# TRL levels explained: what TRL 1 to 9 really mean, and how to place your technology

URL: https://zernyxhub.com/blog/trl-levels-explained
Author: Dr. Abhay Tharmatt, Research Scientist, R&D, Zernyx Innovations
Published: 2026-10-17
Category: Technology readiness

> Technology readiness levels (TRL 1 to 9) explained with plain examples, the TRL 4 vs 5 vs 6 gap, India's new NTRAF framework and common mistakes.

**Short answer:** Technology readiness levels (TRLs) are a nine-step scale for how mature a technology is, from TRL 1 (basic principles observed) to TRL 9 (a proven system in real operation). TRL 1 to 3 is research, TRL 4 to 6 is validation and prototyping, and TRL 7 to 9 is demonstration and deployment. A technology sits at the highest level for which you can show evidence, not the level you expect to reach.

Ask a researcher, a startup founder and an investor what TRL a technology is at, and you will often hear three different numbers. That is not because the scale is vague. It is because people grade on hope rather than evidence. This guide explains each level in plain terms, shows where the hard jumps are, and tells you how Indian funders and frameworks now use the scale.

## What technology readiness levels are

Technology readiness levels are a nine-point scale for measuring how mature a technology is. NASA developed the scale for space programmes, and it has since been adopted by research funders, defence agencies and industry around the world, including the European Union's research programmes.

In India, BIRAC uses TRLs to select and monitor projects, and has adapted the NASA scale into separate definitions for areas such as drugs, vaccines, medical devices and diagnostics, agriculture and software. In December 2025 the Office of the Principal Scientific Adviser, working with CII, unveiled the National Technology Readiness Assessment Framework (NTRAF) to give the country one common way of assessing TRL. More on that below.

The scale is useful because it gives everyone a shared language. A licensee, a grant committee and a TTO can all see that "TRL 4" means something different from "TRL 6", and price, fund or plan accordingly.

## The nine levels, in plain terms

Wording differs slightly between frameworks. The meaning is broadly the same.

| TRL | Standard description | What it looks like in practice |
|---|---|---|
| 1 | Basic principles observed and reported | A scientific effect is seen and written up. No application yet. |
| 2 | Technology concept or application formulated | Someone proposes how the effect could be used. Mostly paper and analysis. |
| 3 | Analytical and experimental proof of concept | The key function works in a lab experiment, usually on separate parts. |
| 4 | Component or breadboard validated in the laboratory | The parts are put together and work as a rough system on the bench. |
| 5 | Component or breadboard validated in a relevant environment | A higher-fidelity version is tested in conditions that simulate real use. |
| 6 | System or prototype demonstrated in a relevant environment | A representative prototype performs all its key functions in near-real conditions. |
| 7 | System prototype demonstrated in an operational environment | A near-final prototype works in the actual setting: a plant, a hospital, a field. |
| 8 | Actual system completed and qualified | The final product passes its tests and, where needed, its certifications. |
| 9 | Actual system proven in operation | The product runs in real use across the full range of expected conditions. |

A simple way to remember it: **TRL 1 to 3 is "does the science work?", TRL 4 to 6 is "does the technology work outside the textbook?", and TRL 7 to 9 is "does the product work for customers?"**

**The nine technology readiness levels**
- TRL 1: Basic principles observed and reported
- TRL 2: Concept or application formulated
- TRL 3: Proof of concept in a lab experiment
- TRL 4: Parts work together on the lab bench
- TRL 5: Close-to-final version in realistic conditions
- TRL 6: Prototype does the whole job in near-real conditions
- TRL 7: Near-final prototype in the actual setting
- TRL 8: Final product passes its tests and certifications
- TRL 9: Proven in real use across expected conditions
(Valley of death: levels 4 to 6)
TRL 1 to 3: does the science work? TRL 4 to 6: does it work outside the textbook? TRL 7 to 9: does the product work for customers?

## A worked example

Take an illustrative case: a university lab developing a low-cost membrane that removes arsenic from groundwater.

- **TRL 1:** The team observes that a modified polymer binds arsenic ions. They publish the result.
- **TRL 2:** They propose using it as a filter membrane for village hand pumps.
- **TRL 3:** In the lab, a small sample removes arsenic from a prepared solution. That is proof of concept.
- **TRL 4:** They cast a small membrane, fit it in a bench rig with a pump and housing, and show the parts work together on lab-prepared water.
- **TRL 5:** They run the rig on real groundwater samples carried in from affected districts, at realistic flow rates and temperatures, with a membrane made the way it would be made at scale.
- **TRL 6:** They build a full-size prototype unit and run it at a test site in conditions close to a real installation.
- **TRL 7:** A near-final unit is fitted to a working hand pump and used by a community for an extended period.
- **TRL 8:** The final design is manufactured and passes the relevant water quality standards and certifications.
- **TRL 9:** Units are installed and running across many sites, in different seasons and water conditions.

Each step is not more effort on the same thing. Each step changes what is being tested, or where.

## TRL 4 vs TRL 5 vs TRL 6: where most technologies stall

The jump from 3 to 4 is usually manageable inside a lab. The stretch from 4 to 6 is where many good technologies stall, because it needs money, engineering skills and test conditions that a research group rarely has. It is the part of the journey people call the valley of death.

The differences are specific.

- **TRL 4 is lab environment, low fidelity.** The parts are integrated, but the setup is ad hoc. It may use lab-grade materials, a hand-built rig and idealised inputs. It shows the concept holds together.
- **TRL 5 is relevant environment, higher fidelity.** The configuration is much closer to the final design, and the test conditions are realistic: real samples, real temperatures, realistic loads. The question is no longer "do the parts work together?" but "do they still work when conditions are messy?"
- **TRL 6 is a prototype demonstrated in a relevant environment.** You now have a representative model of the actual system, at or near the right scale, doing all the jobs it needs to do. This is where engineering development properly begins.

[Image: Three cards comparing TRL 4 (parts work together on an ad hoc bench rig with ideal lab inputs), TRL 5 (close-to-final design tested in realistic conditions with real-world samples) and TRL 6 (a working prototype does the whole job in a near-real setting)]

The test to apply is simple. If your best data comes from ideal inputs on a bench rig, you are at TRL 4, however good the results. If it comes from realistic inputs on a configuration close to the final design, you can argue TRL 5. If a representative prototype has done the whole job in near-real conditions, you are at TRL 6.

This gap matters commercially. Industry licensees are usually much more willing to commit to a technology once it has evidence at TRL 5 or 6, because the remaining risk becomes an engineering problem rather than a scientific one. That is also why [deals stall at the readiness stage](https://zernyxhub.com/blog/university-tt) when nobody has said clearly what it would take to move up.

## India's NTRAF framework

Until recently, Indian ministries, funders and institutions each used their own modified TRL definitions, which made assessments hard to compare. The National Technology Readiness Assessment Framework from the Office of the Principal Scientific Adviser is meant to fix that.

Based on what PSA has published:

- It provides a single core framework, with additional questionnaires for **healthcare (pharma and biotech)** and **software**.
- Assessment runs in stages: screening questions to place a technology in a TRL band, top-level questions to set an "anticipated TRL", and detailed verification questions that need **supporting documents**.
- It is designed to underpin future online assessment and to support R&D funding decisions under national missions.

The draft was opened for public comment until 31 January 2026. Check psa.gov.in for the current version before relying on any specific wording.

The direction is clear even if details evolve: Indian funders want TRL claims backed by evidence, not just a number on a proposal form.

## How to assess your own TRL honestly

1. **Define exactly what you are assessing.** A component, a subsystem or the whole product? Each can have its own TRL. A product is only as ready as its least ready critical part.
2. **Start at TRL 1 and work up.** For each level, ask: what document, data set or test report proves we have done this? Stop at the first level where the answer is "not yet".
3. **Describe the test environment precisely.** "Tested in the lab" and "tested on real samples under field conditions" are different TRLs. Write down which one you did.
4. **Check scale and materials.** Results on hand-made samples do not prove that a manufactured version will behave the same way.
5. **Look beyond the technology.** Manufacturing readiness, regulatory approvals and supply of materials can hold a project back even at a high TRL. Sector frameworks such as BIRAC's build regulatory milestones into the levels for this reason.
6. **Get a second opinion.** Someone outside the team will spot the gap you have stopped seeing.

If you want a structured way to do this, [TRL Pro](https://app.zernyxhub.com/trl-calculator) walks you through level-by-level questions and records the evidence behind each answer, which is useful for grant proposals and licensing discussions.

## Common mistakes

- **Grading on the plan, not the evidence.** "We will test in the field next quarter" is not TRL 7.
- **Taking the TRL of the best component.** A world-class sensor in an untested system does not make the system TRL 6.
- **Confusing a demo with validation.** One good run for visitors is not the same as repeatable test data.
- **Ignoring the environment.** A prototype that only works in an air-conditioned lab is not validated for a dusty plant floor.
- **Using one framework's wording for another's form.** BIRAC's sector definitions, NTRAF and international scales differ in detail. Use the one your funder or partner uses.

## Why TRL matters for licensing and value

TRL is one of the strongest drivers of what a technology is worth to a licensee. Each level up removes risk and reduces the time and money the licensee must spend before it earns anything. That is why the same patent can attract a modest upfront fee at TRL 3 and a much stronger offer at TRL 6. Our guide on [how to put a price on a technology](https://zernyxhub.com/academy#value-tech) shows how readiness feeds into the cost-to-market estimate, and [how to license a technology from an Indian university or lab](https://zernyxhub.com/academy#license-tech) covers how buyers check readiness during due diligence.

Being honest about TRL also builds trust. A licensee who finds that "TRL 6" was really TRL 4 will question everything else in your pitch.

## FAQ

### What is the difference between TRL 4 and TRL 5?

At TRL 4, integrated components are validated in a laboratory with a low-fidelity setup. At TRL 5, a higher-fidelity version close to the final design is validated in a relevant environment that simulates real conditions.

### What TRL is a working prototype?

It depends on where it has been tested. A prototype that does all key functions in near-real conditions is TRL 6. If it has worked in the actual operating environment, it is TRL 7.

### Does a granted patent mean a higher TRL?

No. A patent can be granted at TRL 2 or 3, because patentability depends on novelty, inventive step and industrial application, not on how developed the technology is.

### Which TRL framework should I use in India?

Use the one your funder or partner asks for. BIRAC has sector-specific definitions for biotech areas, and PSA's NTRAF aims to become a common national standard. The international NASA-style scale remains the common reference for global partners.

### Can software have a TRL?

Yes. Several frameworks, including BIRAC's and NTRAF, have software-specific criteria, where the "relevant environment" might mean real user data, real integrations and realistic load.

---

Published by Zernyx Hub. More: https://zernyxhub.com/llms.txt