How Fast Is the UPC? Time-to-Decision Benchmarks Across Divisions
The UPC promises a merits decision in ~12–14 months. Our data puts the real median near 13.7 — but preliminary injunctions land in weeks and divisions diverge. The benchmarks.
The Unified Patent Court was built to do one thing national systems struggle with: deliver a final, cross-border decision on patent infringement and validity in about a year. That is the signature promise — an oral hearing roughly twelve months after the claim is filed, and a written judgment within about six weeks of the hearing. The headline UPC time to decision figure most litigants quote is "12 to 14 months," and on the evidence so far it is broadly holding: across the merits cases in our database that reached a substantive infringement or validity ruling, the median time from filing to final decision is 417 days — almost exactly 13.7 months.
That single number is why the UPC has reshaped European patent strategy in three years. But like every headline average, it hides the variance that drives decisions. Provisional measures resolve in weeks, not months. Revocation actions run longer than infringement actions. And the busiest divisions — the German local divisions carrying most of the load — show the first signs of strain. This piece sets out the benchmarks by stage, case type, and division, using the decisions our platform has classified.
The headline: ~12–14 months to a merits decision
Start with what the court set out to achieve. The UPC's Rules of Procedure are deliberately front-loaded: the heavy lifting happens on paper, early, under strict deadlines, so the oral phase can be short. A defendant has three months to file a statement of defence; the written procedure (defence, reply, rejoinder) is fixed and compressed; a judge-rapporteur then runs an interim conference to tee up the issues; and the oral hearing itself is typically one day — a "mini-trial" where both sides argue an already-defined case. The court then "endeavours to issue the decision in writing within six weeks of the oral hearing." The stated aim, per the UPC's own guidance, is for cases to be concluded "within a total period lasting no longer than a year."
How is that aim playing out? Across the 73 infringement and revocation actions in our database that reached a substantive outcome — infringed, not infringed, revoked, maintained, or amended — the median filing-to-decision time is 417 days, about 13.7 months, with the middle half of cases falling between roughly 11 and 17.5 months. The Munich Central Division's decision in the consolidated Sanofi v Amgen revocations in about 13 months is a representative merits timeline.
The UPC is not hitting a clean twelve months — but for a court issuing decisions enforceable across 18 states, a real-world median under fourteen months is the fast-litigation promise substantially kept. The pressure point is the busy German local divisions, whose heavy caseloads pull the longest cases toward the upper end of that range.
The variance: provisional measures in weeks, merits in months, appeals on top
The merits average is only one speed the UPC runs at. The court is really three different clocks, and conflating them is the most common mistake in forum planning.
| Stage / case type | Indicative time (filing → decision) | Our data (median) |
|---|---|---|
| Provisional measures (preliminary injunction) | days to a few months | ~3 months |
| ↳ fastest quartile of preliminary injunctions | within ~2 weeks | 13 days (Q1) |
| Infringement on the merits | ~12–14 months | ~13 months |
| Revocation on the merits | longer — adds validity workup | ~11–14 months |
| Merits, substantive ruling (inf. + rev.) | the headline benchmark | ~13.7 months |
| Appeal to the Court of Appeal | adds roughly 8–14+ months | varies by track |
Category medians shift with how "decision" is counted — first order versus final judgment — so treat the merits-substantive row as the load-bearing figure and the others as directional.
The spread between the first two rows is enormous. A preliminary injunction is its own front-loaded fast-track: the applicant puts the whole case in the opening application, the court sets a hearing "as soon as possible," and the oral hearing — where one is held at all — runs a half-day to a day. In our data the median provisional-measures case is decided in about three months, and the quickest quarter of preliminary injunction applications within roughly two weeks of filing. On that timetable a patentee can block operations across Europe in under four months. For a defendant facing a credible preliminary injunction, the relevant clock is not next year's trial — it is the next few weeks.
At the other end, appeals add time on top. The Court of Appeal in Luxembourg moves quickly on procedural questions (suspensive-effect requests, security orders), but a full merits appeal extends the dispute well beyond the first-instance window. The two-years-in framing is instructive on outcomes too: of the merits decisions reviewed on appeal, about a third were overturned — "fast to first-instance decision" is not the same as "fast to final certainty."
The divisional breakdown — where the merits clock runs faster and slower
Here is a cut rarely published anywhere: filing-to-decision for substantive merits rulings, by division. These are per-division cuts, so read every row as directional — a tendency rather than a settled rate. But the pattern is a clear one, and it points squarely at caseload pressure.
| Division | Merits median (filing → final decision) |
|---|---|
| Munich CD | ~10.5 months |
| Paris CD | ~13.7 months |
| Munich LD | ~14.5 months |
| The Hague LD | ~15.2 months |
| Milan LD | ~17.0 months (directional) |
| Düsseldorf LD | ~17.9 months (directional) |
| Mannheim LD | ~18.4 months (directional) |
Two things stand out. First, the central divisions — Munich CD and Paris CD, which carry the standalone revocation work — turn out merits decisions at or inside the headline target. Second, the German local divisions handling the most infringement volume (Düsseldorf, Mannheim, and to a lesser extent Munich) sit at the slower end of the merits range. That is exactly the strain Bird & Bird flags: Munich LD is the single busiest division in our corpus at 626 cases, and the German LDs together absorb the bulk of UPC litigation — more cases, longer queues. The Court of Appeal is the second-largest venue at 614 records, reflecting how much of the court's activity is now appellate and procedural review rather than fresh first-instance filings.
The honest caveat: these per-division merits samples are too thin to publish as hard service-level numbers — we frame them as direction, not guarantee, and recompute as the pools grow.
Why speed is itself the strategy
These benchmarks matter because speed at the UPC is not a convenience — it is leverage. German litigation has historically been bifurcated: an infringement court can grant an injunction long before a separate validity court rules, creating the famous "injunction gap." The UPC abolishes bifurcation in the ordinary case — infringement and validity decided together, in one ~12–14-month proceeding, enforceable across 18 countries. UK proceedings are thorough but slower and single-jurisdiction. The UPC's offer is the combination national systems can't match: broad territorial reach, validity and infringement in one judgment, and a calendar measured in months.
That changes settlement dynamics. The court's first merits decision — Kaldewei v Bette, Düsseldorf, 3 July 2024 — issued a cross-border injunction roughly thirteen months after the UPC opened, signalling the timeline was real. Once defendants grasped that a pan-European injunction could land inside fourteen months — or, via provisional measures, inside a single quarter — the incentive to settle early rose sharply. The court's "summer of settlements" — high-profile disputes such as Ocado/AutoStore, Abbott/Dexcom and Huawei/Netgear resolved before judgment — flowed directly from that combination of tight timelines, territorial reach, and judicial readiness to issue injunctions. Many disputes never reach a merits ruling: among datable main cases in our data that settled or withdrew, the median exit comes around six months in — well before trial, under the pressure the clock creates.
The practitioner takeaways are concrete. Price the settlement against the clock, not the trial. For the patentee, the credible threat is the speed of relief, especially a preliminary injunction; for the defendant, the protective letter and an early validity attack are time-sensitive defences. And choose the division with the calendar in mind — a central-division revocation and a German-LD infringement action do not run on the same timetable.
What to watch
The open question is durability: will the speed hold as volume rises? The signs are mixed. The court is still broadly hitting its target, but the rolling averages have crept above twelve months, roughly 29% of infringement actions are now over a year old without a decision (per practitioner analysis), and the busiest German divisions are the ones bending first. New filings keep arriving — our corpus already spans well over a thousand case records for 2025 — and judicial capacity is finite. If the merits median drifts from ~13.7 toward the high-teens, the UPC's core value proposition narrows. Watch three things through the rest of 2026: whether the German LDs' merits timelines stretch further, whether the six-week post-hearing writing target keeps slipping (only ~40% of decisions currently meet it), and whether the Court of Appeal's own backlog turns "fast first instance" into "slow final certainty." We track filing-to-decision, filing-to-hearing, and the intervening milestones by division on our pendency dashboard, and will update these benchmarks as the next cohort of decisions lands.
The one thing unlikely to change soon: at the UPC, how fast you can reach a decision is as much a strategic variable as what you can win. Plan the calendar first.
Related reading
- How the Unified Patent Court Works: A Litigator's Guide to Divisions, Timelines & Outcomes
- UPC vs. National Courts: Where to Litigate Your European Patent in 2026
- Where Should You File? UPC Preliminary Injunction Grant Rates by Division
Want filing-to-decision benchmarks, pendency by stage, and preliminary injunction timing for a specific division before you file? That is what UPClytics is built for — explore the data.
Sources
- Unified Patent Court, "Frequently Asked Questions: Oral Procedure" (oral hearing ~1 year after filing; decision within ~6 weeks; one-year overall aim) — https://www.unifiedpatentcourt.org/en/faq/oral-procedure
- Bird & Bird, "The UPC in Numbers: 32 Months of Action" (rolling first-instance average just over 15 months; German LD caseload) — https://www.twobirds.com/en/insights/2026/the-upc-in-numbers-32-months-of-action
- Bird & Bird, "The Unified Patent Court (UPC) at 21 months" — https://www.twobirds.com/en/insights/2025/the-unified-patent-court-(upc)-at-21-months
- Pinsent Masons (Out-Law), "The UPC: lessons learned from the first 18 months" (12–14 month target; Sanofi v Amgen ~13 months) — https://www.pinsentmasons.com/en-gb/out-law/analysis/upc-lessons-learned-first-18-months
- Bird & Bird, "Applications for provisional measures before the UPC – what may be granted" (front-loaded preliminary injunction procedure; one-day oral hearing) — https://www.twobirds.com/en/insights/2025/applications-for-provisional-measures-before-the-upc--what-may-be-granted
- UPClytics.com — internal dataset of classified UPC decisions (snapshot June 2026 build).