Five lazy tips for grant writing

I've been meaning to write a comprehensive guide to grant writing but it's a tall ask. Here's a lazy "5 tips for better grant writing".

Do not bullshit

Just don't. We can tell. Anyone who has been involved in a research project (so anyone who has 2+ years research experience) understands what makes the essence of a research project, and while they might not be able to explain what is wrong with a proposal they will know when something is amiss or incomplete. Regardless of how many exotic techniques and marvellous machines at your disposal, you must answer the fundamental question of feasibility and purpose. And we know what a feasible research project looks like. We need to know the research aim or question, how data will be collected, how it will be interpreted, and how will you know that your interpretation is not an artefact of your method. People often fail to explain the last part. Ideally, you will use multiple approaches to cross check your findings. As long as this is assured to us, as the reviewer, we do not need to fully understand the specifics around the method used.

Follow the brief

The easiest way to dismiss a proposal is to determine it is out of scope. There is usually a clear specification provided by the funding body and reviewers can lean on that if needed. In a decision where two proposals are equally good, it can be difficult to argue that one is more important or more likely to succeed than the other. It is easier to prefer one that is clearly a better fit for the scope of the call.

So read the description of the call very closely. Try to read between the lines, there is usually a single ideal project they really want to see and a spectrum of other projects they will entertain. For instance, they might write in the initial blurb that they will fund projects “with a clear translational impact but will also welcome proposals focusing on fundamental science.” These mixed messages are common. If we submit a project that leans more toward fundamental science, how welcoming would they be? In this case, I would have a look at the guidance of what to include in proposals and how proposals will be assessed. It is not a good sign if the guidance has lengthy explanations of what they consider societal impact with detailed examples. This does not mean that we shouldn't submit the proposal, but the idea had better be a good one.

Clearly signal your intent and link different sections together

Research proposals can be complicated and difficult to follow. While reading about details of what work will be done, it's easy to lose track of the original motivation. Your proposal should include a concise, straightforward statement that captures the essence of what the project is about, serving as a quick reference point for reviewers. Opinions vary whether this should be the first lines of the proposal, or whether it should be after some background information.

The overall aim will be broken down into objectives, which have a series of tasks that will meet those objectives. There are many ways to make it clear how these interconnect. It can be as easy as good headings/subheadings, or a diagram. One implicit trick is to have the same number of objectives as the number of tasks, and make it clear that objective 1 is served by task 1 and so on. Three is generally a good number of objectives and tasks.

The proposed work should follow a clear trajectory. For instance, the initial two objectives could focus on characterising a mechanism in the laboratory, leading to the final objective of investigating its broader applicability in natural settings. Although, this does raise the question of risk migration - if objective 2 depends entirely on results from objective 1, what will you do if objective 1 fails? - and do not say it won't fail because you are just that good. The trajectory does not have to be a single line, it can diverge or run in parallel; you may establish at the beginning that there are three major facets of a particular problem and understanding each of those facets becomes an objective. Anyway, workflows and diagrams can help if the words escape you.

Give equal weight to different tasks

You can tell a lot about someone's background by which tasks they emphasise in the proposal. Every task has to be fairly explained, in terms of the work and how they relate to achieving the objectives. An applicant with mainly wet-lab experience in a proposal that has a laboratory and a computational component may carefully explain the lab work, but then brush over the computational tasks. Likewise, a data analyst could agonise over the analytics without giving much information about the initial sampling. Even if I do not know the subject area very well, when I see this, I immediately know which part of the projects are weak - the lack of clarity on paper reveals the lack of clarity in the mind of the applicant. Maybe that's mean but it is tough grifting in the marketplace of ideas.

Why you? Why now?

The world is full of problems to solve. Simply stating that there is a problem is not enough. Too often I read proposals where there is no problem stated but simply there is something we do not understand. Why is your problem the one we must tackle, and why right now?

Once it is clear that the problem is indeed timely and important, why are you (and your people/research environment) the ones who should do the work? Are you clearly capable? It's on you to prove it. It can be difficult to do given the limited amount of space. Some ingenuity is required. Try to make your words serve two purposes throughout, for instance, when presenting preliminary data, this does not only explain the current state of research but if you have been working in this area, it's an opportunity to highlight your expertise. People tend to write passive statements like, “It has been previously shown that phenomenon A is linked to condition Y”, with a citation. The sentence could do a lot more, that is, if it's your work, then tell us! “Using the X facility here, we have shown phenomenon A is linked to condition Y”.

Be single minded about the problem at hand, even bloody minded. Everything written should be in service to the singular aim of the project. The aspects of the expertise you mention, the preliminary work, and the facilities around you should contribute to the success of this project. There should almost be a feeling of compulsion in the proposal. It should feel as if you care. It's infectious. And if you don't care, why should I?

My checklist

Here's my checklist of things you need to ensure before submission. If you can agree strongly with each point, you're in good shape.

• I care about the problem and the research work outlined to address it.
• I can state the overarching aim as a short freestanding statement (in maybe 40 words).
• The outcome of this project is more than "adding to human knowledge" by "revealing something we don't know".
• Each work package and task are fairly and equally described.
• The overall singular aim is broken into objectives which are achieved by a series of tasks.
• I have shown why I (and my organisation) are capable of delivering this work.
• I have mentioned a clear contingency for the single biggest risk in the project.
• I have not made a statement where I acknowledged a risk and then immediately dismissed it.
• I have only mentioned the aspects of my organisation/work environment that specifically support the work described.
• It is abundantly clear which objective(s) each task serve. Or that the task is a vital preparatory step towards the objective(s).
• My costings are realistic and clearly justified. I have shown colleague(s) who have done similar work and they say these numbers are reasonable.

Good hunting.

Questions or comments? @ me on Mastodon @happykhan@mstdn.science or Twitter @happy_khan

The banner image is an AI generated picture (Midjourney) with prompt; 'A lightbulb in the style of Wassily Kandinsky'. You can share and adapt this image following a CC BY-SA 4.0 licence