Analysis
How to Tell if a Patent Is Strong or Weak
A patent is a legal right, and like any legal right, some are formidable and many are paper tigers. Whether you are an investor doing diligence, a competitor sizing up a threat, or an inventor judging your own filing, you can read a patent's real strength from a handful of signals.
Educational framework, not legal advice. A real validity or infringement opinion requires a patent attorney.
- 01
Claim breadth
What to look at
Read claim 1 (the broadest independent claim). Count its limitations — the distinct elements or steps it requires.
How to read it
Fewer limitations = broader = covers more products = more valuable, but also more vulnerable to invalidity (more prior art can anticipate it). Many narrow limitations = easy to design around. The sweet spot is a claim broad enough to matter but supported well enough to survive a challenge.
- 02
Number of independent claims
What to look at
Count the independent claims (the ones that do not refer back to another claim).
How to read it
Each independent claim is a separate shot at coverage. A patent with one narrow independent claim has a single point of failure — knock it out and the patent is gutted. Multiple independent claims of varying breadth provide layered, more resilient protection.
- 03
The prior art it cited
What to look at
Look at the references cited on the front page and in the prosecution history.
How to read it
A patent that issued over a lot of close prior art has been battle-tested — the examiner saw the nearby art and allowed it anyway. A patent that cited almost nothing is a yellow flag: it may not have been thoroughly examined, meaning undiscovered prior art could invalidate it later in an IPR.
- 04
Prosecution history
What to look at
How long did it take to grant, and how many office actions were there? How much were the claims amended?
How to read it
Heavy amendment narrows scope and creates prosecution-history estoppel (surrendered territory the patent can never recapture). A patent allowed instantly with no rejections might be narrow, or might have had thin examination. The arguments made to win allowance also limit how the claims can be read.
- 05
Forward citations (cited by)
What to look at
How many later patents cite this one? This is the "cited by" count.
How to read it
High forward citations mean the invention was foundational — later inventors kept building on it. It is one of the strongest signals of importance and often of value. A patent cited by hundreds of later patents sits at the root of a technology tree. (PatentBrief shows this count on each patent page.)
- 06
Patent family size
What to look at
How many related patents exist — continuations, divisionals, and foreign counterparts in other countries?
How to read it
A large family signals the owner invested heavily and built layered, overlapping protection that is hard to design around. It also means pending continuations may still be adding claims tailored to competitors. A lone patent with no family is far easier to circumvent or invalidate.
- 07
Remaining term
What to look at
When does it expire? Take the earliest effective filing date plus 20 years, then add any Patent Term Adjustment shown on the cover page.
How to read it
A patent with 15 years of life left is worth dramatically more than one with 2. A patent near expiry has little forward value regardless of how strong its claims are — and once expired, the invention is free for anyone.
- 08
Specification quality
What to look at
Is the written description rich and full of alternative embodiments, or thin and sketchy?
How to read it
A detailed spec supports broad claims, survives Section 112 (written description and enablement) challenges, and leaves room for valuable continuations. A thin spec is fragile — it limits claim breadth, invites Section 112 attacks, and forecloses future claim flexibility.
- 09
Section 101 eligibility risk
What to look at
For software and biotech especially: are the claims directed to an abstract idea, a law of nature, or a natural phenomenon — or to a concrete technical solution?
How to read it
Post-Alice, subject-matter eligibility is the first thing a challenger attacks on software patents. Claims that recite a specific technical improvement are far stronger than claims that merely apply an abstract idea on a generic computer. An eligibility-vulnerable patent can be killed early and cheaply.
- 10
Who owns it
What to look at
Look at the assignee and any reassignment history. Is the owner active and litigious, a dormant entity, or a known licensor?
How to read it
Strength on paper is only a threat if someone will enforce it. A patent held by an active operating company or an aggressive licensing entity carries real practical risk. A patent held by a defunct or disengaged owner is a much smaller threat in practice, however broad its claims.
Putting it together
No single signal decides it — strength is a composite. A broad claim sitting on a thin specification with little prior art cited is a paper tiger: it looks scary but folds under an invalidity challenge.
Contrast that with a moderately broad claim that was battle-tested over close prior art, sits in a large patent family, has a decade of term remaining, rests on a rich specification, and is held by an active enforcer. That is a formidable patent — and the kind worth taking seriously.