Technology Readiness Level is a nine-point scale describing how mature a technology is, from an idea on paper to hardware proven in operational use. Agencies use it to scope solicitations, compare proposals, and decide whether a technology is ready for a program to adopt. Founders routinely rate themselves two or three levels above where the government would put them.
What the levels roughly mean
The scale originated at NASA and has been adopted, with variations, across federal agencies. In plain terms:
| Levels | What exists | Typical evidence |
|---|---|---|
| 1–3 | Basic principles, a concept, early analytical or lab proof | Papers, models, a bench experiment |
| 4–5 | Components or a breadboard validated in the lab, then in a relevant environment | Subsystem test data under representative conditions |
| 6–7 | A system prototype demonstrated in a relevant, then an operational, environment | Field or flight demonstration of an integrated prototype |
| 8–9 | The actual system completed, qualified, and proven in operations | Qualification test reports, operational use |
A Phase I typically sits around the low middle of the scale and a Phase II pushes toward the upper middle, but the expectation is stated in the topic rather than fixed by the program, so read the Phase I and Phase II comparison alongside the actual solicitation.
How agencies use the scale
Topics are usually written with an entry and an exit level: the agency states what maturity it expects you to start from and what the award should deliver. Reviewers use that to judge whether your plan is credible, and proposing to jump four levels in one phase reads as a lack of engineering realism — a point worth checking against the rest of our SBIR application guidance. Program offices then use the scale as a gate; many will not consider adopting a technology until it has been demonstrated at a defined level in an environment they recognize.
Why your number is probably too high
The disagreement almost always comes down to the word environment. A founder with a working prototype on the bench feels close to done. A program office asks whether that prototype ran in the actual environment — temperature, vibration, dust, electromagnetic conditions, an untrained operator — and whether it was a production configuration or a hand-built one. Those two questions typically cost several levels, and the gap between them is where most of the remaining engineering budget goes. Our overview of defense product development covers what qualification in a real environment involves.
Agencies also apply the scale differently. Definitions for software, for biomedical products, and for manufacturing readiness diverge meaningfully from the hardware scale, and some organizations track a separate manufacturing readiness measure alongside it. Ask the program office how it defines the levels for your topic rather than assuming — a good use of a conversation when talking to a program manager. Then rate yourself conservatively: an honest, slightly low self-assessment with evidence behind it persuades far better than an optimistic one a reviewer quietly discounts.
Projects House helps founders assess where a technology actually sits and plan the testing that moves it up. Start with the contact form.