Companies that buy patents are not one market. They are several, with different motivations, and the difference determines who will pay for a given asset.
An operating company buys what protects its products. An assertion entity buys what reads on other people's products. A defensive aggregator buys to take patents off the board.
The same patent can be worth very different amounts to each, or worth nothing to most of them.
But they all run the same four checks first, and those checks are free.
The buyer types
| Buyer | Buys for | Pays most for |
|---|---|---|
| Operating company | Product protection, blocking | Direct relevance to their business |
| Non-practising entity | Licensing and assertion | Claims reading on shipping products |
| Defensive aggregator | Removing assertion risk | Patents threatening their members |
| Competitor | Removing a threat, freedom to operate | Assets that constrain them |
| Existing licensee | Ending royalties, certainty | The patent they already pay for |
| Investment funds | Return on licensing | Portfolios with demonstrated revenue |
| Universities and research bodies | Rarely buyers; usually sellers | — |
Existing licensees are the most overlooked buyer and frequently the best. They already understand the technology, already value it enough to pay, and diligence is shorter.
What each actually wants
Operating companies
| Want | Detail |
|---|---|
| Coverage of their own products | Protection |
| Coverage of competitor products | Blocking |
| Cross-licensing currency | Portfolio strength |
| Freedom to operate | Removing a constraint |
| Ignore | Anything outside their field |
They are narrow and they pay well within that. A patent squarely in their space is worth far more to them than to a general buyer; one outside it is worth nothing.
Non-practising entities
| Want | Detail |
|---|---|
| Claims reading on products in the market | The whole thesis |
| Documented evidence of use | Reduces their work |
| Meaningful remaining term | Assertion takes years |
| Robust validity | Assertion invites challenge |
| Clean ownership | Standing to sue |
Evidence of use is what they are buying. A patent with a documented chart against shipping products is a categorically different asset to an NPE than one without.
Their reputation varies widely. The category spans licensing businesses operating reasonably and entities whose model depends on settlement economics. See patent trolls.
Defensive aggregators
| Want | Detail |
|---|---|
| Patents that threaten their members | Removing risk |
| Assets likely to reach assertion entities | Pre-emption |
| Not maximum licensing value | Different objective |
They buy to neutralise, not to assert, which means they price against the risk removed rather than the revenue generated. That can be more or less than an NPE would pay depending on the asset.
What every buyer checks first
| Check | Source | Eliminates if |
|---|---|---|
| In force? | USPTO Patent Center | Lapsed |
| Recorded ownership | USPTO Assignment Search | Chain gap |
| Remaining term | Front page + PTA + family | Under ~3 years |
| Claims read on products? | Their analysis | No evidence |
| Encumbrances | Records and disclosure | Unreleased lien |
| Prosecution history | Patent Center | Scope surrendered |
| PTAB and litigation history | Public records | Narrowed or challenged |
| Ipiry Patent Survival Curve v1.0 | Rate |
|---|---|
| Survive the 3.5-year fee (2022 cohort) | 85.8% |
| Survive the 7.5-year fee (2018 cohort) | 64.6% |
| Reach full term (2014 cohort) | 41.4% |
| Abandoned before full term | 58.6% |
Computed from 27,273,654 USPTO maintenance fee records covering 8,262,336 US utility patents — see the patent survival curve.
Buyers know that base rate. They check status first because a meaningful share of what is offered to them is already dead.
Investment funds and litigation finance
| Model | Detail |
|---|---|
| Patent acquisition funds | Buy portfolios, monetize through licensing |
| Litigation funding | Fund assertion in exchange for a share of proceeds |
| Buys | Assets with demonstrated or demonstrable revenue |
| Diligence depth | High — they are underwriting an outcome |
| Timeline | Slower |
Litigation funders are not buyers, strictly. They fund a case rather than acquire the asset, which suits an owner who has a strong patent and no budget to enforce it.
Both groups underwrite outcomes, so their diligence is the most demanding in the market. A documented chart is close to a prerequisite.
Universities and research institutions
| Role | Detail |
|---|---|
| Usually sellers, not buyers | Technology transfer offices |
| Occasionally acquire | To strengthen a licensing position |
| What they offer sellers | Rarely a market |
Technology transfer offices are worth knowing as counterparties even though they seldom buy, because they license actively and understand the mechanics.
What buyers do not care about
| Seller believes matters | Buyer weighting |
|---|---|
| Prosecution cost | Almost none |
| Patent count | Low — coverage matters |
| How novel the idea felt | None — the claims decide |
| Awards, press, prototypes | Minimal |
| Length of the specification | None |
| Time invested | None |
Money spent on prosecution says nothing about coverage. An expensive patent with narrow claims covering an approach nobody uses is worth less than a cheap one covering what everybody does.
The claims decide. See patent claim.
What raises the price
| Factor | Effect |
|---|---|
| Documented claim chart against shipping products | Largest single factor |
| Several independent claims of different breadth | Robustness |
| Family depth — continuations, foreign members | More leverage |
| Coverage in manufacturing jurisdictions | Supply-chain reach |
| Long remaining term | Longer runway |
| Clean, recorded chain of title | Removes the main blocker |
| Pending continuation | Family stays open |
| Prosecution history without heavy narrowing | Less estoppel |
Build the chart before approaching anyone. It is the difference between "here is a patent" and "here is a patent that reads on these three products". See claim chart example.
Approaching a buyer directly
| Step | Detail |
|---|---|
| 1. Identify who practises the claims | Free, from product docs and their filings |
| 2. Verify your own asset first | They will |
| 3. Lead with the evidence, not the patent | A chart, not a number |
| 4. Consider the notice risk | An approach can invite a declaratory judgment suit |
| 5. Decide sale or licence before contact | Different conversations |
An approach that reads as an assertion changes the dynamic. Framing it as a sale opportunity rather than an infringement allegation keeps it commercial and reduces the risk of a pre-emptive filing.
What stops deals
| Blocker | Fixable? |
|---|---|
| Chain of title gap | Sometimes |
| Dissolved entity in the chain | Frequently not |
| Unreleased security interest | Usually, with the lender |
| Undisclosed exclusive licence | Restricts what you can sell |
| Patent lapsed mid-process | No |
| Term too short | No |
| Inventor never assigned | Sometimes |
Chain defects stop transactions rather than reducing prices. Buyers do not negotiate around them; they move on.
Run the search on your own patents early. See assignment search.
Timing shapes who is interested
| Remaining term | Buyer appetite |
|---|---|
| 12+ years | All types |
| 8–11 years | Strong |
| 5–7 years | NPEs and competitors mainly |
| 3–4 years | Thin |
| Under 3 years | Most buyers decline outright |
Assertion takes years, so an NPE needs term remaining after litigation timelines. That is why short-term patents lose their largest buyer group first.
Reaching buyers
| Route | Access |
|---|---|
| Brokers | Buyer-side relationships — the scarce input |
| Direct to companies practising your claims | Identifiable yourself |
| Existing licensees | You already know them |
| Listing platforms | Wide, low conversion |
| Auctions | Concentrated, public |
| Competitors | Direct |
If you can name the buyer, approach them. The broker's value is knowing who to approach when you cannot.
Companies practising your claims are findable for free through product documentation and their own patent filings. See patent infringement search.
Worked example: one patent, four valuations
A sensor calibration patent, 9 years remaining, reading on products from two manufacturers.
| Buyer type | Interest | Rationale |
|---|---|---|
| Manufacturer A (practises it) | High | Removes a threat, gains freedom to operate |
| NPE | High | Two identified targets, documented chart |
| Defensive aggregator | Moderate | Only if members are exposed |
| Unrelated operating company | None | Outside their field |
| Existing licensee | N/A | None exists |
What changed the outcome
| Before | After |
|---|---|
| Patent offered with no evidence | Chart built against both products |
| One interested party | Three |
| Manufacturer A's position | "Why would we pay?" |
The chart created the competition. Manufacturer A's interest depended on the asset being credible in someone else's hands.
The unrelated operating company was never a buyer, regardless of quality. Field relevance is binary for that group.
How buyers price
| Input | Effect |
|---|---|
| Income from licensing or avoided cost | The core |
| Remaining term | Discounts sharply below ~5 years |
| Maintenance fees inherited | Subtracted |
| Foreign annuities | Subtracted |
| Validity risk | Discount |
| Enforcement cost, where relevant | Subtracted |
| Evidence of use | Multiplies interest |
| Fee | Due after grant | Large | Small | Micro |
|---|---|---|---|---|
| First | 3.5 years | $2,150 | $860 | $430 |
| Second | 7.5 years | $4,040 | $1,616 | $808 |
| Third | 11.5 years | $8,280 | $3,312 | $1,656 |
| Total | $14,470 | $5,788 | $2,894 |
The fees are a real deduction. A buyer acquiring a patent with two windows remaining is taking on up to $12,320 in large-entity fees before anything else.
Cost approaches are not used. What you spent on prosecution does not appear in any buyer's model. See intellectual property valuation.
Terms matter as much as price
| Term | Why buyers care |
|---|---|
| Whole family included? | Continuations and foreign members |
| Warranty of ownership and authority | Standing |
| No undisclosed licences | What they are actually getting |
| Fees current at closing | Immediate risk |
| Seller licence back | Reduces value if broad |
| Cooperation on foreign recordation | Needed after closing |
| Files and prosecution records | Enforcement preparation |
A broad licence back reduces what a buyer will pay, because it limits exclusivity. Reserve only what you actually need.
Companies that buy patents: the checklist
- Identify which buyer type fits your asset. Most patents suit one or none.
- Start with existing licensees and companies practising your claims.
- Verify in-force status and the recorded chain first. Buyers do.
- Fix chain defects before approaching anyone. They stop deals outright.
- Build a documented claim chart. It is the largest single value factor.
- Stop weighting prosecution cost. Buyers do not.
- Present the whole family, including foreign counterparts.
- Disclose encumbrances up front. Discovery in diligence costs more.
- Keep any licence back narrow. Broad reservations reduce the price.
- Start twelve months before a fee deadline. Lapsing mid-process ends everything.