Skip to content
PatentBrief

Patent Intelligence · Claim Evolution

Are patents getting broader or narrower?

The number of claims in a patent is a rough proxy for how broadly the inventor tried to protect their invention. More claims = wider protection. How has this changed since 2010? Which domains are writing broader IP today than they were a decade ago?

Global avg claims

23.8

2010–2014 avg

28.2 claims

2020–2024 avg

22.7 claims (-19.5%)

Overall direction

narrowing

Average Claims Per Patent by Grant Year

↓ Narrowing (-19.5%)

Showing avg claims (purple), P75 (light bar). More claims = broader filing strategy.

25

2010

34

2011

29

2012

31

2013

24

2014

22

2015

24

2016

24

2017

21

2018

24

2019

20

2020

21

2021

25

2022

24

2023

23

2024

Year-by-Year Claim Statistics

YearAvgMedianP75Broad (21+)Narrow (≤5)
202422.92223480
202323.92223350
202225.12224291
202121.32223303
202020.32223322
201923.62224375
201821.42323351
201723.52227252
201623.52327281
201521.62127171
201423.72329201
201331.22632240
201228.62530201
201133.62734120
201025.02436111

Claim Evolution by Domain

Three-period averages (early 2010–14, mid 2015–19, recent 2020–24) with trend direction.

healthcare

→ Stable0.0%

—

2010–14

—

2015–19

28.7

2020–24

Semiconductors

↑ Broadening+14.8%

22.2

2010–14

23.4

2015–19

25.5

2020–24

Gene Editing

→ Stable0.0%

—

2010–14

—

2015–19

24.3

2020–24

Consumer Electronics

↓ Narrowing-22.3%

30.8

2010–14

24.4

2015–19

24.0

2020–24

Software

↓ Narrowing-28.1%

33.0

2010–14

23.7

2015–19

23.7

2020–24

Automotive

→ Stable0.0%

—

2010–14

24.5

2015–19

23.6

2020–24

Telecommunications

↓ Narrowing-26.6%

31.7

2010–14

22.2

2015–19

23.3

2020–24

AI & Machine Learning

↓ Narrowing-5.1%

23.8

2010–14

24.5

2015–19

22.6

2020–24

finance

→ Stable0.0%

—

2010–14

27.3

2015–19

22.2

2020–24

robotics

→ Stable0.0%

—

2010–14

18.9

2015–19

21.7

2020–24

gaming

→ Stable0.0%

—

2010–14

16.7

2015–19

21.3

2020–24

aerospace

↓ Narrowing-29.8%

29.3

2010–14

19.0

2015–19

20.6

2020–24

Mechanical Engineering

↑ Broadening+28.3%

15.8

2010–14

22.9

2015–19

20.2

2020–24

E-Commerce

↓ Narrowing-35.4%

31.0

2010–14

21.7

2015–19

20.1

2020–24

Clean Energy

↑ Broadening+46.0%

13.7

2010–14

17.2

2015–19

20.0

2020–24

materials

→ Stable0.0%

—

2010–14

17.7

2015–19

20.0

2020–24

Biotechnology

↓ Narrowing-30.2%

28.4

2010–14

20.3

2015–19

19.8

2020–24

Pharmaceuticals

↓ Narrowing-32.5%

28.3

2010–14

17.8

2015–19

19.1

2020–24

medical_devices

→ Stable0.0%

—

2010–14

17.1

2015–19

16.4

2020–24

oncology

→ Stable0.0%

—

2010–14

17.8

2015–19

—

2020–24

What claim count measures

Each claim in a patent defines a distinct scope of protection. More claims = more chances to catch infringers (different claim types cover different embodiments), but also more prosecution cost and more opportunity for the USPTO to invalidate individual claims. High claim counts are an aggressive protection strategy.

Why broadening ≠ bad

Broader filings (more claims) often indicate that inventors and their attorneys are getting more sophisticated at capturing every dimension of an invention. They may also reflect the trend toward continuation patent families — filing multiple continuations with different claim sets to build a defensive moat.

Domain variation explained

Pharmaceutical patents tend to have more claims because each formulation variant, dosage form, and method of treatment can get its own claim. Software patents are constrained by Alice doctrine, which invalidates abstract software claims, leading to tighter average counts. Biotech sits in between.

Claim Breadth Analysis →Patent Vintage Analysis →Patent Complexity Index →