UPC vs. National Courts: Where to Litigate Your European Patent in 2026
UPC vs national court in 2026: a side-by-side framework on reach, speed, cost, validity risk and injunctions — and when each forum is the smarter choice.
The choice is no longer whether to take the Unified Patent Court seriously — it is whether, for this patent and this dispute, the UPC or a national court is the sharper instrument. Nearly three years in, the answer is rarely "always the UPC." The UPC offers reach that no national court can match — a single action that can yield an injunction across roughly 18 states at once — but national systems still win on predictability, granular cost control, and, in the right case, raw speed. This is a decision framework, not a verdict: a dimension-by-dimension comparison of the UPC against the forums that still matter most — Germany's bifurcated courts (Mannheim, Munich, Düsseldorf), the Netherlands (The Hague), and, for contrast, the UK, which sits outside the UPC entirely.
The headline trade-off — reach for risk
Strip the strategy back to one sentence and it reads like this: the UPC concentrates both your upside and your downside into a single proceeding. A national court hands you one country's worth of remedy and one country's worth of exposure; the UPC hands you a continent of each.
| Dimension | Unified Patent Court | National courts (DE / NL / UK) |
|---|---|---|
| Geographic reach | One action → injunction across all UPC states where the patent has effect (up to ~18), plus an emerging long-arm reach over non-UPC parts (incl. the UK) where the defendant is UPC-domiciled | One country per action; cross-border relief possible but exceptional (NL, and now DE in biosimilars) |
| Speed (merits) | ~12–14 month target to the oral hearing; in practice ~12.5 months, decision ~1 month later (Bird & Bird, 32 Months of Action) | DE infringement: ~10–18 months, but validity runs separately (24+ months); UK: ~12 months to trial; NL: comparable on the merits |
| Speed (provisional) | Preliminary injunction in days–weeks; our data: ~57% of decided preliminary injunction motions granted | NL kort geding: preliminary injunction in ~6–12 weeks at low cost; DE: ex parte preliminary injunctions possible in days |
| Validity risk | Central revocation can kill the patent everywhere it has effect, in one ruling | Nullity is country-by-country; no single knockout (DE bifurcates validity to the Federal Patent Court) |
| Cost | Higher fixed + value-based fees (€11k infringement fee and up); recoverable-cost ceilings €38k–€2M | Lower and more predictable for a 1–2 country dispute (esp. DE's RVG-capped regime) |
| Predictability | Young but fast-maturing case law; CoA reversal ~33% of merits appeals | Decades of settled doctrine, esp. Germany |
The rest of this piece unpacks each row — and, crucially, when each makes sense.
Geographic reach — the UPC's structural advantage
This is the dimension on which the UPC is simply in a different class. File once at a UPC division and a single panel can rule on infringement and validity for every UPC state in which the patent is in force, and issue one injunction that runs across all of them. Replicating that nationally means parallel suits in a dozen jurisdictions — separate counsel, separate fees, separate (and potentially contradictory) judgments.
What has changed the calculus in 2026 is long-arm jurisdiction. Building on the CJEU's BSH Hausgeräte v Electrolux ruling (25 February 2025), UPC divisions have started reaching beyond the UPC's own borders. In Fujifilm v Kodak, the Düsseldorf division awarded damages and a permanent injunction covering the UK part of a European patent against a Germany-domiciled defendant — a "historic first," in Pinsent Masons' phrase, for UPC relief extending beyond member states. The Hague division went further still in HL Display v Black Sheep, granting relief spanning UPC states, signed-but-unratified states such as Ireland, Lugano Convention countries, and the UK. As Baker McKenzie frames it, the UPC may now hear infringement actions over European patents validated in non-UPC countries — and may even assess their validity, though only inter partes, without touching the national register.
National courts are not standing still. Pinsent Masons notes the CJEU's reasoning has "empowered some national courts, notably the Munich Regional Court, to award cross-border injunctions" in the biosimilar aflibercept disputes — and the Dutch courts have offered cross-border relief for years. But for sheer default reach, nothing matches a single UPC action.
Speed — fast everywhere, but the German split still matters
Speed is where the comparison is closest, and where the German system deserves careful handling. The UPC's front-loaded procedure targets an oral hearing 12 to 14 months from filing, with the decision typically within about a month of the hearing; Bird & Bird's tracking confirms the court is broadly hitting roughly 12.5 months to the hearing in practice. That is genuinely fast — but it is not categorically faster than every national alternative.
The UK Patents Court aims to bring patent cases to trial within 12 months, and its Shorter Trials Scheme is built for proportionate, expedited disputes (with a costs-capping pilot — £500k on liability, £250k on quantum — running through end-2026). The Netherlands moves comparably on the merits and can be dramatically faster on provisional relief, with a kort geding decision often within a week or two of the hearing and an injunction obtainable in roughly 6 to 12 weeks, at relatively low cost.
Germany is the structural outlier because it bifurcates: infringement is decided by the regional courts (Mannheim, Munich, Düsseldorf), validity by the Federal Patent Court, on separate clocks. Infringement judgments often land in 10–18 months, but nullity proceedings take 24+ months — producing the famous "injunction gap," where a patentee can win and enforce an injunction before validity is finally resolved. For a patentee, that gap is a feature; for a defendant facing a questionable patent, it is the system's signature risk. The UPC, by contrast, hears infringement and validity together by default, closing the gap — which is precisely why some patentees still prefer the German split and some defendants flee it.
Validity risk — the one-shot problem
Here the trade-off inverts, and it is the single most important point for any patentee weighing the two systems. At the UPC, a central revocation action can invalidate the patent across every state where it has effect, in one ruling. Lose once and the protection is gone continent-wide. National nullity is the opposite: a patent struck down in Germany survives in France, the Netherlands, and beyond; an invalidity attacker must win the same fight country by country, with no single knockout available.
Our own outcome data sharpens the warning. Across the infringement actions our platform has classified as decided on the merits, just over half end with the patent valid and infringed, about a quarter not infringed, and about a quarter invalid — validity is attacked successfully in a large share of cases. A one-in-four chance of losing the patent everywhere at once is exactly the exposure that drove more than 500,000 classic European patents to be opted out of the UPC. For a crown-jewel patent of uncertain strength, the country-by-country resilience of national litigation can be worth far more than the UPC's reach.
Injunction posture and predictability — the maturity gap
On injunctions, both systems lean patentee-friendly, but differently. UPC divisions have shown a clear willingness to grant preliminary injunctions — our data puts the grant rate across decided preliminary injunction motions at roughly 57%, with a wide divisional spread (Munich and Düsseldorf among the more grant-prone) — and to issue pan-European final injunctions. Germany's appeal is the injunction gap; the Netherlands' is the speed and (now) cross-border reach of its provisional relief.
Predictability is where national courts still hold the edge. German patent doctrine is the product of decades of case law; UK and Dutch jurisprudence are similarly mature and densely reasoned. The UPC's body of law is being written in real time — fast, but young, and visibly still settling: its Court of Appeal has overturned a meaningful share of the merits decisions it has reviewed (on the order of a third). For a dispute that turns on a finely balanced point of law, the comfort of a developed national doctrine — and a more predictable cost ceiling — can outweigh the UPC's breadth.
When each forum makes sense
There is no universal answer, but the factors that should drive the call are clear (and largely the ones Pinsent Masons flags: location of the infringing acts and manufacturing, market share, speed, and judicial experience).
Lean UPC when:
- The infringement spans multiple UPC states and you want one injunction, not twelve suits.
- Your patent is strong and you can absorb the central-revocation risk.
- You need pan-European deterrence quickly, and the long-arm reach over non-UPC parts (incl. the UK) is strategically useful.
Lean national when:
- The dispute is confined to one or two countries — national litigation is usually cheaper and more cost-predictable there.
- The patent is valuable but vulnerable, and country-by-country resilience beats a single point of failure.
- You want Germany's injunction gap, the Netherlands' fast, low-cost provisional relief, or a UK forum (the only route to a UK-wide judgment from a UK court).
- Predictability on a knife-edge legal issue matters more than reach.
The two systems are complements, not substitutes — and during the transitional period (running to at least 2030), classic European patents that have not been opted out can be litigated in either, so the choice is live on a case-by-case basis.
FAQ
Is the UPC always cheaper than litigating nationally? No. The UPC is typically more cost-efficient only when it replaces multi-country litigation against the same infringer. For a dispute limited to one or two countries, national courts — especially Germany's value-capped RVG regime — are usually less expensive and more cost-predictable. UPC court fees start at a €11,000 fixed fee for infringement, plus value-based fees above a €500k case value, with recoverable-cost ceilings running from roughly €38k to €2M.
Can a UPC court really decide UK infringement when the UK left the UPC? In the right circumstances, yes. Where the defendant is domiciled in a UPC state, divisions have asserted long-arm jurisdiction over the UK part of a European patent (Fujifilm v Kodak). Any ruling on the validity of a non-UPC patent binds only the parties (inter partes) and does not alter the national register. The precise limits are being tested on appeal in 2026.
If I lose a validity challenge at the UPC, is the patent gone everywhere? Across all UPC states where it has effect, yes — that is the central-revocation risk, and it is the mirror image of the UPC's reach. National nullity, by contrast, is country-by-country, so a loss in one state leaves the patent standing elsewhere.
What to watch
The framework above is stable; one variable could move it materially. The Court of Appeal's ruling on long-arm jurisdiction — the Dyson v Dreame appeal was heard on 22 January 2026, with a decision expected in the first quarter — will define how far the UPC can reach into non-UPC territory, and how much an authorised representative can pull an international defendant into the court. If the long arm holds, the UPC's reach advantage widens and the case for a single UPC action strengthens further. Two other threads bear watching: the friction between UK courts and the UPC over interim-licence declarations in SEP disputes (the "anti-interim-licence injunction" / "anti-anti-suit injunction" stand-off), and whether national courts — led by Munich in biosimilars — keep expanding their own cross-border relief, narrowing the UPC's edge. The destination is not "UPC wins"; it is a two-track European system in which the smart money matches the forum to the patent.
Want the division-level grant rates, time-to-decision benchmarks and outcome base rates behind these comparisons? Explore the data on our platform.
Related reading
- How the Unified Patent Court Works: A Litigator's Guide to Divisions, Timelines & Outcomes
- Opt-Out or Stay In? The Strategic Calculus Before the 2030 Sunset
- Long-Arm Jurisdiction at the UPC: Can a German Court Decide UK Infringement?
Sources
- Pinsent Masons (Out-Law) — "Expect patent litigation strategies to be shaped by the UPC in 2026": https://www.pinsentmasons.com/out-law/analysis/patent-litigation-strategies-upc-2026
- Baker McKenzie — "EMEA: A Shifting Landscape on European Patent Litigation (the UPC's long-arm jurisdiction)": https://www.bakermckenzie.com/en/insight/publications/2026/02/emea-upcs-long-arm-jurisdiction
- Clifford Chance — "The dawn and rise of the long-arm jurisdiction of the UPC": https://www.cliffordchance.com/insights/resources/blogs/ip-insights/2026/01/the-dawn-and-rise-of-the-long-arm-jurisdiction-of-the-upc.html
- Bird & Bird — "The UPC in Numbers: 32 Months of Action" (time-to-hearing benchmarks): https://www.twobirds.com/en/insights/2026/the-upc-in-numbers-32-months-of-action
- Bird & Bird — "Unified Patent Court Fees increase from 1 January 2026": https://www.twobirds.com/en/insights/2025/unified-patent-court-fees-increase-from-1-january-2026---are-you-ready
- Bird & Bird — "Spotlight on the Netherlands: Dutch cross-border injunctions" (kort geding speed, BSH): https://www.twobirds.com/en/insights/2025/netherlands/patent-litigation-in-practice-series-spotlight-on-the-netherlands-dutch-cross-border-injunctions
- CJEU — BSH Hausgeräte GmbH v Electrolux AB, Case C-339/22 (25 February 2025).
- Unified Patent Court — official site (structure, jurisdiction, transitional period): https://www.unifiedpatentcourt.org/en
- UPClytics.com — internal dataset of classified UPC decisions (snapshot June 2026 build).