Twenty tips for writing a research proposal

4 05 2015

Proposal FormatThis post’s title might promise a lot, but it would be disingenuous of me to imply that I could cover all of the essential components of this massive topic in one blog post. Many people (my wife included) have made careers out of teaching people how to write successful grant proposals, so I won’t pretend to be comprehensive and insult their expertise. That said, I’ve been reasonably successful on the grants’ side of the science game, and I’ve assessed a fair few grant proposals in my day, so I think I can offer at least a few pointers. As usual, each person probably has her or his own way of doing things, so there’s unlikely to be a single, winning method. Approaches will also vary by funding agency and country of origin. I am therefore targeting the earlier-career people who have yet to get fully indoctrinated into the funding cycle, with generalities that should apply to most grant proposals.

1. A proposal is not an article, so don’t try to write it as one.

In the huge list of things ‘they never taught you as a student, but need to know to be a successful scientist’, this has got to be one of the biggies. Now I’m mainly talking about science here, but grant proposals cannot and should not follow the standard format of peer-reviewed articles. Articles tend to put an elaborate background up front, a complex description of hypotheses followed by an even more complex description of methods and results. Do not do this for a proposal. A proposal should be viewed more as a ‘pitch’ that hooks the assessor’s attention from the get-go. More on this aspect below.

2. Understand what the funder actually funds

Many neophytes to the funding cycle have the mistaken perception that funding agencies exist to fund the sort of research that the researcher want to do. Sorry – nah! A funding agency exists to fund the type of research that the funding agency wants to see happen. It doesn’t matter if it’s the Australian Research Council, the Gates Foundation, the Wellcome Trust, the National Science Foundation, or the Slovenian Science Foundation – they all have (often government-set) funding priorities. Know these and follow them. If you must alter your research desires to fit within these frameworks, so be it. Don’t be offended that no one wants to fund what you want to do.

3. Read the guidelines, and follow them to the letter.

It might seem self-evident, but few people actually read these before launching into the draft proposal. Avoid this mistake. Spend days, if not weeks, poring over ever single element of the guidelines that will tell you more than just the margin or font size they expect. There are many, subtle hints in the guidelines that tell the applicant what and how to write. More importantly, they are usually fairly explicit about what you shouldn’t include.

4. Assume the assessor has no knowledge whatsoever of your field, because most of them will not.

If you write a proposal to your immediate supervisor, closest colleague or a specialist reviewer, they will have an implied background against which they can judge the merits of your work. Most assessors will not have this (or they will for a limited field only), so you have to be (almost) patronising about explaining yourself in the proposal. Do not assume any existing knowledge, explain all your terminology (see more below), and tell them repeatedly why the subject is important and compelling.

5. You’re selling yourself (and your team) as much, if not more, than the research project you are proposing. Track record, track record, track record.

A pessimist might justifiably conclude that as long as the proposed research isn’t flawed and fits the funding model in question, many agencies don’t give a rat’s hairy bollocks about the subject matter per se. Instead, they focus mainly on the reputation and track record of the person(s) proposing the research. An early-career researcher could then (again, justifiably) conclude that the system is stacked against them because if they don’t yet have a great track record, what chance would they ever have of getting funded in the first place? It’s a truism in science as pretty much everywhere else in society that the rich get richer, but you can avert this to some extent by careful alignment with more established colleagues (see below) and an emphasis on what makes you different.

The take-home message of this point is that you matter so very much in a proposal that you must sell yourself. After all, a research grant is an award, so think of the proposal as an application for a prize. Make your assessor think that you are the greatest thing since E=mc2, and focus on your career highlights and contributions to date. The funder needs to know that they will be funding a person whose proven reputation, skills and output will guarantee the return on their investment (i.e., the success of your proposed research).

6. Never underestimate the value of a good team of collaborators.

This now might seem obvious from the previous point, but it’s not merely a recommendation to latch onto the best person in your field to maximise your success vicariously. A careful selection of the key people to fill any weaknesses in your expertise, reputation and knowledge will make a huge difference to how your team’s capability and the project’s feasibility are assessed. On that note, it’s equally important to exclude weak collaborators who, at least from the perspective of the assessor, bring little to the team’s expertise or reputation.

7. Never underestimate the value of a good title.

Remember that point about ‘selling’ above? Just like in product marketing, a catchy title and a clever opening will potentially get you the attention you need to stand out among the hundreds or thousands of other researchers vying for the same pot of gold. The title should (i) avoid questions (it’s not going to make people inherently more curious – after all, you won’t have the results yet to answer the question), (ii) immediately understood by all and sundry, (iii) be short and to the point and (iv) include something to tell us all why it’s so bloody important.

8. You’ve already won or lost the game in the first page.

Including the title, an assessor will at the very least get bored, or abandon any further assessment at worst, if you have not explained on the very first page (i) what you are going to do (see below), (ii) why it’s so exciting and (iii) what major societal problem it will solve (more on applied versus theoretical subjects below). You have to keep the assessor’s attention, so provide some tantalising information up front that will spur them on to more reading.

9. Please tell us immediately what you are intending to research.

I’ve singled out this element of the above point because it is the most important. If you prattle on for pages setting up the background and theoretical construct of the problem you propose to address before the assessor has any idea what you are actually proposing to do, you’ve lost all hope of convincing anyone that it’s important and do-able. An assessor wants to see immediately that you’ll be proposing to do xy and such that you can answer big questions ab and efficiently, effectively and convincingly.

10. Tell us why the research is exciting.

You might think it is, so might your partner and your grandmother, but does anyone else? Do not take it as given that your chosen topic (Point 2 notwithstanding) is of any interest whatsoever to anyone else. You have to explain, in gory detail, what makes it so bloody fascinating and essential that you do the research now. As an assessor, I want to end up being as excited by it as you, so sell yourself. If you’re not inherently a good salesperson, you’ll have to become one.

11. Explain the applied outcomes of the research, if there are any.

I’m not going to enter into the debate about the relative merits of so-called blue-skies versus applied research (I think they’re both necessary – it’s the ratio that’s up for debate), but chances are that if you have followed Point 2 there will have to be some application to your work. In other words, why does society need to invest in your research if nothing practical will result? Spend a good deal of time explaining how your results will affect the real world, either through policy, technology or remediation, and never, ever state that it will simply provide humanity with more ‘knowledge’. Please.

12. Funding agencies are generally risk-averse, so make sure that you (and/or your team) have some history in the area of the proposed research.

Coming back to Point 5, you must understand that most granting agencies aren’t willing to take a punt on your potential as a researcher, no matter how wonderful you truly are; instead, they want a guarantee that you’ll be able to do what you say you will do. Remember, you are competing with many others for a paltry sum. Proposing a difficult, elaborate and risky research project will only lead to disappointment. It might sounds a little jaded on my part, but it’s true to some extent that funding agencies only fund what’s already been proven to work. Sadly, if your project is too innovative (i.e., ‘risky’ seen through the eyes of the assessor), it’s unlikely you’ll receive funding. A working rule of thumb is that if you have some established track record in the area of proposed research (e.g., a previously published paper in the subject), then you have a much better chance of success than proposing something you’ve never done before.

13. Hypotheses! State them.

It’s worth repeating that hypotheses are testable assertions and not merely aims. It’s one thing to aim to solve the energy crisis, it’s quite another to say how you will do it. Be careful to list your main hypotheses and their predictions, and how you will test them with data/models, etc. If you do not propose testable hypotheses, your risk of failure (as deemed by the assessor) increases, and so too does your probability of not being funded.

14. At risk of sounding like a broken record, avoid jargon as much as possible.

Jargon is for specialists and in my opinion, should be avoided at all costs. It is even more important to avoid jargon of all types (and I include abbreviations, initialisms and acronyms in this list) in a research proposal. If the assessor does not know immediately what you mean, even after defining a term, you will lose her/his attention. Be clear, and even if it requires more words, explain everything simply.

15. Avoid motherhood statements and subjective qualifiers. Quantify where possible.

A motherhood statement is defined as a vague, ‘feel-good’ platitude with which few would inherently disagree. In science, it’s usually associated with some outcome (e.g., ‘we must preserve species’). Why ‘must’ we do something? Do not assume that your assessor has the same values as you, or that the funding agency upholds the same morality. Likewise, avoid subjective qualifiers like ‘a lot’, ‘a multitude’, ‘very’ and ‘significant’ if they have no quantifiable meaning. Make sure you demonstrate that the research will quantify a phenomenon or process and that you do not inadvertently demonstrate your biases or lack of understanding by using such subjective terminology.

16. Be methodologically specific.

Some people try to hide the dust under the rug in a proposal by including throw-away lines regarding how they will achieve their objectives. One in particular that I see far too often is “… and then we will model the system’. How will you model system? What model will you use? How will you parameterise it? Do you have the necessary expertise in your team to construct such a model? Likewise, ‘… we will measure …’ and ‘… we will construct …’ statements without the corresponding methodological detail (exactly how) are a clear demonstration to the assessor that you don’t know what you’re doing. If you don’t, make damn sure that you do by including someone who does and then describing it in detail.

17. Be realistic.

Proposals can often verge on the fantastical because the proponents purport to solve the mysteries of life, the universe and everything in 10 or fewer pages. Without lessening the impact of why the research is important, don’t venture too far into Faerieland and claim that you will be able to solve all elements of the problem under investigation. Stick to the hypotheses and do what you can within the budget and timeline proposed.

18. Give some serious attention to your communication strategy.

Many funding agencies (all?) want to know how you’ll make the results of your research known to the public and not just to them or the few specialists who might actually read the resulting scientific articles. Communication is becoming more and more important these days as society in general becomes more and more adverse to scientific endeavour. While social media might not be everyone’s cup of tea, spend a little more time explaining how you’ll reach a much broader range of people than most researchers achieve. Think of clever and innovative ways of reaching out, and dedicate more than a passing thought to this section of the proposal.

19. Have an experienced colleague read it. Better yet, have two or three of them read it.

After it’s all said and done, give it to someone with more experience than you to read and critique it fully. In many ways, this is the most important part of the process before you even submit. More opinions are better than one.

20. Ask whether your father/mother/auntie/best mate, as a taxpayer, would fund your research.

Most research these days is publicly funded, so asking a few lay taxpayers that you can arm-wrestle to read it will tell you if it strikes a chord, or bores them shitless. If you can’t convince the punter in the street, it’s less likely that the funding agency will deem your research worthwhile.

CJA Bradshaw



11 responses

13 05 2022
A few insights into the inner workings of the Australian Research Council |

[…] think it’s essential to understand not only how to maximise your probability of success (see this post for some generic tips), but also how your grant is treated once you submit it. I’ve heard […]


10 04 2018
How to Write a Successful Grant Application - Blogademix

[…] will be expected to use jargon specific to their area. If the research proposal forms part of a grant application, authors would have to write it so that reviewers who might not be as familiar with their specialty […]


18 03 2016
Writing successful grant applications: power in numbers | The Applied Ecologist's blog

[…] the session have been captured by others: interesting links include this highly retweeted post by Conservationbytes, this post by the Guardian, and the tips shared by the […]


10 12 2015
Science is beautiful |

[…] grind (bad reviews, profiteering of academic publishers, shitty university administration, the constant pressure to beg for money, poor pay, feelings of futility, et cetera ad nauseam), there’s nothing quite as comforting […]


2 10 2015
The sticky subject of article authorship |

[…] Scientists spend a huge amount of time begging (errhm, … requesting … ) funding to do their research, which inter alia requires original thought, […]


10 09 2015
Scientist offers 20 tips for writing a research proposal | OARS Research News

[…] pleased to reblog this Conservation Bytes post by CJA Bradshaw. […]


10 08 2015
Joanna Candler

Reblogged this on jocand0.


13 05 2015

Great tips! Especially about asking your father/mother/auntie/best mate, as a taxpayer :) If you are looking for proposal templates, I suggest to take a look at a Proposal Toolkit:


13 05 2015

Great tips! Especially about asking your father/mother/auntie/best mate :) If you are looking for proposal templates, I suggest to take a look at a Proposal Toolkit:


12 05 2015
Dbytes #195 (12 May 2015) | Dbytes

[…] Recommended by anna Renwick [Editor’s tip: If your proposal includes lists, make sure they contain no more than 5-7 tips.] […]


8 05 2015
Frontiers Academy

Reblogged this on Frontiers Academy .


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: