify your product to avoid infringing claims, is almost always feasible if you find the problem early. At the prototype stage, a design-around costs engineer time. At Series A due diligence, the same finding costs deal velocity, investor confidence, and potentially valuation. Post-launch, it costs product recall, manufacturing retooling, and re-certification, assuming litigation does not arrive first.

The practical framework for hardware founders: run a focused FTO search before you commit to a manufacturing architecture, not after you have locked your BOM and placed tooling orders. If your product touches wireless communications, MEMS, or medical device hardware, treat the FTO search as a line item alongside mechanical prototyping, not as a legal expense to defer.

Investors conducting Series A diligence will ask whether you have done an FTO analysis. Disclosing a known risk that you have already designed around is a manageable conversation. Having a diligence attorney discover an unanalyzed infringement exposure in a densely patented hardware category is not.

The Public Disclosure Clock

The moment you disclose your invention publicly, a clock starts. In the U.S., that clock runs for one year under 35 U.S.C. § 102(b)(1) before your own disclosure becomes prior art against you. Founders treat this as a strategy. It is not. It is a safety net with holes, and relying on it systematically destroys international patent rights before you ever file.

What Counts as a Public Disclosure

The definition of "public disclosure" under 35 U.S.C. § 102(a)(1) is broader than most founders expect. A trade show demo, an investor pitch deck shown without a signed NDA, a crowdfunding campaign video, a LinkedIn post with technical detail, a conference presentation, a published academic paper: all of these qualify. Courts have consistently held that disclosure to even a single person who is not bound by confidentiality can constitute a public disclosure triggering the prior art clock. The word "public" does not require mass distribution. It requires that the information left your control without a confidentiality obligation attached (Egbert v. Lippmann, 104 U.S. 333 (1881)).

The Grace Period Is Narrower Than You Think

The U.S. one-year window under § 102(b)(1) covers disclosures made by the inventor or by someone who obtained the information directly from the inventor. It does not cover third-party disclosures. If a manufacturing partner shares your technical specifications with a subcomponent supplier, and that supplier publishes or uses the information independently, the grace period does not apply to that disclosure. You are now running against prior art that you cannot argue away.

The on-sale bar compounds the risk. Under Helsinn Healthcare S.A. v. Teva Pharmaceuticals USA, Inc., 586 U.S. 123 (2019), a single confidential commercial transaction, including a manufacturing agreement or volume pricing discussion, can trigger the § 102(b)(1)(B) on-sale bar. A secret sale or offer for sale starts the one-year clock even without any public disclosure. Founders who enter manufacturing negotiations without a patent on file are often running a clock they do not know is running.

The International Picture

Outside the U.S., the margin shrinks to nearly zero. The European Patent Convention Art. 55 provides a six-month grace period, but only for disclosures resulting from evident abuse by a third party or display at an officially designated exhibition. Investor pitches do not qualify. Trade shows do not qualify unless the exhibition appears on a specific approved list. China's Patent Law Art. 24 and Japan's Patent Act Art. 30 are similarly narrow: academic society presentations and government-supervised testing disclosures may qualify, but crowdfunding campaigns, ODM spec-sharing, and investor meetings do not. If you disclosed before filing and you want patent protection in Europe, China, Japan, or South Korea, the analysis is usually over before it starts.

The Provisional as Disclosure Insurance

A provisional application filed before any disclosure event establishes a priority date that predates the disclosure, removing it from the prior art calculation entirely under § 102(b)(1) and § 119(e)(1). This is the actual function of the provisional beyond cost savings: it is a disclosure-proofing mechanism. The IP strategy decisions that matter most for hardware founders are not about which claims to pursue. They are about sequencing: file the provisional before the pitch deck goes out, before the ODM conversation happens, and before the Kickstarter campaign goes live. The rest of the strategy builds on that foundation.

What VCs Actually Want to See in Your IP Stack Before the Raise

Most hardware founders arrive at Series A diligence with some version of the same answer to IP questions: "We have patents pending." That answer communicates filing status. It does not communicate strategy, coverage, defensibility, or whether the company actually owns what it thinks it owns. Institutional investors making hardware bets are not checking a box when they ask about IP. They are evaluating whether the patent position creates a moat that justifies the valuation, and that evaluation requires a specific set of documents, not a summary statement.

The diligence process at a competent Series A fund will include outside counsel reviewing your IP position in detail. That review follows a checklist, whether or not the investor articulates it to you upfront. Founders who have assembled the materials in advance move through diligence in weeks. Founders who have not spend 30 to 60 days locating documents, chasing former employees for signatures, and explaining gaps they should have closed six months earlier. In the worst cases, those gaps produce escrow holdbacks, valuation reductions, or conditional closes that reopen negotiations you thought were settled.

The Components of a DD-Ready IP Binder

Building this binder is not a paralegal task you assign in the final weeks before a term sheet. Each component has a lead time, and several of them require coordination with people who may no longer be easy to reach. The realistic timeline to assemble a complete binder from scratch is six months — which means the work should begin as soon as you know a raise is on the 12-to-18-month horizon.

Inventor Assignments: The Gap That Kills Deals

Under 37 CFR § 3.73, recorded inventor assignment documents are required for the assignee — meaning your company — to have standing to enforce patent rights. An assignment that was signed but never recorded at the USPTO is not sufficient for diligence purposes, and an assignment that was never signed at all leaves the company unable to enforce the patent without joining the inventor as a necessary party. If that inventor has since left the company on bad terms, "necessary party" becomes a serious litigation vulnerability. Series A investors understand this, and their counsel will check USPTO assignment records directly, not just ask whether assignments were obtained.

The fix is straightforward but time-sensitive. Every person named as an inventor on a pending or issued application needs a signed assignment agreement on file and a recorded assignment at the USPTO. For current employees, this is typically covered by a properly drafted PIIA agreement executed at hire. For contractors, co-founders who departed early, or collaborators from the prototype phase, you may need to track down individuals who have no current obligation to cooperate. That process is far more tractable six months before your raise than it is in week three of due diligence.

University IP Clearance: The Bayh-Dole Exposure

If any inventor worked at a university, a research hospital, or a federally funded laboratory at any point during the development of the technology, the Bayh-Dole Act (35 U.S.C. §§ 200–212) creates a potential institutional ownership claim. Bayh-Dole gives universities the right to retain title to inventions developed using federal funding, and university IP policies routinely extend that claim to inventions developed using university facilities, time, or resources — even without direct federal funding. The university does not need to have filed anything. The potential claim exists by operation of law and institutional policy.

Series A investors will ask whether any inventor had an institutional affiliation during development. If the answer is yes and you cannot produce a clearance letter from the institution confirming it has no ownership interest, that is a diligence blocker. Not a negotiating point — a blocker. Universities can take months to respond to clearance requests, and some require their own legal review before issuing them. Start this process early, or better yet, obtain clearance letters at the time the work is being done.

FTO Currency and the Stale Report Problem

A freedom-to-operate analysis that is more than 18 to 24 months old presents a specific risk that founders often overlook: it may be more damaging to your diligence position than having no FTO analysis at all. An investor whose counsel identifies a relevant patent that postdates your FTO report will interpret the old report as evidence that you stopped monitoring the space, not as evidence of a rigorous compliance program. The patent landscape shifts continuously, particularly in hardware categories with high filing volume. Your FTO analysis needs to reflect the landscape as it exists near your close date, not as it existed when you first got organized.

What "We Have Patents Pending" Actually Communicates

Filing status is table stakes. The investors you want to close are evaluating IP as a strategic asset: what is covered, how broadly, what the FTO exposure looks like, whether the claims survive a design-around attempt by a well-funded competitor, and whether the company actually owns everything it thinks it does. A coherent IP narrative — here is what we filed, here is what it covers, here is what we have designed around, here is what the FTO shows — is worth more in a Series A conversation than a large application count without supporting documentation. The binder is the evidence that the narrative is true.

Hardware IP Checklist

Hardware IP strategy is fundamentally a sequencing problem. The decisions that determine your patent position — filing dates, disclosure timing, FTO coverage, assignment completeness — are almost entirely made in the first 18 months of your company's life, and the mistakes made in that window cannot be fully corrected later under the pressure of a fundraise or an infringement claim. This checklist organizes those decisions by the order in which they must be made.

Phase 1: Before Any Technical Disclosure

1. File a provisional patent application under 35 U.S.C. § 111(b) before any NDA-free conversation with investors, ODMs, contract manufacturers, or component suppliers. The provisional establishes your priority date and functions as the foundation for every protection strategy that follows.
2. Draft the provisional to cover what you have built, what you might build, and the underlying principle your design is based on — include alternative embodiments, materials substitutions, and use cases that have not yet been validated. Narrow provisionals lose priority on the features your product actually ships with.
3. Once the provisional is on file, mark prototype materials and investor decks "Patent Pending" — confirm that at least one application is filed before using this marking, as marking without a pending application is a civil violation under 35 U.S.C. § 292.
4. Execute NDAs with all manufacturing partners, component suppliers, and technical advisors before any technical disclosure, and verify that NDA coverage is complete before beginning any disclosure conversation. A single unprotected disclosure in a foreign jurisdiction eliminates international patent rights in absolute-novelty jurisdictions with no remedy.
5. Run a focused patentability search to identify related prior art in your technology area — this informs how to draft claims in the non-provisional and surfaces design considerations before manufacturing architecture is locked.

Phase 2: Within the 12-Month Provisional Window

6. Decide whether to file a PCT application by month 12 from the provisional priority date — a PCT filing under 35 U.S.C. § 363 extends national-phase entry deadlines to 30 months from the original priority date, buying time to validate product-market fit before committing to country-by-country filing fees across Europe, China, Japan, South Korea, and other critical manufacturing and market jurisdictions.
7. Commission a freedom-to-operate search before committing to a manufacturing architecture or placing tooling orders — a design-around at the prototype stage costs engineer time, while the same finding at Series A diligence costs deal velocity, investor confidence, and potentially valuation.
8. Have every person who contributed to the invention sign an inventor assignment agreement, and record the assignment at the USPTO per 37 CFR § 3.73 — unrecorded assignments are insufficient for diligence purposes, and unsigned assignments leave the company unable to enforce the patent without joining the inventor as a necessary party.
9. Ensure all founders, early employees, and contractors who contributed to the technology have signed PIIA agreements assigning IP rights to the company — PIIA coverage must reach back to the earliest development work, not just current employees at the time of the raise.
10. File the non-provisional utility patent application with formal claims before the provisional expires at 12 months — the provisional-to-non-provisional conversion is a hard deadline with no extension, and subject matter not disclosed in the provisional receives no priority benefit under 35 U.S.C. § 119(e)(1).

Phase 3: Before the Series A Raise

11. Assemble the DD-ready IP binder: filed application copies with official USPTO filing receipts, recorded inventor assignment documents, PIIA agreements covering all contributors, an FTO analysis report, a no-lien certificate confirming no security interest against IP assets, and university or prior-employer clearance letters for any inventor who held an institutional affiliation during development.
12. Update the FTO analysis if the existing report is more than 18 to 24 months old — a stale FTO report can be more damaging to your diligence position than having none, because it signals that you stopped monitoring the patent landscape rather than demonstrating a rigorous compliance program.
13. Confirm PCT national-phase entry deadlines running 30 months from your original priority date and decide which jurisdictions to pursue before filing fees are incurred — national-phase entry decisions cannot be deferred past the deadline without permanently abandoning rights in those countries.
14. Develop a written IP narrative documenting what your claims cover, what competitors have been designed around, and what the FTO analysis shows — investors conducting Series A diligence evaluate IP as a strategic asset, and a coherent documented narrative is worth more than a large application count without supporting evidence.

The hardware IP window is real, and it runs earlier than most founders expect. The cost of building the foundation correctly — provisional filing, FTO analysis, inventor assignments, PIIA coverage — is a fraction of what it costs to repair those gaps under diligence pressure, when the parties who need to cooperate may no longer be reachable and the clock is running against a term sheet expiration.