Two NSF Grandfather Clauses Expire on the Same Day: What November 4, 2026 Does to Companies Stranded by the SBIR Shutdown

There is a specific kind of grant deadline that does not feel like a deadline: the one where you already have the thing. You hold an NSF Project Pitch invitation. Or you hold an NSF Phase I award with a Phase II you have always intended to write. Nothing is pending. Nothing is due. The asset just sits there.

On Wednesday, November 4, 2026, two of those assets expire simultaneously for two entire cohorts of companies — and both cohorts exist only because the SBIR and STTR programs lapsed at the end of FY2025 and NSF's competition went dark for months.

The transition language is buried in NSF's restart materials rather than announced as a deadline, which is exactly why it is worth spelling out.

The two clauses

Clause one — legacy Project Pitch invitations. Companies holding a Project Pitch invitation issued on or after July 3, 2025 that have not yet been converted into a submitted proposal may submit their full Phase I proposal — or a Fast-Track proposal, where permitted — no later than November 4, 2026.

Clause two — legacy Phase I awardees. Phase I awardees whose award start date was on or after November 6, 2023, and who have not already submitted a Phase II proposal, may submit that Phase II proposal no later than November 4, 2026.

Both are accommodations, not penalties. NSF is extending validity to people the shutdown stranded. But an accommodation with a date on it behaves exactly like a deadline once the date passes, and the consequences on the two sides are different.

Why these cohorts exist

The standing rule is that a Project Pitch invitation is valid for the next two submission deadlines after the date of the official invitation. Under normal operations that is a comfortable window — roughly eight months of runway across NSF's quarterly cycles.

Then the cycles stopped. NSF's SBIR/STTR competition went quiet through the reauthorization lapse and restarted in mid-2026 with two new solicitations: NSF 26-510, the main deep-technology program, and NSF 26-511, a new pilot emphasis on scientific instrumentation. New Project Pitches began being accepted again on June 2, 2026. The first full-proposal deadline under the restarted program was July 27, 2026. The next is November 4, 2026, then March 4, 2027, then July 7, 2027 — and annually thereafter on the first Wednesday in November, the first Thursday in March, and the first Wednesday in July.

Do the arithmetic on a company invited in, say, August 2025. Its "next two submission deadlines" would ordinarily have arrived in late 2025 and early 2026. Neither happened. Rather than void those invitations, NSF is letting the July 3, 2025 cohort ride to November 4, 2026 — which, counted against the restarted calendar, is precisely two deadlines: July 27 and November 4. The rule was not suspended. It was re-anchored to the deadlines that actually occurred.

Same logic on the Phase II side. A company whose Phase I started in, for example, early 2024 would normally have had a defined Phase II submission window keyed to its award. Those windows evaporated when there was nothing to submit into. The November 6, 2023 anchor date sweeps in essentially everyone whose Phase II window was disrupted and gives them one shared, final opportunity.

What is actually at stake

For the Project Pitch cohort, the stated consequence of expiration is clean and unforgiving: if a Project Pitch invitation expires, the PI must restart the Project Pitch process.

That restart is not a formality, because Project Pitch submissions are rationed. A company may submit a maximum of two Project Pitches per 12-month period, and no more than three cumulative submissions for the same project or technology. If you have already used pitches on this technology — and many companies burned one or two during the pre-lapse period trying to land the right framing — a restart may put you against the cumulative cap on the concept you actually want to fund. In the worst case, letting an invitation lapse does not cost you a cycle. It costs you the concept.

For the Phase I cohort, the stakes are larger in dollars and just as final. NSF's Phase II ceiling under the restarted program is $1,250,000 over roughly 24 months, up from the $1,000,000 that longtime applicants have in their heads. Related mechanisms stack on top: Phase IIB adds up to $500,000 against a third-party match, and the Technology Enhancement for Commercial Partnerships supplement runs up to 20 percent of the Phase II award. A Phase I awardee who misses November 4 is not deferring that money to March. Phase II eligibility flows from the Phase I award, and the transition clause is the mechanism keeping these particular awards eligible at all.

If you are in either cohort and any part of your eligibility is ambiguous — an invitation you are not certain of the date on, a Phase I start date near the November 6, 2023 boundary, a Fast-Track question — the move is to email your NSF program director now rather than to reason it out from the solicitation text. Boundary cases are exactly what program officers exist to resolve, and they are far more useful in September than in the last week of October.

The rest of the November 4 window is worth knowing anyway

Even for companies outside both cohorts, November 4 is the next real door, and the restarted program looks different enough from the pre-lapse version that stale assumptions are a liability.

The award ladder. Phase I is up to $305,000 over 6 to 18 months, inclusive of direct and indirect costs, the small business fee, and Technical and Business Assistance. Phase II is up to $1,250,000. Fast-Track — Phase I and Phase II proposed together, where permitted — runs to $1,555,000. And the restart added a Strategic Breakthrough tier reaching up to $30 million, gated on NSF program director approval and 1:1 matching funds. That last tier is not a normal application path; it is a mechanism for a small number of nationally significant efforts, and treating it as a lottery ticket wastes everyone's time.

Two solicitations, one calendar. NSF 26-510 anticipates roughly $210 million and about 340 awards across all categories, funding across nearly all technology areas — NSF explicitly does not solicit specific technologies. NSF 26-511, the instrumentation pilot, carries up to $40 million, with roughly 57 to 59 Phase I SBIR awards, 11 to 12 Phase I STTR awards, about 10 Phase II SBIR and 2 Phase II STTR awards annually. The two share deadline dates, so the choice between them is a positioning decision, not a scheduling one.

The instrumentation pilot's actual thesis. NSF's rationale for 26-511 is unusually candid: traditional venture capital may not be incentivized to support next-generation scientific instrumentation because the market is small, even where building the capability domestically matters for national security and U.S. science leadership. That is an invitation, and it should change how you write. In 26-510 you are expected to argue a large addressable market. In 26-511, a modest market is the premise — the argument you owe is about who is scientifically stranded without this instrument, not about a hockey stick. Companies building experimental platforms, detectors, or the instrumentation that AI-driven discovery depends on have been squeezing themselves into the wrong narrative for years. This lane exists so they can stop.

Eligibility fundamentals that still trip people. The 500-employee small business ceiling includes affiliates — a fact that surprises venture-backed companies whose investors hold controlling positions across a portfolio. The PI's primary employment must be with the small business at the time of award and throughout, defined as at least 51 percent employed by the company. One proposal per organization and one per PI at a time. Deadlines are 5 p.m. submitter's local time, which is one of the very few federal deadline conventions that is not Eastern.

What to do this week

If you hold a Project Pitch invitation dated on or after July 3, 2025 and have never submitted against it, find the invitation email today and confirm the date. If it qualifies, November 4 is not your target — it is your wall. Work backward: a Phase I proposal you have not started needs the intervening weeks for the project description, the commercialization plan, and the registrations, and Research.gov has been under active revision to accommodate the restarted solicitations' requirements. Discovering a portal problem on November 3 is a category of self-inflicted loss that happens every cycle.

If your Phase I award started on or after November 6, 2023 and you have not submitted Phase II, treat this as the highest-value item on your fall calendar. It is a $1.25 million decision with a fixed expiry, and unlike most grant deadlines, you are competing from a position most applicants would trade for — you already have the award.

And if you are in neither cohort, use the window for what it is: the second real opportunity since NSF turned the program back on, with a genuinely new instrumentation lane that has not yet developed the applicant pileup the main program has. We covered the restart and its two-solicitation architecture when it landed; the November cycle is where the strategy gets tested.

Deadlines that only apply to a specific cohort, anchored to a date in a solicitation nobody re-reads, are the ones that quietly cost companies the most. Surfacing the ones that actually apply to you is the problem Granted was built to solve.

Sources: NSF 26-510 solicitation · NSF 26-511 solicitation · NSF America's Seed Fund solicitations page · Submitting a full SBIR/STTR proposal to NSF