UPC Analytics
ENDE
Analytics1Cases1Decisions3Judges3Parties5Firms
Filing trendsOutcomesPendencyTechnologyForum strategyLeading casesLegal issuesSecurity for costs

Outcome base rates

What's normal — PI grant rate, infringement rate, revocation rate, settlement rate. Honest denominators using motion type.

CategoryrevocationLegal principlemotivation to combine prior art references required for obviousnessClear all
Patentee win rate
Share of merits decisions where the patentee prevailed — infringement cases finding infringement, revocation cases upholding the patent. Settled, withdrawn, and procedural-only outcomes excluded from the denominator.
100%patentees prevail on the merits

1 merits decision (small sample)

1 won · 0 lost · Insufficient prior-period data

Win rate by year
Patentee win rate by year of first decision.
  • 2024: 100% (1/1)
Win rate by division
Top divisions by merits-decision volume.
  • Paris CD
    100%
    (n=1)
PI grant rate
—
PI grant rate (conservative)
—
Infringement rate
—
Revocation rate
0%
0 revoked / partially · 1 maintained / amended
Settlement / withdrawal rate
Settled / withdrawn / dismissed as a share of all non-pending outcomes.
0% 0 / 1
Outcomes by category (detailed)
Stacked breakdown using sharper outcome enums — revocation cases split into revoked_full / revoked_partial / maintained_as_*, etc.
By technology sector
Top sectors by case count (filter scope applied).
By case category
How outcome rates differ across the six L2 buckets.
  • Revocation1
By division
PI grant rate · infringement rate · revocation rate per division (within scope).
  • Paris CD1 casesPI grant rate: —Infringement rate: —Revocation rate: 0%
Recent decisions
Most recent decisions in scope.
  • 2024-12-18ACT_589997/2023Patent maintainedThe Paris Central Division dismissed Tandem Diabetes Care's revocation action against Roche's EP 2 196 231 B1 (ambulatory drug infusion system), maintaining the patent as granted and ordering the claimants to bear the costs of proceedings, finding the grounds for invalidity (including lack of inventive step over prior art Diaz/Robertson/Glejboel) were not proven.