All insights
Strategy8 min read

Opt-Out or Stay In? The Strategic Calculus Before the 2030 Sunset

More than 500,000 European patents are opted out of the UPC. Inside the Article 83 mechanics, the lock-in risk, and the calculus before the 2030 sunset.

More than half a million European patents have been pulled out of the Unified Patent Court's reach. On the register data published one year into the new system, opt-outs had climbed past 530,000 — roughly 418,000 lodged during the three-month sunrise period before the Court opened, plus another 112,000 in the first operational year. As of early 2025 the figure sits comfortably north of 500,000. That is the single most consequential strategic decision European patent owners have taken since 1 June 2023, and for a large minority of portfolios it is reversible only once — if at all.

This piece walks through the UPC opt-out strategy that produced those numbers: the Article 83 mechanics, the trade-off every proprietor is implicitly pricing, how to segment a portfolio, and the lock-in risk that sharpens as the transitional window counts down toward its 2030 sunset.

The opt-out by the numbers — and what it signals

The scale is hard to overstate. The opt-out rate spiked to a peak of 36,336 patents and applications on 30 May 2023, the day before the Case Management System went offline ahead of the Court's opening — a stampede for the exit by proprietors who wanted to keep the new court at arm's length until they could see how it ruled.

Two readings of that surge are both true. The cautious one: owners feared central revocation — a single UPC action that can wipe out a patent across all participating states at once — and chose the familiar, slower, country-by-country national system. The tactical one: the opt-out is cheap, free to file, and reversible (subject to the conditions below), so opting out early preserved optionality at almost no cost. Either way, the headline is that the patents you can see being litigated at the UPC are, by definition, the ones their owners chose not to shield.

How the opt-out works — Article 83 mechanics

The opt-out lives in Article 83 of the UPC Agreement, the article that governs the transitional period. The essentials:

  • It applies only to "classic" European patents (and applications and SPCs based on them) — not to Unitary Patents, which are locked into the UPC by design and cannot be opted out.
  • It lasts for the transitional period, which runs seven years from 1 June 2023 and may be extended once by up to a further seven years — a maximum of 14 — on the basis of a broad consultation with users. On the current schedule the window is due to expire on 31 May 2030; that "~2030 sunset" is the date the whole opt-out architecture is built around.
  • It must be lodged before any UPC action concerning the patent. If a competitor files at the UPC first, the door is shut — the patentee is "locked" into the Court and must defend there.
  • It can be withdrawn once — "opting back in" under Article 83(4) — which restores the dual jurisdiction of the UPC and the national courts. But withdrawal is permanent and one-time: a patent that has been the subject of a withdrawal cannot be opted out again. And withdrawal is unavailable if a national action concerning the patent has been brought during the transitional regime — the mirror image of the lock-in above. (A UPC Court of Appeal ruling has since clarified that, for these purposes, "action" means one brought during the transitional regime, so pre-June-2023 national litigation does not, on its own, block a later withdrawal.)

That asymmetry — opt out freely, but opt back in only once and only if nobody has sued you nationally in the meantime — is the hinge on which the entire strategy turns.

The core trade-off: pan-European upside vs. central-revocation risk

Strip away the procedure and the decision reduces to a single bet, which can be set out plainly:

Stay in (no opt-out)Opt out
Enforcement reachOne action → injunction across all UPC statesCountry-by-country in national courts
Invalidity exposureOne central revocation can kill the patent EU-wideMust be revoked country-by-country; no single knockout
Cost & speedSingle forum, ~12–14 month merits targetParallel national suits, slower, costlier at scale
Reversibilityn/aWithdraw once only; lost if sued nationally first
Best fitPatents you want to assert broadly and fastPatents you mainly want to hold and defend

The upside of staying in is genuine and is exactly what the UPC was built to deliver: a single pan-European injunction from one proceeding, replacing the old patchwork of parallel national suits. The downside is the symmetrical risk — a single central revocation, or a revocation counterclaim inside an infringement case, with pan-European effect. As Bardehle Pagenberg frames it, there will "hardly be a clear-cut case" either way; the considerations pull in different directions for every individual patent.

Portfolio segmentation: crown jewels vs. the rest

Because the calculus is patent-by-patent, the practical answer is rarely "opt out everything" or "stay in everywhere." It is segmentation. The pattern that has emerged in firm guidance — Freshfields, among others — is to treat the crown jewels differently from the long tail:

  • Crown jewels / blocking patents you have no near-term intention of asserting: many owners opted these out, refusing to expose a portfolio-defining asset to a single EU-wide revocation while the Court was untested. The cost of caution is low; the cost of losing the patent everywhere at once is catastrophic.
  • Patents you intend to enforce against pan-European infringement: stay in, and use the UPC's single-shot injunction as the whole point. Opting these out throws away the reach you actually want.
  • The long tail — lower-value, narrowly-practised, or geographically concentrated patents: the decision is driven by where protection actually matters, the strength of the patent, and the technology field. A patent that only matters in two countries gains little from UPC reach and loses little by opting out.

The segmentation is not static. A patent opted out today can be opted back in (once) when an enforcement need arises — provided no national action has intervened — which is why many sophisticated owners parked assets on the opt-out register as a default and kept the opt-in as a live option.

What the litigated set tells us

This is where our own data offers a narrow but honest lens. Our platform has classified 2,830 UPC case records and identified 542 distinct patents-in-suit across them. By construction, every one of those patents was not opted out — or was opted back in — at the time it was litigated. We see only the patents owners chose to expose.

That selection effect is the point. The half-million-plus opt-outs are the dog that didn't bark: the vast reservoir of European patents whose owners decided the UPC's pan-European injunction wasn't worth the pan-European revocation risk — at least not yet. The 542 patents we can see being fought over are the opposite population: assets their owners were confident, or aggressive, enough to put in front of the new court. As the Court's validity record becomes more predictable, the live question for every opted-out portfolio is which side of that 542-vs-530,000 line each patent now belongs on.

Two clocks now sharpen that question, and they pull in opposite directions. The first is the lock-in clock: the opt-out only works if it is on the register before anyone sues you at the UPC. A patent that is "in" the system is exposed every day to a competitor filing first — at which point opting out is off the table and a central-revocation counterclaim is on it. The second is the sunset clock: the opt-out exists only during the transitional period, which on the current schedule closes on 31 May 2030 (later if extended to the full 14 years). Once it shuts, every classic European patent falls under UPC jurisdiction automatically and the lever is gone for good. Because withdrawal is one-time-only, the sequence of moves matters as much as the moves themselves — which is why timing is not a footnote to the opt-out decision, it is the decision.

What to watch

  • The extension question. Whether the Administrative Committee extends the transitional period beyond 31 May 2030 (to up to 14 years) is the single biggest variable. An extension buys every opted-out portfolio more optionality; a hard 2030 close forces decisions.
  • Opt-back-in volume. Watch for a rise in withdrawals as owners grow confident in UPC validity outcomes — a leading indicator that the strategic mood is shifting from "shelter" to "enforce."
  • Lock-in litigation. More disputes over what counts as an "action" that blocks withdrawal, and over the timing edge cases, will keep refining how much room owners actually have to opt back in.
  • The revocation track record. As central-revocation outcomes accumulate, the perceived risk that drove half a million opt-outs will either harden or soften — and the whole calculus moves with it.

The opt-out was the first big strategic call of the UPC era. With the sunset now inside a five-year horizon, it is about to become the last one too.

Want to see which patents and parties are actually being litigated at the UPC — the non-opted-out set, mapped by division, outcome, and adversary? Explore the platform.

Sources

Opt-OutProcedure

Related insights

See the data behind the analysis.

Every article here is built on UPClytics — cross-case UPC analytics on divisions, judges, firms, and outcomes.