Grant writing tips — a view from the other side
Recently I was reviewing a large stack of proposals for a small travel award. In the process of breaking up this review process, I wrote a very long tweet thread that’s proven popular.
To help people, here’s the list unpacked and with fewer typos:
- Discuss how your grant is going to be value for money, with a view towards leaving a sustained legacy after the proposal has ended. Be reasonable in the legacy, as within this proposal you need to justify how this new money will enable you do exciting and new.
- Write your proposal with the knowledge that your reviewer is an expert, but not an expert in your niche of work. I am a qualified scientist and engineer, but it is unlikely I work on your specific area of research. You can break down barriers through the use of clear and simple language, and reduce your use of acronyms and jargon.
- If you must use acronyms, ensure that they are adding value & not alienating your reviewer. It is especially important that you explain them on first use. This is a rookie error, and it will annoy your reviewer.
- Avoid the use of acronyms and very technical language in your grant title. This is the first thing a reviewer will read, and it’s important to sell your vision to the reviewer quickly.
- Ask a colleague, who perhaps is too familiar with your field, to read your proposal, and to establish whether it makes sense to a “semi-expert”. Your colleagues are busy, so make sure it’s a near to final draft that you share.
- Make sure you answer key questions: Why do this? Why this way? Why you? Why now? Who will benefit? (this is unpacked more completely in my previous post).
- Your proposal is likely going to focus on developing an idea, but people will be involved too (as a resource, in growth of a network, and your collaborators) and so please address how the people will be developed alongside the idea. Many of us care about the people and the science.
- If you want to use metrics to quantify your success, be aware that we all use metrics differently, e.g. the h-index is field dependant, it depends on database, and is tied to how long you have been researching. Also avoid the use of the journal impact factor, as it is bogus for predicting quality of individual work or individual scientists.
- If you are going to list your papers, make them accessible by the reviewer. It is easy to post them as pre- or post- prints, make them open access, make them green accessible, and give links that work in your proposal.
- Update your ORCiD to enable you to have a commercial free CV that is readily linked and easy to search for.
- If you want to qualify your amazing skills, provide underpinning evidence & do not brag. Preferably write your proposal so that the reviewer can come to their own conclusions that you are amazeballs. If you write something that makes you vomit a little, the reviewer (who is only human, and a peer, i.e. like you) will likely want to vomit too. Note that you should also consider systemic gender issues here — there is evidence that men are more self-assured than women.
- Print your proposal, and read it aloud. This will help you spot howlers, like the repetition of particular phrases in a particular manner to spell out a particular point.
- Read the rubric of the proposed call, and ensure that you have answered each technical query. If the proposal has a specific question, make sure you answer it well. If you fail on answering key questions, it is easy to disqualify your proposal and this wastes everyone’s time.
- Reviewers will often be very quick to evaluate proposals, especially if there are a large number to evaluate. I took less than 30 mins a proposal (and it has been suggested by the panel chair that I could take 15 mins for each one!).
- Avoid vacuous words, such as ‘various’ and ‘etc.’. Be specific and ensure that your words add substantive value, name checking specific ideas or technical points to provide context. It is more useful to employ words like ‘such as’ to conjure up specific examples, and highlight that you understand how to break down an idea or a group of concepts. You can address this with a red pen and a printed copy, striking out ‘fluff’ words.
- Be specific and careful with your language. Reviewers will often read many grants side-by-side, and the details matter. If you can illustrate your writing clearly and distil your vision into a short proposal, using tight and specific language, then I have confidence that you have clarity of thought and will deliver this proposal well.
- Be careful to not make your language ‘too tight’ or ‘too technical’, as you do not want to alienate your reviewer and make it hard for them to read. As a reviewer, I should not have to read every sentence several times to understand it, and spend further time to unpick its meaning.
- If you are going to use the ‘brand’ of your University or Department, make sure that you put this into the context of how specific attributes of your institution will add value to your proposed body of work.
- You can strengthen the use of associated brands by inclusion of qualified success or metrics that are important, and to highlight value added to the substance of your proposal. This may include REF scores, graduated students, or even prizes.
- If you are going to add prizes or awards, please ensure that you detail how they are important for your expertise or know-how for this proposal. It is tedious, and annoying, to have to google these during a reading of your proposal.
- In your proposal plan, break down the work into individual work packages or streams. This enables the reviewer to understand the relative difficulty of each task, and to highlight that you have considered the order of delivery and the relative risks involved.
- Read up on data management. This is an increasingly important aspect of modern science, e.g. if you are doing something to exiting data, and producing new information, you are generating new data and this must be managed. (Tips here: https://nsf.gov/eng/general/ENG_DMP_Policy.pdf)
- Research data can be physical data (samples, equipment, film plates) or electronic/digital (what we normally consider ‘data’). This data has to be managed before it is captured, as well as during processing, curation, and for longer term archiving. Consult your funding agency documentation or research institution for further guidance and suggestions.
- If your proposal is going to benefit others, and it last a substantive period of time, please address how the beneficiaries will help develop and evolve the proposal as it progresses.
- As a reviewer, reviewing only part of my job and it is largely unrewarded (more on this in the post script), so please make my life easier. Add hyperlinks where possible, and keep things simple and straight forwarded.
The origin of this advice is because I spent a couple of days undertaking this stack of reviews. The process was tedious, but valuable.
The value added to my grantsmanship, and yours if you’ve enjoyed these tips, is an important proxy benefit of giving back to the community, but note that if you want to be reviewed in a timely fashion, fairly and reasonably, you must sign-up and conduct reviews.
If you worry about how peer review will help your career, consider it part of your continued training to see how people write good (and bad) proposals/papers. Furthermore, these days you can also seek to be recognised for your hard work. For publications, you can easily sign up for a Publons account. For grants, this is a bit more tricky but you can be added to the lists as a panel member for most of the major granting agencies (and highlight this on your CV).
You can also advocate and discuss the value of peer review with others, and encourage and support them to get involved. Speak with your journals about their reviewer process, recommend your peers for reviewer panels, and encourage training and understanding of the value of importance of this process.
If you are a senior academic, ensure you are picking up your fair share. You can encourage your more junior peers to get involved, but don’t over burden them, and please extend an invitation to review with support and understanding that it takes time and it’s difficult for the first few goes.
How many items should you be reviewing?
Let us assume that each grant, or paper, has a 20% acceptance/success rate, and it requires 3 reviewers to consider it. This means that you must be reviewing (at least!) 5–10x as many papers/grants that are accepted/you win. This is a minimum to make the system sustainable and reasonable for everyone.
For these reviews, the quality is as important as the quantity. Please consider how you can review better. Make your comments actionable, reasoned, and enable the panel chair to understand how you have made numerical decisions.
You might want to consider the idea of signing your review reports. It is also important that you remember that humans will read your review, so keep this professional and do not abuse your privileged position.
As a final thought, ultimately it is not sustainable for our communities for individuals to shirk the responsibility of peer review, but temper this with a need to ensure self-care that you are not extending yourself too thin.