I am post-doc from biological sciences writing a grant proposal. One of the sections of the application is dedicated to risk assessment of the given project. There were no further instructions.

I can think of very general risks and their solutions. For example:

  1. Experiment failed. → Troubleshoot; try alternative method.
  2. Manuscript publication delayed. → Get help from supervisor or grants office.
  3. Insignificant results. → Discuss with experts; try again!

Is this the kind of risk assessment funding agencies (like NIH, Welcome Trust, or EU) are looking for? Can you suggest some extra points (research or managerial) which I can consider here?

  • The meaning of "Risk Assessment" can vary considerably by location. In Australia, they are part of a formal safety management system. Aug 31, 2019 at 6:30

2 Answers 2


There are two classes of risks that are typically considered in grant-writing:

  1. Programmatic risks that the money will be wasted because something doesn't go as you hoped.
  2. Safety and ethics risks based on the specific contents on your research.

The second is generally covered under a checklist of declarations regarding specific topics (e.g., "human subjects research," "dual-use agents of concern," "vertebrate animal research"), rather than general "risks", so I'm going to write my answer addressing the first class.

With regards to programmatic risks, the big issue here is research always involves the unexpected, most often emerging in the form of things not working as well as you thought they would. Progress might be slow, effects might be small, vital supplies might become unavailable, things just might not work, etc.

What grant reviewers are generally looking for here, then, is evidence of clear thinking about how to handle setbacks of this sort. You really don't want to give money to somebody with no "plan B", and then 3 months into the project something goes wrong and they end up just frittering away the rest of the money banging their head against the same wall.

How I recommend assessing risks, then, is to sit down with a your project plan and asking yourself the following questions:

  • For each step in the plan, what materials/people/facilities/information is needed before I can begin the step? What happens if they aren't available as expected?
  • For each step in the plan, what does "failure" look like? How many different ways can it fail?
  • Which of these problems are most likely to endanger the overall goals of the proposal?
  • For all of the "important" problems, what is the plan B for how to ensure money is spent in a worthwhile manner if they are encountered?

Doing this assessment may also cause you to reorganize your plan: sponsors often like to see the most "critical path" or high-risk work up front, so that the most significant and productive problems can be encountered sooner rather than later (and the project terminated if necessary).

  • 2
    In EU grant writing often detailed abort criteria (e.g. figure of meritc) at midtime of a project plan are necessary (for 3-5 years projects) as well as distinct sections where you have to argue what reviewers could possibly critisize in your proposal. A plan B is always nice to have, but in fundamental research projects often not possible I fear developing/evaluating a new method/hypothesis. Then it's even more important to think in small steps as you described through the virtual project and allocate personnel and money appropriately so a final report describes exactly why it didn't work. Aug 31, 2019 at 14:52

I would suggest you to use the risk-assessment methodology of the PMI (Project Management Institute): Lavanya, N. & Malarvizhi, T. (2008). Risk analysis and management: a vital key to effective project management.

It's widely accepted in business directly and in academia as reference since it's a paper and proven methodology. More importantly, there are various other tools you can use to organize, estimate, and present the risk, and it has all the nice graphics and tables that academics love to see any proposal.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .