The weakest page in most STTR proposals
Read enough STTR submissions and one page fails more often than any other: the commercialization section. It is usually a market size, a growth rate, a chart of a total addressable market, and three sentences about a transition partner who has not been contacted. Every number on it was generated by the people asking for the money. An evaluator reading that page has no way to tell a serious plan from a hopeful one, so the safe move is to discount all of it. Meanwhile, the same proposal has a research institution attached to it whose transition history is sitting in a public database, unmentioned.
The STTR program exists because Congress wanted federally funded research to leave the building. The statute sets it up structurally: an STTR award requires a small business and a nonprofit research institution working together, with the small business performing at least 40 percent of the work and the institution at least 30 percent. The remaining 30 percent can go to either party or to a third. The program is funded by a 0.45 percent set-aside on the extramural research and development budgets of the five participating agencies, and the small business, not the institution, holds the contract and carries the commercialization obligation.
That structure creates a fact most teams never use. The institution on the proposal has spent decades producing exactly the kind of evidence the commercialization section is asking for, and it has been legally required to record it.
Why the institution's record is third-party by construction
Under the Bayh-Dole Act, 35 U.S.C. §§200 through 212, and its implementing regulations at 37 CFR Part 401, a nonprofit or small business contractor may elect to retain title to inventions made under federal funding. Election comes with duties. The contractor must disclose each subject invention to the funding agency within two months of the inventor's written report, elect title within two years of disclosure, file patent applications on schedule, and report on efforts to bring the invention to practical application. The standard patent rights clause at FAR 52.227-11 puts these obligations in the contract itself.
The reporting flows through the federal interagency invention reporting system, iEdison, now administered by the National Institute of Standards and Technology. Utilization reporting under 35 U.S.C. §202(c)(5) is treated as confidential and is not something an outside party can pull, so the raw agency file is not the source to use. The useful source is what the institution publishes about itself.
Nearly every research university reports annually to the AUTM Licensing Activity Survey: invention disclosures received, patents issued, license and option agreements executed, startups formed, and licensing income. Most technology transfer offices publish an annual report on their own site with the same figures. Department of Energy laboratories report through the Office of Technology Transitions annual report, which counts cooperative research and development agreements, licenses, and agreements for commercializing technology. Federally funded research and development centers report through the Federal Laboratory Consortium. Patents and their assignment history are in USPTO Patent Public Search and the Patent Assignment Search. Sponsored awards are in NIH RePORTER, the NSF award search, and USAspending.gov. A spin-out's first outside raise usually appears in an SEC Form D.
Every one of those records was created by someone other than the person writing the proposal. A licensee signed and paid. A sponsor committed money. A registrar issued a patent. A company filed with the Commission. That is what makes the record third-party evidence rather than a projection, and that distinction is the entire argument of this article.
Evidentiary weight — transition artifacts as an evaluator reads them
Editorial weighting from public sources and practitioner reading — illustrative, not a measured statistic.
What counts, in order
An executed license in the same technology family. The strongest single row. Somebody with a budget read the disclosure, negotiated a field of use, and signed. Even when the licensee's name and terms are confidential, the technology transfer office can usually confirm the year, the class of licensee, and the field of use in writing.
A spin-out with outside money. A company formed on the group's intellectual property that then raised capital or won its own award clears two hurdles at once: the technology was worth building a company around, and an investor or a contracting officer agreed. The formation date is in state registration records and the raise is often in a Form D.
Follow-on funding from a non-academic sponsor. A program office, a prime, an industry consortium, or a state agency that put money into the same line of work after the basic research ended. This is the single most underused row. It is easy to verify in USAspending.gov or an agency award search, and it answers the question the commercialization section is actually asking, which is whether anyone outside the academy has ever wanted this.
Adoption without a dollar figure. A model, dataset, code base, standard, or method from the group that a government organization or a standards body picked up. A tool cited in an agency handbook. A dataset that became a community benchmark. A method written into a test procedure. Adoption is a transition even when no money changed hands, and it is checkable.
Issued patents with live prosecution. Useful, and weaker than the rows above, because a patent proves novelty and institutional investment rather than demand. Continuations and foreign filings raise the weight, since institutions do not pay international filing fees on a portfolio nobody wants.
People. Where the group's graduates went. When former members of the lab now work at the transition target, the integrator, or the program office, that is a real pathway and it is worth one line.
The exhibit itself
Build it as one table, on one page, with a verification pointer in every row. Rows without a pointer come out. The discipline of requiring a pointer is what turns a paragraph of adjectives into a document an evaluator can audit in four minutes and then stop questioning.
| Artifact | What it proves | Where an evaluator checks it |
|---|---|---|
| License executed | A buyer with a budget wanted this enough to sign | TTO annual report; written TTO confirmation of year and field of use |
| Startup formed | The technology supported a company, not just a paper | State registration; SEC Form D; the institution's startup roster |
| Sponsored follow-on | A non-academic sponsor continued the same work | USAspending.gov; NIH RePORTER; NSF award search |
| Patent issued and assigned | Novelty plus institutional willingness to pay for it | USPTO Patent Public Search; Patent Assignment Search |
| Lab or FFRDC agreement | A federal laboratory chose to co-develop | DOE Office of Technology Transitions annual report; FLC listings |
| Field adoption | Somebody uses the output operationally today | Agency handbook, test procedure, or standards citation |
Format each row the same way: year, artifact, counterparty class, what changed hands, pointer. Five columns, no prose. Then write two sentences under the table connecting the pattern to the proposed work, and stop. The table does the persuading.
Getting it in ten working days
Our teams run this as a fixed sequence, because the long pole is never the writing. It is the technology transfer office's queue.
Building the transition-record exhibit
Two details save the schedule. Ask the technology transfer office for the principal investigator's portfolio rather than the institution's totals, because institutional totals invite the objection that the number belongs to somebody else's lab. And ask for clearance in the same message as the request, so the office reviews once instead of twice.
The allocation of rights agreement is part of the evidence
Every STTR award requires a written agreement between the small business and the research institution allocating intellectual property rights and rights to carry out follow-on research, development and commercialization. Most teams treat it as a form to clear before award. It is better read as the closing argument of the commercialization section, because it answers the question the transition record raises next: who can actually sell this.
An agreement that names the background intellectual property, states who owns what arises under the award, grants the small business a defined commercialization license, and sets a field of use turns the lab's transition history into a path the small business can walk. An agreement that defers everything to a later negotiation leaves the evaluator with a good history and no route from it. Data rights work alongside this: the protection period for SBIR and STTR technical data and software under DFARS 252.227-7018 runs 20 years from award, which is long enough for a real product line and worth stating plainly.
Where the line is
The institution's record is not the small business's past performance, and presenting it as such is a fast way to lose credibility with a contracting officer. Say what it is: the transition history of the research lineage this award builds on. That framing is both honest and strong.
It is also directly supported by the acquisition rules. FAR 15.305(a)(2)(ii) provides that a solicitation should authorize offerors to submit past performance information for predecessor companies, for key personnel with relevant experience, and for subcontractors performing major or critical aspects of the requirement. On an STTR, the research institution is a subcontractor performing a statutory minimum of 30 percent of the work, and the principal investigator is key personnel. The record of both is squarely in scope. Separately, most Department of Defense SBIR and STTR Phase I evaluations weigh technical merit, the qualifications of the team and facilities, and commercialization potential, without a past performance factor at all, which makes the transition record a contribution to the third criterion rather than a substitute for something else.
Keep two more boundaries. Do not present confidential license terms; present the year, the counterparty class, and the field of use, cleared in writing. And do not stretch the technology family. A license on an unrelated technology from the same department proves institutional competence at licensing and nothing about this work, and a sharp reviewer will notice the gap.
Why this outperforms a market projection
Program managers see market projections constantly and have learned what they are worth. The numbers come from the same handful of analyst reports, the growth rates are rounded up, and the addressable market is drawn so broadly that a one percent capture reads as a business. None of it is checkable, so none of it moves a score.
A transition record is checkable, which changes the reader's posture from judging a forecast to confirming a fact. It also speaks to the risk the program manager actually carries. A Phase I is a small bet. A Phase II is above two million dollars under the statutory guideline amounts that the Small Business Administration adjusts for inflation each year, and Phase III is where a technology either enters a program of record or quietly stops. Under 15 U.S.C. §638(r)(4), an agency may award a Phase III sole source to the firm that developed the technology, with no further competition required. The program manager evaluating a Phase I proposal is already thinking about whether that path exists. A lineage that has produced licenses, spin-outs and sponsor follow-ons is direct evidence that it does.
The best version of the commercialization section makes both arguments in order. Here is what this lineage has already transitioned, verified. Here is the rights structure that lets this small business carry the next step. Here, briefly, is the market. The projection stops being the case and becomes the last paragraph of it.
Common questions on using the record
Our licenses are confidential. Can we still use them?
Yes. Terms and licensee names are frequently confidential; the existence, year, and field of use usually are not. Ask the technology transfer office to confirm in writing what may be stated, and cite the office's own published aggregate figures for context. A row reading "2023, exclusive license, medical device manufacturer, imaging field of use, confirmed by the Office of Technology Transfer" is strong and discloses nothing protected.
What if the group has never licensed anything?
Then the record is built from the other rows. Sponsored follow-on funding from a non-academic sponsor, a federal laboratory agreement, adoption of a tool or dataset by an agency, and where former group members now work are all transition evidence and all verifiable. Many strong groups have thin licensing histories and deep sponsor histories, and sponsors are the more persuasive row for a defense customer anyway.
Does this apply on an SBIR without a research institution?
The same method applies to any partner whose record is external to the offeror: a federal laboratory under a cooperative agreement, a consultant with a documented product history, a hospital system, or a commercial partner. The mechanism is not the STTR structure itself. It is that somebody outside the proposal team already committed resources, and the commitment is on the record.
How long should the exhibit be?
One page, six to ten rows, with the strongest row first. If the table needs a second page, the technology family definition is too loose. A tight table with six checkable rows outscores a long table with twenty rows an evaluator has to sort.
Frequently asked questions
The small business must perform at least 40 percent of the work and the research institution at least 30 percent, measured in the manner the agency specifies. The remaining share may go to either party or to a third performer. The small business is the awardee and holds the contract.
The AUTM Licensing Activity Survey and the institution's own technology transfer annual report carry disclosures, patents, licenses, startups and income. Patents and assignments are in USPTO Patent Public Search. Sponsored awards are in USAspending.gov, NIH RePORTER, and the NSF award search. Department of Energy laboratory activity is in the Office of Technology Transitions annual report.
Not as the offeror's own record. Present it as the transition history of the research lineage and of key personnel. FAR 15.305(a)(2)(ii) contemplates past performance information from key personnel and from subcontractors performing major or critical aspects of the requirement, which is what the institution is on an STTR.
It allocates intellectual property rights between the small business and the research institution and addresses rights to carry out follow-on research, development and commercialization. A strong one names background intellectual property, assigns ownership of what arises under the award, and grants the small business a defined commercialization license with a stated field of use.
Under DFARS 252.227-7018, the protection period for SBIR and STTR technical data and computer software runs 20 years from the date of the award that generated it. Marking practice matters, so the legends should be checked before delivery rather than after.
Bottom line
The commercialization section is the easiest place in an STTR proposal to gain ground, because most submissions fill it with numbers nobody can check. The research institution already carries a record that outside parties created, that public databases hold, and that a technology transfer office will confirm in about a week. Put it in a one-page table with a pointer in every row, pair it with an allocation of rights agreement that shows who can sell what, and the section stops being the weakest page and starts being the reason the proposal is believed.