For me, one of the most useful insights was the issue of 'introducers' rejections'. As I'm sure you know, when you submit an application to the ESRC it doesn't go straight to panel. It goes through a sifting process, including:
- Office sift: roughly 10% of applications get rejected at this stage on technicalities, such as not having the right attachments, sections not being filled, format not being adhered to, etc;
- Reviewers' sift: roughly 30% get rejected at this stage. If the reviewers identify substantial flaws, and grade the applications accordingly;
- Introducers' sift: I think this has been in place for some time, but I hadn't realised the scale of it before now. Each application is allocated to two introducers, who will have the responsibility of introducing the application to the panel. However, they can reject applications before they get to panel if they think that, realistically, they don't stand any real chance of getting funding, and it would be wasting the panel's time to discuss them.
The ESRC said that, after these sifts, they would only expect 30% of applications to go to panel. Given this figure, it looks like the introducers are expected to strip out 30% of the applications. That's quite a substantial figure.
In practice I imagine that this is fairly straightforward. There will be obvious applications that don't have anything wrong with them, but are never going to fly. However, it does make me worry slightly that the responsibility for identifying these is bestowed on so few people. Given how 'political' sub disciplines can be, what happens if your application is sifted out be someone who disagrees with your work, rather than allowing a wider range of views to input?
Anyway, the lesson to take away from this is to look at the grants panel membership (pdf) and try and identify the two people who are likely to be the introducers. Do a bit of background research on them and their interests, and try to key your proposal in with what makes them tick. They are the keepers of the keys to the kingdom - or at least to the grants panel.