

Guest article by Dominic Davies, CEO and Co-founder of Lightbringer, the AI-native patent service
Defensive patents are patents you hold primarily to deter infringement claims and strengthen your position in negotiations. If another company asserts a patent against you, a patent covering its products may give you leverage to negotiate a settlement or cross-license. Owning a patent does not itself give you freedom to operate, however: your products may still infringe someone else’s rights.
Most people file patents for one of two reasons. The first is offensive: you want to stop a competitor selling something that copies your invention. The second is defensive: you want to make sure nobody can stop you.
Researchers who ran the Berkeley Patent Survey describe the defensive motive plainly. Companies that patent defensively see the patents not as weapons but as shields; if a competitor sues, you can counter-sue, and this may support a settlement or cross-license rather than a courtroom. Ideally, a defensive patent never has to be used: its value comes from the fact that it could be.
The same survey found that the top reason startups patent at all is to prevent competitors from copying them, followed by reputation and securing investment. Defensive value rarely tops the list at filing. It tends to emerge later, once the market has settled and you can see who is building where.
That timing matters for renewal decisions. A patent filed as an offensive bet on a product that never shipped can turn into a useful defensive asset if a competitor later moves onto the same technical route. And a patent filed as a shield can become dead weight if the threat it was meant to deter never materialized.
Renewal is the moment to ask whether the shield still covers anything. Ask these three questions of each defensive patent, and write down the answers.
A defensive patent has greater leverage when its claims cover commercially important activity that a competitor cannot easily avoid. If you can name the competitor, name the product or technical approach they're using, and explain why your claims sit across that approach, the patent is doing its job. If the honest answer is "someone in our space might one day," document the likely future use and the evidence supporting it.
This is also where patent count misleads. Twenty patents wrapped around a technology no competitor uses deter nobody. If you don't know who is building on your route, that's a monitoring problem before it's a renewal problem. Competitor patent monitoring can reveal technical priorities and filing activity. Combine it with product evidence and claim analysis to assess whether your patents provide useful leverage.
Think about the two or three companies most likely to assert a patent against you in the next five years. For each one: if they sent a letter tomorrow, which of your patents would you send back?
If a patent appears on that list, assess whether its likely leverage justifies the remaining cost. Note that a counterassertion may concern a different product line from the one involved in the original dispute. If it doesn't appear on that list, ask why you're paying for it. Deterrence only works against parties who make products you can point your claims at. A defensive patent has no leverage against a non-practicing entity that sells nothing, so if your main litigation risk comes from that direction, the patent isn't protecting you from it.
Some defensive patents earn their keep through agreements rather than threats. If a patent is named in a cross-license, a joint development agreement, or a supply contract, it has a concrete commercial role, and letting it lapse may breach a term or remove your leverage at renewal of the agreement itself. Check the contract before you check the fee schedule.
Optionality is a weaker version of this argument. "It might matter in a licensing conversation" is a reason to keep a patent for a year or two after the product it covers was retired. It's not a reason to keep paying into a patent's second decade.
These are indicators of defensive value, not automatic renewal rules. Also consider claim validity and scope, evidence of infringement, design-around options, remaining patent term and the commercial stakes.
Failing the three tests doesn't automatically mean lapse. There are other ways to hold defensive ground, and part of a good patent portfolio strategy is knowing when to switch:
The middle rows are where most of the savings hide. A European patent validated in ten countries becomes ten renewal fees paid to ten offices; cutting that to the three markets where you actually compete keeps the shield where it matters and drops the rest.
PatentRenewal.com's guide to which patents to keep and which to let lapse walks through the geography decision in detail.
Renewal fees are designed to climb. In the US, large-entity maintenance fees run from $2,150 at 3.5 years to $4,040 at 7.5 years and $8,280 at 11.5 years under the current USPTO fee schedule. PatentRenewal.com's US renewal fee guide covers the payment windows and grace period in detail.
The owners are responding. The share of US patents maintained through all three fee stages peaked at roughly 52% around 2012–13 and has fallen to about 40%, according to Dennis Crouch's analysis at Patently-O. A defensive patent has no product manager arguing for it. It gets dropped because nobody could explain what it was for, not because anyone checked whether the competitor it deterred was still around. The three tests exist so the decision is made on purpose.
A defensive patent you let go after failing the three tests is patent pruning. A defensive patent you lose because a renewal deadline slipped is an unplanned loss of rights and potential negotiating leverage.
Most jurisdictions allow a six-month grace period with a surcharge; PatentRenewal.com's grace period checker shows where a late payment still lands. After that, restoration depends on local rules and usually needs legal work. Treat the grace period as insurance against mistakes, not as extra time to decide.
You don't need a committee. Once a year, pull every patent that has a renewal due in the next 18 months and tag the ones you're holding for defensive reasons. Run the three tests. Where the answers are mixed, look at geography before you look at lapse. Where the call turns on claim scope or a contract term, get your patent attorney's read. Then write one sentence per patent explaining the decision, so next year's review starts from a record rather than a guess.
If you're not sure which patents in your portfolio are defensive in the first place, that's the first job. A freedom to operate search against your current products, and a portfolio review that maps each patent to a product, a competitor, or a contract, will sort them quickly. At Lightbringer, that mapping is what Insights is built to do, with every recommendation reviewed by a patent attorney before it reaches you.
What is a defensive patent?
A defensive patent is one you hold to protect your own position rather than to stop others copying a product. Its value is potential leverage: if a competitor asserts a patent against you, a defensive patent gives you a counter-claim, which usually pushes the dispute toward settlement or cross-licensing. However, owning a defensive patent provides potential leverage, not guaranteed freedom to operate.
Are defensive patents worth the maintenance fees?
Sometimes. A defensive patent is worth maintaining when it covers a technology a named competitor uses, when it would be your counter-claim in a dispute you can realistically foresee, or when it's named in a license or partnership agreement. If none of those apply, a narrower territory, a pending continuation, or a defensive publication may protect you for less.
What's the difference between a defensive patent and a defensive publication?
A defensive patent gives you the right to exclude others and costs renewal fees for as long as you hold it. A defensive publication puts the invention into the public record so nobody else can patent it, costs almost nothing, and gives you no right to exclude anyone. Note that defensive publications do not guarantee freedom to operate. Choose the publication when you need to prevent others from patenting the technical disclosure but have no realistic need to stop a competitor.
Do defensive patents deter non-practicing entities?
No. Direct counterassertion generally offers little leverage against entities with no relevant infringing activities. A defensive patent works as a counter-claim against companies that make and sell products your claims can reach. An entity that sells nothing has nothing to counter-claim against, so defensive patents offer little protection from that direction.
Can a defensive patent become an offensive one?
Yes. If a competitor moves onto the technical route your patent covers, a patent filed as a shield can become the basis for an infringement claim. This is one reason to keep defensive patents under review rather than treating the label as permanent.
How often should I review my defensive patents?
Once a year, covering every patent with a renewal due in the following 18 months. That gives you a full review cycle plus the lead time most IP annuity payment providers need to lock instructions before a deadline.
Compare your fees against optimal market rates. Receive an audit covering all your renewal-related costs with a detailed itemized breakdown.