Provisional Patent at Month 12: Convert, PCT, Refile, or Abandon?

This guide shows how to choose among four paths at the 12-month provisional deadline using a month-9 checkpoint.

It explains when conversion, PCT filing, a stronger refile, or deliberate abandonment is supported by commercial validation and US market priority.

How the 12-Month Clock Actually Runs

A provisional application buys a 12-month priority date and "patent pending" status, but it does not mature into a patent by itself. Under USPTO practice described in CASRAI's worked example, the applicant must affirmatively file a non-provisional application claiming benefit of the provisional before the 12-month date; nothing converts automatically. The provisional is a placeholder for the date, not a substitute for the later filing. Check the filing receipt and the claimed benefit language before relying on the date.

The clock is not a courtesy deadline. The provisional lapses under 35 U.S.C. § 111(b) after 12 months, and that lapse kills the priority date. As the MPEP § 711 analysis cited by Patently Strategic frames it, abandonment at the 12-month mark is statutory rather than discretionary: inaction does not keep the application alive, and the earlier filing date is lost unless a qualifying non-provisional (or other proper filing) is made in time. The distinction matters because a missed deadline is not cured by intent; the statute runs from the filing date, not from the inventor's calendar.

CASRAI's worked example illustrates the practical shape of that clock: a university invention disclosure becomes a provisional draft, and the 12-month deadline to file a non-provisional is the operative constraint. Use it to confirm who owns the disclosure, what was filed, and when the 12 months expire. The example is useful as a checklist of dates and documents, not as a promise that filing the provisional alone preserves rights.

If the deadline passes without the required filing, the provisional is abandoned. Revival is not automatic either; the MPEP § 711 route cited by Patently Strategic points to a petition under 37 CFR 1.137. That is a separate, discretionary request, not a reset button, and it does not restore the lost priority date as a matter of routine. Also distinguish statutory abandonment from an express abandonment under MPEP § 711, which requires a written declaration identifying the application; the two are not the same procedural event.

Practically, treat month 12 as a hard statutory boundary, not a reminder. Calendar the date from the provisional's filing date, verify the non-provisional's benefit claim, and decide the path before the deadline; once § 111(b) has run, the priority date is gone. A practical check: set the internal deadline at month 9, leaving time to prepare claims, drawings, and the benefit statement.

The Evidence: Minority Convert

The converging evidence is that only a minority of provisionals are ever converted, and the three sources below reach that conclusion from different angles.

Commercify's May 2026 deep read states directly that only a minority of provisional applications are converted into non-provisional applications, and that a substantial fraction of the conversions that do occur would have been better filed differently—or not at all.

Arapacke Law's timeline guide frames the inverse failure mode: filing mistakes can cause provisionals to expire unused, so a lapsed deadline sits alongside deliberate non-conversion as a path. Read together, the sources describe a funnel in which many provisionals never become non-provisionals; the sources do not state a precise rate, so treat the funnel as qualitative.

CASRAI's worked example traces a university invention disclosure through a provisional draft to a non-provisional filing. It is useful as a single affirmative chain, not as evidence that conversion is automatic; the second filing remains a separate decision point.

The practical check is to treat conversion as the exception to be justified, not the default to be executed. If an invention has not earned real commercial validation and the United States is not the priority market, the evidence points away from automatic conversion even before costs are considered.

A useful audit is to ask, for each provisional, which of the three source patterns it resembles: a justified conversion, an unused expiration, or a conversion that should have been filed differently. That question keeps the minority-conversion statistic from becoming a slogan.

Four Month-12 Paths, Compared

At month 12, Commercify names four real options rather than one default: convert the provisional to a non-provisional, file a PCT application, refile a stronger provisional, or deliberately let the application lapse. The calendar does not choose among them; the two-test rule does, and the choice should be made before the deadline forces it.

Deliberate abandonment is a legitimate strategic tool, not a failure mode. USPTO MPEP § 711 states that an application "may be expressly abandoned by filing a written declaration" identifying the application, so the applicant controls the exit as well as the entry. That makes abandonment a decision, not an accident.

Under the two-test rule, CONVERT wins only when both binary tests pass: real commercial validation and the United States as the priority market. If validation is real but the US is not the priority market, the rule points to PCT; if the US is the priority market but validation is not yet real, it points to refile; if neither test passes, it points to deliberate abandon.

Score the invention at month 9, not month 12. By then the evidence of validation and market priority is usually clear enough to choose a path, while there is still time to prepare the filing that path requires. Waiting until month 12 turns a strategic choice into a rushed default.

The practical check is simple: if either test fails, do not let the provisional roll into a non-provisional by default. Choose PCT, refile, or abandon on purpose, and record the reason in the file so the next review starts from a decision rather than a habit.

Costs: The 5–10x Cliff at Month 12

The month-12 conversion is a large cost step relative to the provisional, but no source in the grounding states a specific multiplier, so treat any "5–10x" figure you encounter as unverified. What Commercify's founder guide does establish is directional: the provisional's low price is not a reliable proxy for total cost, because the conversion carries the expensive drafting work. Verify current USPTO fees before relying on any ratio.

Commercify's founder guide names the inversion: "the cheap provisional + expensive conversion pattern usually costs more total than the expensive provisional + cheap conversion pattern, and produces weaker rights." In practice, the cheaper first filing can push drafting, claim shaping, and correction work into the later, more expensive filing, while a stronger provisional can make conversion more of a filing step than a rebuild.

The budget test is therefore not whether the provisional is affordable today, but whether the substantially larger non-provisional cost is fundable at month 12. If that later cost cannot be funded, the provisional was the wrong instrument for the invention, not merely an inexpensive first step. Price the actual conversion with current USPTO fees and your attorney's estimate before committing.

Because fees change, verify current USPTO filing fees directly before relying on any budget. Treat any third-party fee summary as a starting point, and confirm the current schedule, surcharges, and entity-status discounts with the USPTO before committing to a path.

A useful internal check is to price both patterns side by side: one budget for a minimal provisional plus a full rebuild at conversion, and one for a stronger provisional plus a lighter conversion. The comparison should include not only filing fees but also the cost of rework, since the inversion claim is about total cost and right strength, not filing fees alone.

What the Evidence Does Not Establish

No source in the available sources set states a precise conversion rate; “a minority” is the strongest claim on record. If counsel or investors ask for a number, treat that phrase as qualitative and verify any percentage against USPTO or SBA statistics before quoting it.

The two-test rule also assumes a U.S.-domiciled applicant. For foreign filers, Paris Convention and PCT national-phase deadlines can layer on top of the 12-month clock, so the month-9 score is not the whole calendar. Confirm your own deadline chain with counsel; the available sources here does not cover non-U.S. priority chains.

The table below is the only new claim this section adds: it shows where the two-test conversion rule breaks and where it still wins. It is illustrative, not exhaustive, and it does not replace counsel’s review of your facts.

Copy the worksheet into your own tracker, assign one owner to each gate, and set calendar invites 30 days before each date so a missed gate is visible before it becomes expensive. Keep one evidence folder with the customer name, market ranking, budget approval, and claim draft; if a gate fails, do not release the conversion budget until the failure is resolved or the path changes.

Decision Rules to Serve the Two-Test Rule

The first rule is negative: no provisional converts automatically. The number that kills the myth is zero—there is no automatic conversion. Under 35 U.S.C. § 111(b), silence after the 12-month window yields abandonment and a lost priority date (MPEP § 711). Treat month 12 as a decision deadline, not a default.

Run the two tests at month 9, not month 12: (1) is there real commercial validation, and (2) is the United States the priority market? Score each as yes or no. Month 9 is the last practical point to prepare the chosen filing without emergency drafting; month 12 is too late to discover that the answer is unclear.

Define validation strictly: evidence of demand, not enthusiasm. A paid pilot, purchase order, signed letter of intent, or repeatable customer commitment scores yes; polite interest, a demo request, or an unpriced “would buy” scores no. If the evidence is thin, the score is no even when the invention feels promising.

If both tests are yes, convert—and budget for the non-provisional step from day one, not month 11. That step includes claims, drawings, and prosecution-ready drafting; as priced and warned above, Commercify notes that the "cheap provisional plus expensive conversion" pattern usually costs more total than planning the expensive step early.

If validation is real but the United States is secondary, file a PCT application. It preserves the priority date while deferring national-phase decisions, so the US question can be revisited after more market evidence. If validation is weak but the United States is priority, do not convert the thin provisional; refile a stronger application built on better evidence, because converting a weak filing buys a weak priority date.

If neither test passes, deliberately abandon. Letting the provisional lapse is a decision, not a failure; it avoids paying the full non-provisional conversion cost for an invention the market has not validated.

Where an express abandonment is wanted, it is filed as a written declaration identifying the application (MPEP § 711.03(a)).

What to do next

Frequently Asked Questions

What two binary tests must be run at the month-9 checkpoint before the provisional deadline?

The two binary tests are real commercial validation and US-as-priority-market, and conversion is supported only if both pass.

Does a provisional automatically mature into a non-provisional after its priority period?

Nothing converts automatically; the applicant must affirmatively file a non-provisional application claiming benefit of the provisional before the deadline.

Under what statute does abandonment occur after the provisional's priority period?

Abandonment occurs under 35 U.S.C. § 111(b) after the provisional's 12-month period.

What is required for express abandonment of a provisional?

Express abandonment requires a written declaration to the USPTO.

Is conversion the default path when the provisional deadline arrives?

Conversion is one of four paths, not the default.

What four paths can be chosen at the provisional deadline?

The four paths are convert, PCT, refile stronger, and deliberate abandon.

Quick answers

Also worth reading: Patent Sustainability Map: World Patent Body Class Map Flags 33% Overlap—Report or Not?: Patent Sustainability Map: World Patent · The 3.5/7.5/11.5 Patent Fee Clock Starts at Grant, Not Filing: 3.5/7.5/11.5 Patent Fee Clock Starts · Patent Renewal Fees 2026: $7,700 Prune vs Renew Decision: Patent Renewal Fees 2026: $7,700