Where Should You File? UPC Preliminary Injunction Grant Rates by Division

Across the preliminary-injunction decisions our platform has classified, the Unified Patent Court grants the injunction roughly 57% of the time — 49 of 86 fully decided preliminary injunction applications. That single number is the one most litigants quote. It is also the least useful, because it hides where the real decision gets made: the division you file in. Restrict the same dataset to first-instance applications and the figure is 59%; group the four German local divisions together and it climbs to 66%. The headline rate is an average of regimes that, in early 2025, were openly diverging on the rule that decides most preliminary injunction applications — urgency.

This is the practical heart of the "UPC preliminary injunction grant rate" question, and it is why forum strategy is the first thing a patentee should settle, not the last.

The headline: a 57% UPC preliminary injunction grant rate

Our database tracks preliminary injunction outcomes as either granted or denied. From the decisions issued between June 2023 and the most recent in our corpus, the split is 49 granted to 37 denied — a 57% grant rate. Two things shape how that number reads. First, how partial grants are treated: we classify each application as granted or denied, where a tracker that carves partial grants into a third bucket will report a lower headline. Second, the trajectory — the UPC has become steadily more willing to grant preliminary injunctions as its case law on infringement and validity has matured, so a figure built only from the early decisions understates where the court now sits.

The figure that should change how you file is not the court-wide average at all. It is the wide divisional spread beneath it — the part covered next.

The divisional breakdown — the spread nobody quotes

Here is the cut competitors rarely publish: preliminary injunction grant rate by division, from the decisions we have classified. Several divisions sit below a handful of preliminary injunction rulings, so treat anything under roughly ten as directional — a signal of tendency, not a stable rate.

Two things stand out. First, the German local divisions — Düsseldorf, Munich, Hamburg — are where the volume and the higher grant rates concentrate; the four German LDs combined run about 66% in our data. Second, the Court of Appeal sits at 50%, lower than the busy first-instance divisions, because appellate review pulls outliers back toward the centre and several first-instance grants are trimmed or reversed on the way up.

Per-division rates are sensitive to how the pool is drawn — whether partial grants count, whether ex parte orders are included, and the exact date of the snapshot can each move a single division's number by 20–40 points, which is why divisional figures vary so widely between published cuts. That sensitivity does not blunt the takeaway; it is the point: the spread between divisions is real and large, and Düsseldorf and Munich are the two divisions worth understanding before you choose.

What's driving it — two preliminary injunction regimes, one court

For much of early 2025 the UPC was running what practitioners openly described as two preliminary-injunction regimes at once. The divisions were not disagreeing at the margins. They were applying a different rule.

A UPC preliminary injunction turns on three questions: is the patent valid "to a sufficient degree of certainty," is there "an (at least imminent) infringement," and does the balance of interests favour an injunction (McDermott). The Court of Appeal has made validity hard to use as a shield — in 10x Genomics v NanoString it held that a preliminary injunction should be refused only if, on the balance of probabilities, the patent is more likely than not invalid, which places the burden on the defendant. So validity and infringement are not, in practice, where most preliminary injunctions are won or lost.

Urgency is. And here the divisions split:

• Düsseldorf demands a fully substantiated application within roughly one month of the patentee learning of the infringement (10x Genomics v Curio), reflecting traditional German practice. Miss it by a day and the application can fail before the merits are reached. Düsseldorf is also the only division issuing ex parte preliminary injunctions — relief without first hearing the defendant.
• Munich applies a more generous standard — a two-month "safe harbour" has been accepted (Dyson v SharkNinja, Syngenta v Sumi) — and has stated in an order that it does not share the Düsseldorf division's view. Munich Judge Tobias Pichlmaier has said the division would "almost never" issue ex parte injunctions, and it has not.

The Court of Appeal has since supplied the connective tissue: in Ortovox v Mammut it held that the clock on "unreasonable delay" starts when the patentee "becomes aware of a violation or should reasonably have known about it." Even that leaves room to manoeuvre — The Hague accepted a 12-month delay in Amycel v Spyro where gathering evidence genuinely took that long. But the structural point stands: in early 2025 the same delay could be fatal in Düsseldorf and forgivable in Munich. That is what "two regimes" means, and it is why a preliminary injunction strategy that ignores division choice is leaving the biggest variable on the table.

What this means for practitioners

The divisional spread converts into a short list of decisions you should make early.

1. Identify your likely division before you act, not after. Where you can file is constrained by the defendant's residence and where infringement occurs, but those constraints often leave a genuine choice — and that choice is worth 20-plus points of grant probability. Map it before the urgency clock is running.
2. Diary the urgency clock to the strictest plausible standard. If Düsseldorf is in play, treat one month from knowledge as the deadline and have a fully substantiated application ready to file. Document precisely when you first became aware of the infringement and every step taken since — Ortovox makes that timeline the battleground.
3. If speed and ex parte relief matter most, Düsseldorf rewards readiness. If your evidence or your knowledge date is messier, Munich's longer runway is the safer forum. Match the division to the shape of your facts, not to its reputation.
4. For defendants: file a protective letter (Rule 207). A pre-emptive brief lodged with the Court — setting out non-infringement, invalidity, and jurisdiction arguments — is the standard counter to an expected preliminary injunction. It is cheap relative to an injunction, and in a court that grants preliminary injunctions more than half the time, it is the difference between being heard before the order and after it.

What to watch

The "two regimes" are a snapshot of a court still converging. Through 2025 and into 2026 the Court of Appeal will keep harmonising the urgency standard — Ortovox was a step, and each further CoA ruling narrows the gap between Düsseldorf and Munich. Watch, too, whether the rising overall grant rate holds as validity attacks sharpen and proportionality arguments (anchored in Meril v Edwards and the Enforcement Directive) gain traction; a preliminary injunction is the most powerful order the UPC can issue before trial, and the balance-of-interests test is where its limits will be drawn. As the per-division pools grow past the small samples behind today's numbers, the directional rates in our table will firm up — and we will update them as they do.

The one thing unlikely to change soon: at the UPC, where you file still moves the odds as much as what you file. Settle the division first.

Related reading

• How the Unified Patent Court Works: A Litigator's Guide to Divisions, Timelines & Outcomes
• Meet the Judges of the UPC: Panel Composition & Ruling Tendencies
• The UPC at Three Years: Outcome Base Rates Every Litigant Should Know

Want preliminary injunction grant rates, time-to-decision, and judge tendencies for a specific division before you file? That is what UPClytics is built for — explore the data.

Sources

• McDermott Will & Emery, "The UPC's emerging approach to injunctions: patterns, pitfalls, and global implications" — https://www.mcdermottlaw.com/insights/the-upcs-emerging-approach-to-injunctions/
• Pinsent Masons (Out-Law), "UPC cases provide guidance on preliminary injunction applications" — https://www.pinsentmasons.com/out-law/analysis/upc-case-provide-guidance-preliminary-injunction-application
• UPClytics.com — internal dataset of classified UPC decisions (snapshot June 2026 build).