Spending six months on a failed grant bid is too much. Things must change

Over the years, I have witnessed many talented and ambitious early-career researchers quit the UK or even leave the university sector altogether because they are unwilling to waste time playing what they perceive to be the mug’s game of applying for funding.

I’m still in that game but, increasingly, I do feel like a mug. You put your heart and soul into an application, but success seems to depend more on the quality and identity of competing bids and the available funds in the particular grant round you apply to, rather than on the quality of the bid itself.

For instance, I just spent six months preparing an unsuccessful bid to the UK’s Biotechnology and Biological Sciences Research Council (BBSRC). Although I was not working on it full-time, a significant proportion of my academic life was devoted to a task that ultimately proved to be a waste of time.

But it wasn’t just my time that was wasted. It was also that of all the internal and external partners I worked with, as well as the support staff involved in the submission process.

And then there are all the other unsuccessful PIs. In the feedback, the BBSRC told me that 71 bids to that particular standard/responsive mode round had been “deemed to be of international quality”. It did not say how many were funded, but if we assume it was 30 per cent, that means 50 bids were unsuccessful.

If we conservatively estimate that a single bid requires two months of person-effort, then that’s almost a combined 12 years of work to prepare internationally competitive bids for just a single call, of which more than eight years of effort is completely wasted. That figure might even be much higher – especially if you include all the other rejected bids, too.

UK Research and Innovation has suggested that, across the research councils, a typical standard-mode call receives up to 350 proposals, of which as few as 20 per cent are funded. This means that, on average, five bids must be submitted for one to be successful. That could mean a combined 46 years of wasted academic time per call (or 138 years wasted across the three annual calls) and 10 solid months of individual effort needed per successful bid.

The true figures doubtless lie somewhere in between. But whatever they are, all this inefficiency is not just demoralising for researchers. It is also a waste of taxpayers’ money – which is not great during a university funding crisis. And it undoubtedly renders the UK less competitive.

Worse still, neither the BBSRC nor the Engineering and Physical Sciences Research Council allows resubmissions for standard-mode calls (unless invited, which is extremely uncommon). My rejection email made clear that resubmissions were not permitted “regardless of how competitive an unsuccessful proposal was” (mine was rated as excellent and included two exceptional reviews).

So consortia with commercial organisations that you have spent so much effort researching and building fall apart, and valuable lines of research enquiry and even promising academic careers are stopped dead – unless the bid team can disguise a resubmission as a new and different bid, involving yet more potentially wasted effort.

Even after several decades of experience as an applicant, reviewer and panel member, I don’t pretend that making the system fairer and more efficient will be easy. But I do have a few suggestions.

First, to reduce the waste – for both applicants and reviewers – we should adopt a two-stage bidding process for standard-mode calls. An initial light-touch expression of interest should be followed by a full-bid stage for only around a dozen applicants (the exact number being dependent on the available budget for the given call).

Double-blinding at the initial stage would also help to address any perceived favouritism linked with who the applicant is or where they work – a bias that currently leads to the same names receiving the bulk of the funding. If double-blinding was felt unacceptable, then I would propose increasing transparency by attributing each review to its author.

Fairness could be further boosted by requiring that all bids have the same number of reviews. Currently, some bids only have two, while others have as many as five. This leads to differing levels of assessment rigour as the chances of getting all good reviews diminishes as the number of reviews increases. It only takes one poor review to scupper a bid with otherwise exceptional grades.

That is why reviews themselves should be subject to greater scrutiny. Poorly written ones should be rejected – the positive ones as well as the negative ones. If it is evident that the reviewer can’t be bothered to spend time properly reading the bid, demonstrably does not have the correct expertise, or is unwilling to write a proper review, then their opinion should be discounted. It does not need subject expertise to spot such reviews, and it is almost an insult to the bidding team to have to spend time responding.

Finally, bids rated as excellent or with two or more reviewer grades of “exceptional” should be allowed a single resubmission if they are unsuccessful the first time.

All this would help minimise the impression that the bidding process is a lottery. Perhaps, then, fewer people would decide not even to bother buying a ticket.

Melvyn Smith is professor of machine vision and co-director of the Centre for Machine Vision (CMV) at the University of the West of England, Bristol.