Why grant reviewers mark TRL down: six common overstatements
Six TRL assessment mistakes that make grant reviewers mark a proposal down, how NTRAF pegs a missing item one level lower, and how to state a TRL that holds.
Part of the guide TRL levels explained: what TRL 1 to 9 really mean, and how to place your technology
Grant reviewers mark TRL down when the evidence does not support the level claimed. Six overstatements are common: rating a system at its best part, calling lab tests a relevant environment, ignoring scale, counting planned work, one run and a skipped regulatory gate. India's NTRAF pegs a technology one level lower if any critical item or document is missing.
A familiar pattern in funding rounds: a proposal claims TRL 6, the evidence supports TRL 4, and the reviewer has to decide whether the applicant misunderstands the scale or is stretching it. Either reading hurts. The project may be excellent at TRL 4, but it now looks either careless or optimistic.
India now has a formal answer to how strictly readiness should be read. The National Technology Readiness Assessment Framework (NTRAF), from the Office of the Principal Scientific Adviser, names optimism bias as a recurring problem: perceived readiness that runs ahead of the actual engineering status. Its rule is blunt. If any critical item at a level is answered no, or its document is missing, the technology is pegged at the next lower TRL.
Here are six overstatements that come up again and again in proposals, with what to write instead.
1. Rating the system at its best component
A device has a sensor validated in a simulated setting, a signal-processing board tested on the bench and an enclosure that exists as a drawing. The proposal says TRL 5.
NTRAF assesses each critical technology element separately, and notes that TRL on its own did not consider how technologies interact, which is why system and integration readiness measures were developed. The honest statement is the level of each critical part, and a system level no higher than the weakest critical part.
Write instead: "Sensor: TRL 5. Processing board: TRL 4. Integrated system: TRL 4, with integration testing as milestone 1."
2. Calling the laboratory a relevant environment
The jump from TRL 4 to TRL 5 is the move from lab validation to validation in a relevant environment. Many proposals treat "tested on real samples in our lab" as the second. It is still the first.
A relevant environment reproduces the conditions that matter for the real use: the temperature range, the sample matrix, the user, the duty cycle. Describe where the test happened and what conditions it reproduced. If the answer is "our bench, our technician, spiked samples", that is TRL 4 evidence, and good TRL 4 evidence is worth having.
3. Ignoring scale
A process that works at 100 millilitres has not shown it works at 100 litres. NTRAF's wording climbs through scale explicitly: lab-scale at TRL 5, engineering or pilot scale at TRL 6, full-scale prototype at TRL 7. Reviewers in chemistry, biotech and materials look for the batch size before they look for the result.
Write instead: the scale of every result you cite, in units.
4. Counting planned work as done
"We will validate the prototype in a hospital setting in the first six months" sometimes leaks into the TRL claim itself. The current TRL is what has been done and documented on the day you submit. NTRAF asks funders to record a starting TRL and a target TRL in the proposal, and to make progression a review parameter. Use that structure: start, target, and the milestones between them.
5. One good run
A single successful experiment shows that something can work. It does not show that it works reliably. Reviewers ask how many runs, batches or units, and how consistent the results were. Three batches within specification say far more than one batch with a better headline number.
6. Skipping the regulatory gate
In regulated sectors, readiness is tied to regulatory steps. For medical devices in India, one common reading of device TRL places a frozen design and a test licence application on Form MD-12 around the advanced proof-of-concept stage, with test batches made under a test licence on Form MD-13 next. Check the scale your funder names. Claiming TRL 6 for a device that has never had a test licence invites a mark-down. Drug projects have the same issue with GLP toxicology and clinical trial permissions. The sector posts on medical devices and drug development set out the gates.
Six overstatements reviewers mark down
How the overstatements map to the levels
| Overstatement | Level usually claimed | Level usually supported | Evidence that would close the gap |
|---|---|---|---|
| System rated at best component | 5 | 4 | Integration test report |
| Lab called relevant environment | 5 | 4 | Test in reproduced use conditions |
| Scale ignored | 6 | 5 | Pilot-scale batches with data |
| Planned work counted | 6 | 5 | Completed validation report |
| One good run | 4 | 3 | Repeat runs within specification |
| Regulatory gate skipped (device) | 6 | 4 or 5 | Test licence (MD-13) and test batch data |
These pairings are typical, not fixed rules. Your sector's definitions decide the exact level.
Why it matters more than it used to
Calls increasingly set a minimum readiness. The ICMR Intramural MedTech Call 2.0 (released January 2026) considered only proposals for technologies that had completed TRL 3, meaning proof of concept demonstrated, or that would cut cost by at least 10% against existing technology. ICMR's June 2026 call for Centres for Advanced Research in priority medical devices and diagnostics said it preferred applicants already working on functional prototypes at TRL 3 or 4 or above. When a threshold exists, an overstated claim is checked against it first.
The same project, stated two ways
The second earns trust, and often the grant.
What reviewers do with an inflated claim
A reviewer who finds an overstated TRL rarely stops at lowering the number. The finding colours the rest of the proposal. If the starting level is lower than claimed, the milestones look too ambitious, the budget looks mismatched to the work and the timeline looks unrealistic. A project that would have been a good fit at its real level can end up scored as a weak fit at the claimed one. That is why an accurate TRL is a strategic choice, and why it is worth having a colleague outside the project read the claim against the evidence before submission.
How to state TRL so it holds
- Identify the critical technology elements, the parts without which the product does not work.
- For each, find the highest level at which every required item is done and documented.
- Take the system level as the lowest of these, unless integration has been tested.
- Attach the evidence: the report name, date, scale and who ran it.
- State the target TRL and the milestones that reach it.
The honest hedge: NTRAF calls itself an initial proposal of guidelines, and many funders still use their own sector definitions, such as BIRAC's. Use the definitions your call names. The discipline of claiming only what is documented applies under any of them. For the evidence itself, how to prove your TRL gives a level-by-level list.
If you want a structured first pass before you write, TRL Pro assesses technology readiness from TRL 1 to 9 using Zernyx's own copyrighted framework (ZTRAF), which covers more than 60 technology domains and subdomains: you answer questions about your technology and its evidence, and get back an assessed level.
Common mistakes in the written proposal
- A TRL in the summary that differs from the TRL in the methodology section.
- Evidence attached without dates, scale or conditions.
- A target TRL two or three levels above the start in a one-year project.
- Using the word "validated" for an experiment that was run once.
My view: reviewers forgive an early TRL far more readily than an inflated one. Claim the level you can prove.
Common questions
What happens if I overstate TRL in a grant proposal?
Usually the reviewer assesses a lower level, and the proposal then looks mismatched with the funding stage it asks for. Some calls set a minimum, such as the ICMR Intramural MedTech Call 2.0 (2026), which considered only technologies that had completed TRL 3, so an unsupported claim can also mean the proposal is out of scope.
How does NTRAF treat a missing piece of evidence?
The National Technology Readiness Assessment Framework, from the Office of the Principal Scientific Adviser, says that if any critical criterion at a level is answered no, or documentation is missing, the project cannot be classified at that level and is pegged at the next lower TRL.
Should I claim the TRL of my best component or the whole system?
Neither as a single number. NTRAF assesses each critical technology element separately. State the TRL of each critical part, and the system's TRL as the level all of them have reached together.
Is it better to understate TRL?
No. Understating can make a project look too early for the call. Claim the highest level at which nothing is missing, and show the next level as a funded milestone with dates.