Yes, and the evidence is stronger than most founders expect. The EPO and EUIPO's 2023 study of European startups found that companies with patents or trade marks at seed stage were 2.6 times more likely to raise funding than those without, rising to 5.2 times at Series A and B, and 10.2 times for startups holding both patents and trade marks. Median seed funding was above €900,000 for startups using both rights against roughly €260,000 for those with none. In the US, Farre-Mensa, Hegde and Ljungqvist (Journal of Finance, 2020) used the near-random assignment of patent examiners to show that winning a first patent causes, not just correlates with, higher growth: 55% higher employment growth and 80% higher sales growth five years later, largely by unlocking funding from VCs, banks and public markets.

What investors actually care about is not the certificate but what it tells them: that the technology is hard enough to be defensible, that the company owns what it is selling, and that a later acquirer will find a clean asset. Deep tech investors, whose bets need long lead times and large capital, weigh patents most heavily; SaaS investors weigh them least, but still check ownership in due diligence. A pending application counts: what investors look for at seed is filed priority, not granted claims. The corollary is that an unfiled invention described in a pitch deck can be both a public disclosure and a due diligence flag, so file before the roadshow. See also What is IP due diligence?

An IP due diligence checklist asks four questions: what do you own, do you really own it, is it still alive, and does anyone else have a claim on it. Expect to produce: a schedule of all patents and applications with numbers, jurisdictions, status and deadlines; assignment documents from every inventor, founder, employee and contractor into the company, with recordal at the patent office; employment and consulting agreements with invention assignment and confidentiality clauses; any licences in or out, including university or previous-employer licences and government funding conditions; an inventory of open-source components and their licences; trade secret policies; records of public disclosures before filing; freedom-to-operate analyses or opinions if any exist; and any threatened or actual disputes, cease-and-desist letters or oppositions.

For a seed round, a lighter version suffices: proof of filed priority applications, signed assignments, and clean founder IP. Series A and later investors, and any acquirer, will run the full list and often a freedom-to-operate check on the core product. Two documents carry the most weight: the assignment chain, because a patent the company does not own is worth nothing to it, and the disclosure record, because a publication before filing can invalidate what is on the schedule. Investors tend to accept pending applications and unresolved office actions as normal; what they do not accept is not knowing. Keeping a portfolio register with deadlines and assignments in one place turns a two-week scramble into a one-hour export. See also What is IP due diligence?

No formula converts a patent into a valuation number, but the best evidence points the same way. The EPO and EUIPO's 2023 study reports median seed funding above €900,000 for European startups using patents and trade marks against about €260,000 for startups with no IP rights, and finds that startups with patents or trade marks are more than twice as likely to achieve an exit by IPO or acquisition. Hsu and Ziedonis (Strategic Management Journal, 2013), studying VC-backed US semiconductor startups, found that larger stocks of patent applications were associated with higher pre-money valuations, with the strongest effect in early rounds and for founders without a prior track record, when investors have least else to go on. Farre-Mensa, Hegde and Ljungqvist (Journal of Finance, 2020) showed a first patent grant more than doubles the probability of eventually going public.

The mechanism is what matters for a founder. A patent raises valuation in three ways: it signals technical quality when there is no revenue yet, it reduces the investor's perceived risk that a competitor can copy the product, and it creates an asset that survives the company, which acquirers pay for and lenders can take as collateral. It lowers valuation in one case: when due diligence finds the patents are not owned, not maintained or already invalidated by an earlier disclosure. The valuation effect therefore comes less from the number of patents than from whether the claims cover what the company actually sells and whether the chain of title is clean. Two well-drafted applications on the core mechanism do more than twenty peripheral ones.

Broken ownership kills more deals than weak patents. The most common findings, in rough order of frequency: an inventor, founder or contractor who never signed an assignment, so the company does not own what it says it owns; a founder whose previous employer or university has a claim on the core invention under an employment agreement or institutional IP policy; a public disclosure, publication, demo or launch before the first filing that invalidates the key application outside the US; an open-source component under a copyleft licence such as GPL or AGPL embedded in a product the company claims as proprietary; lapsed applications or patents where a deadline or maintenance fee was missed; joint ownership with a partner or university with no agreement on who may license or enforce; and a competitor patent nobody checked that the product plainly reads on.

Most of these are fixable, and investors and acquirers distinguish between fixable and fatal. A missing assignment can usually be signed, provided the person is still reachable and cooperative, which is why it should be done at hiring rather than at exit. A prior-employer claim can sometimes be released or licensed. A missed disclosure cannot be undone in Europe, but the US grace period or narrower claims may salvage part of it. What turns a fixable issue into a lost deal is discovering it in the data room instead of disclosing it upfront with a plan. The lesson that recurs in every post-mortem: assignment language matters. Agreements that say the employee "agrees to assign" future inventions create only a promise; agreements that say "hereby assigns" transfer title automatically, the distinction at the heart of the Stanford v. Roche litigation. See also Do I need an invention assignment agreement?