“Each year, there is over $26 billion in investment by the Australian government in research and development. Less than 1% results in a direct commercial outcome. If research funders were more like venture capitalists and treated knowledge as a product, how could this help lift research impact?”
Research organisations rely on various forms of government and industry-managed funds to support their research activities. Typically, there is an announcement or research call with a problem or area to which the funder will commit resources. More applied research applications must increasingly articulate how to deliver on-ground benefits or impacts that address a problem.
Researchers will align their unique and core research capabilities with the problem statement. They will develop a research plan and budget to deliver insights and/or technologies that can be used to address the core problem. A typical research plan will include a section that addresses a plan for 'impact', 'engagement', 'commercialisation' and/or 'extension'. The trick is to ensure the application demonstrates an achievable deliverable that can be measured in a final report.
An investor in a start-up approaches on-ground benefits and impact through the lens of how an insight and/or technology can solve a problem. The investor recognises uncertainty and wants the start-up to quickly remove these uncertainties and find product market fit where the benefits are clear and have demonstrated value for a group of people or market segment. Crucial to investment is presenting market feedback based on market testing or research.
Using strategies to ensure research investments deliver knowledge-based insights and are captured in computer code that can connect with users might help researchers and research funders get 'market' feedback. This 'market' feedback will help drive a greater return on investment from research funding. Here are some tips on how funders can help drive researchers to improve the return on investment from research investments.
Many tools are available to help researchers share access to their code. They don't have to write an app or even hand over source code. Cloud computing infrastructure provides a framework where researchers can allow users to see the results of their data analysis or modelling framework.
When researchers are encouraged to make a direct, living connection with an end-user, this helps them better understand the challenges and opportunities of their research insights.
Artificial is creating a big buzz and no where ore so than for researchers. Large automated sensor sed systems are driving large complex data sets that are ripe for machine learning. This hype is manifested in reseaerch funding applications.
Machine learning and AI tools derive statistical insights. While it may be possible to derive some general underlying principles, the real opportunity is to take complex, high-volume data and create black box models.
If researchers are promising black box models the only benefit is making sure others can pass data in and derive insights. They are not deriving knowledge that can be shared so make sure the researcher has a clear path to share access to the model.
Final reports are a cop-out; they deliver an output, but it's not very useful.
Code-based insights, if done right, provide an opportunity for users to interact with the research insights. The interaction might just be sharing data but could also provide a practical demonstration of the analysis. Ideally, computer code can deliver valuable insights that have practical value.
Technology Readiness Level or TRL is used to assess how mature a technology is. It uses standard descriptions linked to a scale from 1 to 9 that assess the relative maturity of a given technology. It is typically used to assess the commercial readiness of any given technology. More recently, it has been used to assess workflows for research-derived machine learning.
To help researchers develop more impactful research resulting from computer code based tools they can use the TRL framework and roadmap to help identify both good work practices and insights developed from computer code. Supporting the development of practical tools and tips that facilitate impactful work practices that can connect a broad range of insights to a broad range of endpoints will help drive adoption.
Changing the perception and realising the value of code-based insights requires fund administrators to be aware of the opportunities.
Administrative staff don't need to become software engineers, but having familiarity with the process of realising value from research code insights can help develop impact strategies. The idea that all research code needs to be wrapped in a mobile app and shared is no longer true. There are lots of paths to drive impact from research code.
