The short answer
The type depends on what the results are expected to be used for (§2.37). Basic research seeks new knowledge with no particular application in view. Applied research seeks new knowledge for a specific, practical aim. Experimental development uses knowledge already gained to produce new products or processes, or to improve existing ones. All three must first pass the five criteria that make an activity R&D at all – see Is it R&D? A five-question test.
The three definitions
- Basic research
- Experimental or theoretical work undertaken primarily to acquire new knowledge of the underlying foundations of phenomena and observable facts, without any particular application or use in view (§2.25).
- Applied research
- Original investigation undertaken in order to acquire new knowledge, directed primarily towards a specific, practical aim or objective (§2.29).
- Experimental development
- Systematic work, drawing on knowledge gained from research and practical experience and producing additional knowledge, directed to producing new products or processes or to improving existing ones (§2.32).
Two questions that decide the type
The manual's key criterion is the expected use of the results (§2.37). To apply it, it suggests asking two things:
- How soon could the results be applied?
- How wide is the range of possible applications?
The more fundamental the work, the later and the broader its possible uses (§2.37). Basic research results are usually published or shared with colleagues rather than sold (§2.26); results of applied research are meant mainly for possible use in products, operations, methods or systems, and may be protected as intellectual property, including by keeping them secret (§2.31).
Same topic, three types
The clearest way to see the difference is one subject seen three ways. These are examples from the manual (§2.40, §2.41), shortened and in our words:
| Field | Basic research | Applied research | Experimental development |
|---|---|---|---|
| Chemistry | Studying a class of polymerisation reactions under various conditions. | Optimising one reaction to make a polymer with the physical or mechanical properties wanted. | Scaling up the optimised laboratory process and testing production methods and products made from the polymer. |
| Computing | Studying the properties of general algorithms for handling large volumes of real-time data. | Looking for ways to reduce spam by studying how spam works as a business and what drives spammers. | A start-up uses researchers' code to build a software product for online marketing. |
| Agriculture | Studying how genome changes and mutagenic factors in plants affect their traits. | Searching wild potato genomes for genes that give resistance to blight, to improve crop potatoes. | Building a gene-editing tool based on how enzymes edit DNA. |
| Economics | Testing theories of why regions grow at different rates. | Analysing one region's case to inform government policy. | Building operational models from statistical evidence to design policy tools that help a region catch up. |
| Education | Studying what in the environment shapes the ability to learn. | Comparing national programmes meant to close learning gaps for disadvantaged communities. | Developing and testing in class software and tools that help pupils with special educational needs learn mathematics. |
| Music | Developing a theory that treats musical events as operations on basic musical material. | Reconstructing an ancient instrument from historical records and experimental archaeology. | Creating new teaching materials based on neuroscience findings on how people process new sounds. |
Examples from Frascati Manual 2015, §2.40 (natural sciences and engineering) and §2.41 (social sciences, humanities and the arts), paraphrased.
Basic research with a direction
“No particular application in view” does not mean no direction at all. The manual distinguishes pure basic research, done to advance knowledge without seeking economic or social benefits, from oriented basic research, expected to build a broad base of knowledge for solving recognised or expected problems (§2.28). Its example: research on saving energy with no specific use in view is still basic research, labelled “oriented” because it has a defined direction (§2.27). Basic research is done mostly in higher education, partly in government, and sometimes by firms (§2.27).
Not a ladder
The order basic – applied – development looks like a sequence, but the manual rejects that reading. The relationship between the three types is dynamic, not linear, with feedback between them (§2.38): a problem met in development can send people back to basic questions.
Nor is experimental development the same as product development. Product development is the whole path from an idea to the market; experimental development is only the stage of it in which general knowledge is tested for specific applications, and it counts only if the work meets the R&D criteria (§2.33, §2.34). Designing, building and testing prototypes normally counts as R&D (§2.50).
Why the type matters: Poland
Statistics Poland (GUS) reports R&D spending by type. In 2024 experimental development took 59.0% of Poland's gross domestic expenditure on R&D (PLN 30.4 billion), basic research PLN 15.0 billion and applied research PLN 6.1 billion. More figures are on the Frascati Manual page.
Sources
Checked against the sources below on 29 October 2026.
- OECD (2015), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, OECD Publishing, Paris. doi.org/10.1787/9789264239012-en
- GUS, Urząd Statystyczny w Szczecinie (2025), Działalność badawcza i rozwojowa w Polsce w 2024 r. / Research and experimental development in Poland in 2024, wraz z tablicami. stat.gov.pl/obszary-tematyczne/nauka-i-technika-spoleczenstwo-informacyjne/nauka-i-technika/dzialalnosc-badawcza-i-rozwojowa-w-polsce-w-2024-r-,8,14.html