Promises, promises, promises: three reasons for grant writing failure

Professor David Wright, Professor of Pharmacy Practice at the University of East Anglia, has offered his thoughts on what makes a good research application – specifically on the issue of ambitious proposals. A common feedback point from the applications we receive has been that applicants promise to deliver more than realistically achievable. Professor Wright delves into this issue and offers insight into how you can avoid following the same path.

The first rule about grantsmanship is ‘don’t write about grantsmanship’. The process is an art rather than a science, relies on luck as much as it does skill and for every poor grant application there will be a poorly functioning award panel.  Whilst you would like to believe that nationally instituted panels operate in a more standardised manner than small charitable grant awarding bodies this also cannot be assumed.  They all get things wrong at different times, due to different reasons but largely because the breadth of knowledge required to cover the world of health services research is frequently not available on most panels and panellists are frequently being asked to compare chalk and cheese when making eventual funding decisions.

The trick with writing grants is just to keep writing them, listen to feedback and when you eventually get over the frustrating elements which can appear within them (e.g. last year I was criticised for not having a real practising nurse on a grant, when clearly I did and someone who was amazing) you have to just accept that the message you had wanted to convey was not sufficiently effective.

There are however some things that you should not do, as these make it too easy for reviewers from preventing you from getting to the point of funding consideration. This is to not over promise.

Common failures in many grants, which make the selection of the reject button easier, are to promise to do too much overall, deliver too much within each phase or deliver too much at the end.

The natural assumption when writing grants as an early career researcher is that grant funding bodies want value for money and that this is achieved by the quantity of work undertaken, the speed at which it is undertaken and amount of outputs promised as a result of it. This, however, cannot be further from the truth. What funding bodies want is reassurance that projects can deliver on time, are more likely to be successful and at the end will provide sufficient high-quality evidence. The national institutes are measured by the number and quality of publications and additional grants, which result from their funding and if you are over promising then you are unlikely to deliver on any of the things they are routinely measured by.

Too much health services research funding has been historically wasted on projects that have ultimately failed. In many cases significant additional resource has been provided in an attempt to save the project and in providing this, as well as significant additional administrative resources.  The regular finger burning has made the traditional funding bodies far more risk adverse and unwilling to fund anything that makes unsupported or unsubstantiated promises. Below, I consider the three promises that require avoidance and how this can be achieved.

Promising too much

The NIHR believes that a systematic review costs between £100k and 150k, a feasibility study £250k and a randomised controlled study anything from £500k to £5M. The amount it is willing to spend is decided by the ultimate value of the perceived impact – precise words for a very imprecise concept.

In charitable terms however, where overheads or full economic costing are not covered, these numbers can be halved. £50k to £75k systematic review, £125k for a feasibility study and £250k for a definitive study. If you have a high quality PhD student who is already funded and are looking for additional funding to deliver on one of these projects, then the ‘student’ is worth £35k per year in charitable funding. Consequently, asking for £20k to £25k to undertake a systematic review may be reasonable as this would pay for a second reviewer, publication and conference costs and could all be done within one year. A feasibility study may cost a bit more but must be presented as a feasibility study and not a thinly veiled attempt to demonstrate that the intervention is effective. Mixing both a systematic review and feasibility study for a PhD student over 2 years is going to require at least £60k and may still be overambitious.

It however would never be appropriate to offer a charitable foundation to deliver a high quality definitive randomised controlled study using your own PhD student for anything less than £200k. You should also never do this if you haven’t already conducted the systematic review or feasibility study first.

The best way to determine whether you are promising too much is to ensure that you have given yourself sufficient time for each element of the process and without making it obvious you have sufficient latitude to allow for unforeseen problems. This is why funding bodies frequently require a Gantt chart, as this helps reviewers to visualise your plans and therefore their feasibility or ambitiousness. An experienced researcher i.e. someone who has regularly delivered significant grant applications awarded competitively will be able to provide advice with respect to the aspirations within your grant application.

Promising too fast

A red rag to all reviewers is an over ambitious recruitment rate. A major reason for study failure has been failure to recruit and/or failure to accurately estimate attrition rates. Consequently, we see the now standard requirement for a feasibility study prior to any form of definitive study.  Reassurance that a definitive study is likely to recruit and retain in order to meet the required sample size is an absolute must. Consequently, promising to recruit too quickly without evidence to support the assumption is going to result in immediate rejection.

Feasibility studies are however an exception to this. The point of these is to estimate these parameters. If you have no prior experience, then it might be reasonable to assume that you will recruit no more than 1 patient per week at each site. Much less than this is unlikely to make future definitive trial inviable and much more than this you could be being over ambitious.

Promising the world

It is all too easy at the end of writing a grant application to promise numerous publications and extensive impact. This can often be a good indicator of the inexperience of the applicant. The impact from a feasibility study should be negligible as without the definitive trial you won’t have the evidence to promote the intervention. The best you can promise is a publication and a definitive trial grant application.

Systematic reviews are increasingly used to inform intervention design and again societal impact would be expected to the limited. If, however you intend to use them to develop guidelines, you need a detailed plan for implementation using recognised approaches and these have to be appropriately costed. Suggesting that you will present your results at appropriate conferences to encourage adoption is not implementation and adds limited value.

The rule with outputs is to be honest with yourself and consider what you can really deliver and what would be most meaningful.


So, in summary, if you want your grant to have any chance of being considered for funding and the vagaries associated with that, then you first have to focus on reassuring the funder that you will be able to deliver on time, on budget, with sufficient quality and that you will maximise the new knowledge that you will generate from it. You definitely don’t increase your funding chances by promising too much, to deliver too fast or the world.

No comments yet.

Leave a Reply