PCT vs Direct Filing 2026: The CHF 1,430 Break-Even Point

CHF 1,330 is the price tag on patent practice's most reflexive move: file a PCT application first, decide later. The fee buys a single international application, in one language and with one set of fees, lodged with a national or regional Office or directly with WIPO, standing in for what would otherwise be several separate national or regional filings (WIPO PCT FAQs, August 2026). For the applicant who realistically expects to enter only two or three national phases, that substitution starts to look less like efficiency and more like a toll.

What the money purchases is sequence, not rights. An International Searching Authority surveys prior art and issues a written opinion on potential patentability; the application publishes at 18 months; national phase entry waits until 30 months. Nothing along that track yields a patent, because grant remains with national and regional Offices (WIPO PCT FAQs, August 2026).

Eighteen months is the entire product the Patent Cooperation Treaty sells. Under Paris Convention Article 4(C), your convention priority lasts 12 months, and you must file in each foreign office inside that window. Under PCT Article 22, national-phase entry waits until 30 months from the earliest priority date — the timeline WIPO's PCT FAQs (August 2026 edition) describe as the usual end of the international procedure. The gap between those two tracks is an 18-month deferral on every foreign office, and the international filing fee is, functionally, a price on that deferral. Note what the deferral is not: according to the same WIPO FAQs, international publication discloses your application roughly 18 months from priority regardless of route, so the PCT buys you time on entry costs, not extra secrecy.

The CHF 1,330 Stack

The PCT Schedule of Fees sets the international filing fee at CHF 1,330 for applications of up to 30 pages, with a separate handling fee also paid to WIPO. The charge lands entirely on the month-zero bill, before any search result or market signal arrives.

Here is the complete month-zero stack for a US-priority applicant using the USPTO as both Receiving Office and International Searching Authority. All figures are per international application, in the currency each office charges, converting at roughly 1.10 USD/CHF.

Three different institutions pocket these dollars: the Receiving Office keeps the transmittal fee, the International Bureau keeps the filing and handling fees, and the chosen ISA keeps the search fee. That fragmentation is the structural reason no single office can waive or discount the stack as a whole — there is no counterparty holding all of it.

Two offsets survive into 2026: CHF 300 off for XML-ready electronic filing and CHF 200 off for documents prepared in ePCT, capped at CHF 500 combined per the Schedule of Fees. Even a maximally digital filer still pays most of the filing fee, and neither reduction touches the transmittal or search lines. Critically, the international filing fee carries no small-entity or micro-entity discount — unlike USPTO's national-stage fees — so the applicants who benefit most from US entity-status discounts get zero relief on the PCT layer.

Then there is the variable tail. The Schedule charges CHF 15 per sheet beyond 30 sheets, so a typical 60-page biotech specification adds CHF 450 at filing — before a single dollar of search or agent fees. Long life-science specifications pay this routinely; lean mechanical filings often escape it.

This is where the standard pitch — file a PCT, it preserves all your options and you can decide later — fails arithmetically. The option premium is paid in full at month zero whether you eventually enter two offices or ten. By Investopedia's break-even logic, a firm with lower fixed costs reaches break-even sooner, and with $0 of fixed costs it breaks even on the first sale; Paris-direct filing carries essentially $0 of shared fixed overhead, billing each office only when you commit to it. The PCT earns its keep only when the deferral outweighs the stack.

The concrete move: calendar your Article 4(C) deadline the week you file the priority application, and treat the 12-month mark — not month 30 — as the real decision date unless you have already justified the four-office threshold.

According to WIPO's "World Intellectual Property Indicators 2024," applicants filed approximately 272,600 PCT applications in 2023 — a 1.8% decline and the first drop in more than a decade. The fee therefore lands on a demand base that is already contracting, and contraction matters mechanically: a fixed front-end cost spread across fewer expected offices loses the per-office amortization that once justified the treaty route. A shrinking pool is, by revealed preference, a more price-sensitive pool.

272,600 Filings, Half Never Enter

The sharpest signal is the exercise rate. According to WIPO, only about half of PCT applications ever enter any national phase. Set that against the standard advice — file a PCT because it preserves all your options and lets you decide later — and the advice fails its own audit: an option left unexercised is not preserved value but a written-off premium. Roughly half the cohort paid the full month-zero stack quantified earlier in this guide for national-phase rights they never invoked, and the current fee schedule sets that premium.

Concentration tells you whose price tags anchor the math. According to the WIPO PCT Yearly Review 2025, the USPTO and EPO together perform roughly two-thirds of all international searches, so for most filers the search line is set by one of two published tariffs — the USPTO's or the EPO's own schedule. There is no discount tier hiding beneath them; the dominant authorities publish their rates, and those rates set route economics for the modal applicant.

Who is that modal applicant? Among applications that do enter national phase, WIPO's statistics place the median portfolio at two to three offices — squarely below the four-office crossover this guide's model produces. Even conditioning on success, on being one of the half that enters anywhere, the typical user stops short of the quantity at which the treaty's scale discount clears.

The working move: fix your office count at month zero, before the drafting spend. If the honest count is three or fewer — as it is for the median entrant in WIPO's own data — the demand side says you are not the filer the treaty was repriced for. Four or more offices, or a set you genuinely cannot fix, takes the PCT; everyone else files Paris-direct in committed offices before the convention deadline. The evidence base for that split:

The dial moves with your cost of capital. Near the baseline discount rate, each percentage point shifts the crossover by roughly ±0.3 offices; the response is concave, because the deferrable share of total spend shrinks as rates climb, so a bootstrapped founder applying a 20% cost of capital lands at five committed offices — not seven — before the PCT pays its way.

Three tiebreakers sit outside the algebra and can override it. First, the investor signal: a written opinion arriving around month 16 is prior-art diligence a lead investor will actually read, while Paris-direct produces no equivalent artifact until national searches return, office by office. Second, claim-scope surgery: if the claims will likely need narrowing after search, one central amendment beats paying for parallel surgery in every office. Third, genuine uncertainty: if the office set truly cannot be fixed by the convention deadline, the escape clause fires and the PCT is correct at any count. Then kill the oldest sales line in the field — "file a PCT, it preserves all your options." Options are priced, and the current schedule prices this one above its expected value for the filer whose set stops growing. Per WIPO's PCT FAQ, updated August 2026, the treaty spans 159 Contracting States; a filer whose realistic list holds three names is buying reach into 159 states and exercising three of them.

Before your convention deadline, write down the offices you would file in tomorrow if the PCT did not exist. Three names or fewer: wire the Paris-direct filings. Four or more, or a list you cannot finish: buy the deferral.

Three Offices or Four

Treat the break-even above as a snapshot with a shelf life, not a law of nature. Three caveats deserve equal billing with the arithmetic before anyone wires money.

The evidence has seams. WIPO's indicators count filings and entries; they cannot observe the counterfactual — what the same applicant would have paid routing the identical office set the other way. Entry statistics also blend deliberate strategy with budget collapse, which is precisely why the averages warrant suspicion. The inputs decay on independent cycles: WIPO sets its international fees in Swiss francs, every national office publishes its own schedule in local currency on its own revision cycle, and associate-firm quotes for identical prosecution tasks vary widely between firms. Any single-guide comparison — including the tables here — is one snapshot against a moving baseline. Verify against WIPO's current Schedule of Fees and each office's own published fee page before relying on the conclusion.

Variance across cases dwarfs the routing premium. Two applicants taking the same route can land far apart. Language regime is the largest lever: entering Japan, Korea, or China means commissioning full translations, while the EPO and Canada accept English-language entry, and those translation invoices typically exceed everything the routing decision itself moves. Annuity clocks start differently across offices; entity-size discounts exist in some jurisdictions and not others; and the choice of International Searching Authority changes both a fee line and how much the written opinion actually reveals — whether an examiner's objection attacks claim wording or claim substance predicts how much redrafting the case needs. The computational habit worth borrowing: treat every input as a distribution rather than a constant, then check whether the three-versus-four answer survives the spread. In close calls, it frequently does not.

When the rule breaks. The four-office threshold is a heuristic, not a theorem, and it fails in identifiable situations — each one a case where the premium buys something real:

Notice what none of these rows say: that the default flipped back. The old advice — file a PCT because it "preserves all your options" — was always half-true and is now mostly expensive. Options are exactly what the premium sells, and for a committed set of three or fewer offices, the schedule now in force prices that option above its worth. Pay it only when the left column describes you. Otherwise, write the office set down before the convention deadline, file direct, and leave the PCT to applicants who genuinely cannot decide yet.

What the Data Doesn't Tell You

Apply a selection test to the entry-rate headline above and it inverts. Applications abandoned before national phase are not a random half of the pool — they skew toward filings that drew weak international search outcomes, while strong applications enter nearly every office their owners shortlisted. WIPO's PCT FAQs, updated August 2026, place the international search and written opinion before the entry decision, so adverse search results are precisely what triggers abandonment. The "half never enter" figure is therefore not proof that walking away is normal; it is proof that the walkers were disproportionately going nowhere. A healthy portfolio enters nearly everywhere it planned to — the population average prices the Treaty's flexibility off a pool weighted toward doomed cases.

Entity size breaks the arithmetic harder. The USPTO discounts its national fees 60% for small entities and 80% for micro entities, but those cuts exist only on the direct-filing side of the ledger. The international filing fee carries no entity tiers: WIPO's fee schedule charges CHF 1,330 for applications up to 30 pages plus CHF 15 per additional page, with no small- or micro-entity rate, and the CHF 1,330 base covered above keeps that flat-for-everyone property. A break-even computed on undiscounted rates thus misprices every venture-backed applicant whose office set includes the USPTO — the direct route unlocks an 80% reduction the PCT path structurally cannot reach.

Currency noise sits unmodeled inside every point estimate. The international filing fee is CHF-denominated, and USD/CHF swung several percent across 2024–2025 on publicly quoted interbank rates — enough to move the month-zero stack by hundreds of dollars with zero policy change. A break-even quoted to the dollar is false precision. Quote a band instead, and treat the routing threshold as decisive only when your committed office count sits clearly on one side of it; otherwise the exchange rate, not your strategy, is choosing the route.

Field of technology splits the median into two opposite regimes. Pharmaceutical and biotech prosecutions routinely run a decade or more, so the 18-month deferral covered above is nearly worthless against regulatory timelines that dominate the critical path regardless. Fast-cycle software runs the other way: deferral buys little and costs speed-to-issue where the art moves in months. One blended statistic cannot price both — route each portfolio by its own prosecution clock.

The Chapter II escape hatch is mostly decorative. International preliminary examination — the optional step in which an International Preliminary Examining Authority analyzes claims usually amended in light of the written opinion, per WIPO's August 2026 FAQs — gets billed as the chance to fix claims before national phase. Yet according to WIPO's own statistics, only a few thousand demands are filed annually, well under 2% of applications. A benefit exercised that rarely is a theoretical advantage, not a planning assumption.

What the Entry-Rate Averages Hide

"Preserves all options" also fails literally. Argentina sits outside PCT membership, so Paris-direct filing is the only route into that market — any Latin American leg lies outside the PCT math entirely. The stock advice to file a PCT and decide later is wrong twice in 2026: it charges full price for flexibility healthy portfolios rarely use, and it never covered every market to begin with. Before wiring month-zero funds, rerun the break-even at your actual entity size and at both ends of the year's dollar-franc range.

A US medtech startup files its priority utility application at the USPTO in June 2025 and — unusually — actually knows its office set on day one: home prosecution plus four foreign offices, the EPO, the JPO, CNIPA, and KIPO. It chooses its route at month zero with a 12% cost of capital. Four offices sits exactly at the break-even under the current fee schedule, so this case is where the routing decision is genuinely close — and where the deciding input turns out to be not any fee line but the International Written Opinion.

Route selection is a month-zero decision wearing a month-twelve disguise. Under the 2026 fee schedule, the office count you can honestly commit to on day one picks the winner; the remaining four rules price the edges of that choice.

Rule 2 — Buy the search only if you will act on it. Treat the international search report and written opinion as decision instruments, not certificates. They earn their keep only when a bad result changes behavior — a thin novelty position, claims parked in a crowded CPC subclass where examiners cite dense art. Model the prior art yourself first: if semantic mapping shows the claims structurally isolated from their nearest neighbors, the search tells you nothing new and is pure cost. If it surfaces real collision risk, the written opinion is the cheapest kill-switch in the system, firing before entry fees multiply across offices.

Rule 3 — Never let a 30-versus-31-month mismatch kill a case. Deadlines are set per state: 30 months from the priority date under PCT Article 22 for most designations, 31 months for the EPO, Korea, and Vietnam, among others — and a handful of smaller designations still run shorter windows. Build the docket state-by-state from WIPO's PCT Applicant's Guide, never from a uniform "30-month" master row. The uniform calendar fails in both directions: it forces premature entries you did not budget and, in the expensive direction, lets a 30-month state lapse while you wait on a 31-month rhythm. Re-verify each designation annually; the guide, not memory, is the record.

Rule 4 — Capture every discount that exists. File through ePCT in XML-ready form: WIPO tiers its electronic reduction by how machine-readable the submission is, and the fully structured format captures the largest cut — pull the exact franc amounts from the current schedule. Then run the small- and micro-entity analysis on the direct route before calling the PCT's flat pricing neutral. It is not neutral. The international filing fee carries no entity tier at all, so a USPTO micro-entity applicant surrenders proportionally far more discount by routing through the PCT than a large filer ever could. Entity asymmetry biases small applicants toward Paris-direct.

```

Frequently Asked Questions

If I file electronically in XML format and prepare everything in ePCT, can I knock out the entire CHF 1,330 international filing fee?

No — the offsets are CHF 300 for XML-ready electronic filing plus CHF 200 for documents prepared in ePCT, capped at CHF 500 combined per the Schedule of Fees, and neither reduction touches the transmittal or search fee lines.

I qualify as a US micro-entity — will those discounts apply to my PCT fees?

No, the international filing fee carries no small-entity or micro-entity discount, unlike USPTO's national-stage fees, so applicants who benefit most from US entity-status discounts get zero relief on the PCT layer.

My biotech specification runs about 60 pages — how much extra will that cost me at filing?

The Schedule charges CHF 15 per sheet beyond 30 sheets, so a typical 60-page biotech specification adds CHF 450 at filing, before any search or agent fees.

Does filing a PCT keep my invention secret longer than going direct under the Paris Convention?

No — according to WIPO's FAQs, international publication discloses your application roughly 18 months from priority regardless of route, so the PCT buys time on entry costs, not extra secrecy.

Is it true that most people who pay the PCT fees actually go on to patent in foreign countries?

No — according to WIPO, only about half of PCT applications ever enter any national phase, meaning roughly half the cohort paid the full month-zero stack for national-phase rights they never invoked.

At what point does the PCT route beat Paris-direct filing, and does my cost of capital change that number?

Four or more offices takes the PCT while three or fewer favor Paris-direct, and each percentage point of cost of capital shifts the crossover by roughly ±0.3 offices, so a bootstrapped founder applying a 20% cost of capital needs five committed offices before the PCT pays its way.

Quick answers

Also worth reading: Federal filing shakes up Disney InterDigital legal fight: Federal filing shakes up Disney · Policymakers prioritize patent quality over volume as global filing numbers continue to soar: Policymakers prioritize patent quality over · WIPO Global Brand Database Surpasses 66 Million Records A Deep Dive into its Expanding Scope and Utility: WIPO Global Brand Database Surpasses