Which patents should you keep, and which should you let lapse?

This guest article comes from PatentRenewal.com, a company specializing in automated IP renewal solutions. We explored how companies can tell which patents still earn their place in the portfolio and which no longer do.

In brief: Keep a patent when it still protects current revenue, blocks a competitor you can name, supports a live licensing conversation, or matters to financing or acquisition value. Narrow the geography when the invention still counts but some territories no longer do. Let it lapse when none of those holds. The goal is to maintain the portfolio the business needs now, not the one you filed years ago.

Most founders approach patent portfolio management with the wrong default. Once a patent is granted, keeping it feels like the safe, neutral choice, and letting it lapse feels like throwing away an asset. But renewal is actually the easier decision to get right, because you are making it with far better information than you had at filing. By now you know which products reached the market, where your customers turned up, which competitors matter, and which parts of the original roadmap dropped off.

The data backs this up. PatentRenewal.com's Patent Lifecycle Report found that patents are abandoned on average 10.7 years into their life, meaning many patents are still being paid for well into their second decade. For a small, founder-managed portfolio, four questions are usually enough to separate the rights worth keeping from the ones worth reconsidering.

The choice is keep, narrow, or lapse

There are three useful outcomes, not two: keep the patent, narrow its geographic coverage, or let it lapse.

That middle option matters because your original filing decisions were made under uncertainty. Early on, you may not have known where your customers, manufacturing, competitors, or partners would concentrate. Several years later, you usually do. A patent can still be strategically important without justifying protection in every country you originally chose.

Four patent portfolio management questions that show whether a patent still earns its place

Don't start with whether the invention is still technically clever or whether the patent was hard to get. Those facts explain why you filed. They say nothing about why you should keep paying to maintain it.

Ask these four questions for each patent.

1. Is it in something you sell today?

Start with current commercial relevance. Does the patent cover a product, component, manufacturing process, or feature that contributes to revenue right now? And more to the point, do the granted claims cover the implementation that actually matters commercially?

A patent can relate broadly to a current product without protecting the specific feature that gives it a competitive edge. If that link is unclear, mapping the claims with a patent attorney will tell you what the patent actually protects.

Future roadmap value can still count, but it shouldn't carry the same weight as technology already earning money in the market.

2. Does it block a competitor you can name?

Competitive value should be specific. You should be able to point to the competitor, the product or technical route involved, and why your patent matters to that route. If you can't name a plausible competitor, the patent may still have value, but blocking a rival isn't yet the reason to maintain it.

This is also why patent count tells you so little about how defensible you are. A handful of patents sitting on commercially important technical routes can matter far more than a large portfolio wrapped around technology no competitor needs.

3. Is there a live licensing conversation?

Separate real licensing relevance from theoretical potential. An active negotiation, technology partnership, cross-license, or transfer discussion gives a patent a concrete commercial role. "Somebody might want a license one day" is a much weaker reason.

Optionality matters in deep tech. Just don't let it become an open-ended excuse for maintaining rights that have no credible route to revenue.

4. Would an investor or acquirer ask about it?

For deep-tech companies, this one can outweigh the other three. A patent doesn't have to cover a revenue-generating product today to be worth keeping. It might protect a foundational technical principle, a platform capability, or research that anchors ownership around your core technology.

In IP due diligence, an acquirer maps your patent rights against the technical advantage your company's value rests on. Would this patent belong on that map? Would its absence leave an obvious hole? If so, current revenue alone shouldn't decide it.

Most patents land somewhere in between, and mixed answers usually point to geography.

Mixed answers usually point to geography

When the answers are mixed, the useful question shifts from whether a patent still matters to where it still needs to be enforceable.

Review the individual jurisdictions against the company you have today. Do you have customers there? Do you manufacture there? Do important competitors make, sell, or operate there? Does the country matter to a live licensing or partnership strategy? Where none of that applies, continued protection deserves a closer look.

European patents are where this bites hardest, because a single granted patent doesn't stay single. Once it is validated, it splits into a separate renewal in every country you kept, each paid to a different office, under its own rules and in its own currency. One decision becomes ten, and one invoice becomes ten. In the tenth year, Germany's official fee is €430 and the UK's is £230, and while the smaller economies cost a good deal less, the amounts stack.

By year 10, a conventional European patent maintained across the 18 EU states currently covered by the Unitary Patent costs €5,133 in official renewal fees for that year alone. If only three of those markets still matter commercially, narrowing coverage can cut that annual bill substantially while keeping protection where the business case remains real. Narrowing coverage helps to keep protection where the business case is still real, instead of paying for the map you drew years ago.

Rising patent renewal fees eventually force the question

In the US, large-entity patent maintenance fees rise at each stage, from around $2,150 at 3.5 years to $4,040 at 7.5 years and $8,280 at 11.5 years (USPTO fee schedule, 2026). The pattern holds more widely: most national offices charge annual renewal fees that increase year on year, so a patent's most expensive years tend to be its last.

The fee schedule itself isn't the real point. The steadily rising cost is what forces owners to keep asking whether continued exclusivity is still worth paying for. And the aggregate behavior reflects exactly that kind of pruning: the share of US patents maintained through the full term peaked at roughly 52% around 2012–13 and has fallen to about 40% today, with most abandonments at one of the three fee stages, according to Dennis Crouch's March 2026 analysis. A lapse, in other words, is often a sign that a portfolio has been reviewed rather than neglected.

Deliberate and accidental lapse are worlds apart

An abandoned patent that lapsed by accident is the worst outcome: you paid for years and lost it without deciding to.

A deliberate lapse follows an honest assessment of a patent's value before the right expires. A missed renewal deadline takes that decision out of your hands.

Most jurisdictions offer a six-month grace period, the floor set by the Paris Convention (Art. 5bis) during which you can still renew after the ordinary deadline by paying a surcharge. Once that grace period closes, getting the patent back becomes far harder, and restoration may hinge on jurisdiction-specific legal tests, supporting evidence, and extra procedural steps. Treat the grace period as a safety net for genuine mistakes, not as extra time to make the decision.

Review what's due in the next 18 months

For a portfolio of a few dozen patents, you don't need a formal review committee. An annual, forward-looking review does more good. The hard part is usually just seeing the field clearly: knowing exactly which renewals fall due, in which countries, and by when. Across dozens of patents spread over multiple jurisdictions, each with its own deadline and local rules, assembling that view accurately is where most of the effort goes, and where a missed entry quietly becomes a missed patent.

Start with the patents that have renewals due in the next 18 months. Eighteen months is a practical horizon: annuity instructions typically need to be locked around three months before a deadline, so it leaves room for a full review cycle plus lead time. Run the four questions against each one. Where the answers are mixed, look at the individual countries. Where the call depends on claim scope, licensing rights, or legal consequences, bring in your patent attorney. Then record one sentence explaining each decision, so the reasoning still makes sense when you revisit it a year later.

PatentRenewal.com handles the IP renewals once you've made the call.

FAQ

Is it worth renewing a patent that isn't tied to a product yet?

Sometimes. A patent doesn't need to cover something you sell today to be worth keeping. It can still earn its place if it blocks a competitor you can name, supports a live licensing conversation, or protects a foundational position an investor or acquirer would expect to see. If none of those holds, that is the signal to reconsider.

How do I decide which patents to keep and which to let lapse?

Start with the renewals due in the next 18 months and ask four questions of each patent: is it in something you sell today, does it block a named competitor, is there a live licensing conversation, and would an investor or acquirer ask about it? Mostly noes point to lapse, mostly yeses to keep, and mixed answers usually mean the real question is which countries to keep it in.

Can I keep a patent in some countries but drop it in others?

Yes, and it is often the smartest option. Renewals are paid country by country, so you can maintain protection in the markets that matter and let it go where you have no customers, manufacturing, competitors, or partners. This narrows cost without abandoning the patent entirely.

What happens if I miss a renewal deadline?

Where a jurisdiction offers a grace period, you can usually still renew by paying a surcharge. Once that window closes, getting the patent back becomes far harder and may require a formal restoration procedure and proof that specific legal conditions were met. Treat the grace period as a safety net for mistakes, not as extra decision time.

Can I refile the same invention later if I let a patent lapse?

Generally no. Once the original application has been published, that disclosure becomes prior art (published information that counts against novelty), so filing the same invention again later will not restore the novelty you had at your original filing date. A deliberate lapse should be treated as final.

Does letting patents lapse look bad in due diligence?

Not necessarily. Diligence looks at whether your portfolio protects the technology and markets that matter, not at raw patent count. A documented decision to drop irrelevant rights can signal stronger discipline than renewing every historical filing on autopilot.