A Better Ladder for Judging Progress
SBIR’s phase boundaries and its funding gap point to one practical judgment: progress is not a single switch that flips from “research” to “commercialized.” A project can resolve a technical uncertainty, attract an investor, test a customer use, or make an initial sale while still lacking the production, procurement, repeat demand, or economics needed for durable adoption. The useful task here is to connect those forms of progress without treating them as equivalent, then use that model to read reported SBIR outcomes with precision.
From phase movement to evidence movement
The phase structure tells us where public R&D support sits in the project’s life. The valley-of-death problem tells us why movement through that structure can stop: the next source of money, the next technical capability, and the next user decision may not arrive at the same time. Put together, they yield a more useful model than either idea alone.
A phase label describes the setting and purpose of funded work. An outcome describes something that happened to the technology, the firm, or a user. Those descriptions can be related without being identical. A Phase II award may support a prototype, but “prototype exists” is an outcome claim about an artifact, not a claim about a customer. A non-SBIR investment may extend the runway, but “investment received” is a claim about capital, not a claim about sales. A Phase III contract may demonstrate a government procurement path, but it does not automatically show that private customers adopted the same technology.
The distinction matters because the strongest evidence changes as the project moves outward from the research setting. Early evidence is usually about whether an approach works. Later evidence is about whether it works reliably in a relevant setting, whether another organization will commit resources to it, and whether the resulting use can continue. The project does not stop being technical when it approaches a market; rather, technical performance becomes one condition among several that other parties must evaluate.
This also explains why phase progression is not a reliable substitute for an outcome record. A project can receive successive support because an agency believes that additional R&D is worthwhile. That support may preserve expertise and equipment, reduce technical risk, or make a later transition more plausible. But the project still has to encounter external decisions. The next funder, customer, or acquisition office may ask a question that the prior phase was never designed to answer.
The central relationship is therefore a chain of dependencies. Technical work creates something that may be worth testing. Testing can reveal whether a customer or mission user has a real need. A commitment from that user or from a funder can create the resources for production or integration. Production and integration can reveal whether use is repeatable and economically sustainable. Each link strengthens the case for commercialization, but no single link can stand in for all the others.
The outcome ladder
A helpful ladder begins with technical completion and becomes more demanding as it approaches sustained use. It is not a universal scorecard: different technologies have different users, timelines, and forms of value. Its purpose is to make the claim boundary visible.
At the first rung is technical progress. This includes the feasibility evidence associated with early R&D, a working demonstration, or a prototype that embodies the proposed design. Such evidence answers whether the approach can work in the form and conditions tested. It may be exactly the right achievement for an early project. It does not by itself answer whether the result can be produced, purchased, integrated, or maintained.
The next rung is transition readiness. Here the project begins to address the conditions outside the original research setup. Production capacity may increase. A potential customer may test the technology. An agency may provide related follow-on support.