Every once in a while I get an email from someone preparing for an ERC Starting Grant interview asking for advice. I’ve been sending more or less the same reply each time, so I figured I’d just write it up here and point people to it. This is specifically about the Starting Grant, which is what I have experience with, but I imagine much of it generalizes.
Before I get into it, let me be upfront: there are proper courses and resources for this, and you should use them. For example, the FFG in Austria runs preparation workshops, EU-LIFE organizes courses, and virtually every European national funding agency will provide some form of support, as it is in their interest to have their researchers succeed. There are also blog posts and guides out there from people who have gotten multiple ERCs, so they probably know better than me. Still, here are my two cents to contribute to the conversation.
My experience: two attempts
My path to an ERC Starting Grant took two tries, and I think both attempts taught me something worth sharing.
For the first attempt in 2023, I started preparing about six months before the deadline, around March. I pitched the basic idea to our faculty in April, worked on preliminary data and high-level writing through the summer, and finished the B1 and B2 parts by early September when I sent drafts to colleagues for feedback (thank you in particular Christoph Bock, and Giulio Superti-Furga). Then it was polishing until the deadline. As it happened, in 2023 there was a roughly two-week extension due to the war in Palestine/Israel, and despite the precious extra time I ended up doing some useful procrastinating: obsessing over figures instead of tightening the text, which still needed work. Immediately after submitting I noticed three or four typos and one unfinished sentence. Classic.
Still, the first attempt was a genuinely good experience. I was 100% invested in the ideas, and it felt great to pour out my full scientific creativity into a single document. For the interview, I rehearsed with our faculty group mimicking the ERC panel about a month before, and I practiced the presentation over 40 times in a timed manner (seven minutes only). I wrote down around 50 likely possible questions with answers and practiced leading with the punchline, reserving additional detail only if a panelist pushed further. I generally don’t prepare for talks. I never rehearse, I just freestyle. But I knew that for this, anything could count. In the end I was proud I could break out of my usual habit and show myself that I can be studious when it matters. The delivery was, I think, flawless.
Which is why I was desolated when I didn’t get it.
The feedback from the seven reviewers, though, was excellent and very much on point. They all identified my Aim 2 as too unfeasible (and from our preliminary work since then, they were right - sad face emoji) and found the grant overall too grandiose (quasi-pun intended). Fair enough.
For my second attempt the following year, I was far less motivated. I was pessimistic, I procrastinated a lot. But I forced myself to focus on the reviewer feedback: I swapped out Aim 2 entirely and toned down the overall scale of the proposal. For the interview, I didn’t finish the presentation until the day before, and I only rehearsed on the day itself (still about six to eight times though since my interview was scheduled for 17:45 - this whole day was painful). The interview went okay, but I got a couple odd questions from one member of the panel, so I wasn’t particularly confident ‘walking out’. And yet the second time was the charm. I got it!
The lesson I take from this is not that preparation doesn’t matter (it obviously does) but that the content and realism of your proposal matter more than a perfectly polished delivery. My first attempt was better rehearsed by an order of magnitude, but the science was less tight. My second attempt was scrappier in presentation but sharper in substance. The panel noticed. Or did it? Who knows.
General advice
Start early if you can, but know that starting early mostly buys you time for feedback loops. The single most valuable thing I did was pitch the idea and send drafts to trusted colleagues well before the deadline. Fresh eyes will catch the things you’ve been staring past for weeks. And on that note, don’t be afraid to reach out to people, whether they hold ERC grants or not, for advice. Not every institutional environment is the same, and it may be that no one around you has been through this process. Cold emails to past recipients can work too (panel member lists and grant results are publicly available), but do your homework first. People are generally happy to help if you show you’ve already put in the effort.
For the written proposal, focus on the conceptual advance. Technology and methods matter, but the panel wants to fund ideas that shift how we think about a problem, not just new ways to measure things. If your proposal has a known limitation (and it will) it’s better to acknowledge it honestly and explain your mitigation strategy than to hope no one notices.
For the presentation, I’d say really commit to the big picture. You have seven minutes, which is both an eternity and nothing at all. Script it, rehearse it, time it, and then rehearse it some more forgetting about the script until it sounds natural. I know this is painful for those of us who prefer to wing it, but the stakes are high enough to justify the discomfort - and who knows, maybe you’ll enjoy it. Practice with someone who can also throw questions at you afterward, especially if they are not in your field. The transition from presentation to Q&A is can really get you if you’re not ready.
For the questions, I found it helpful to write out my top 30 to 50 most likely questions and draft answers aimed at 30 seconds to two minutes each. Lead with the punchline. If the panelist wants more, they’ll ask. You can look up the panel composition from previous years on the ERC website, and typically many (most?) members repeat every couple of years. That said, I wouldn’t recommend focusing too much on who’s asking. You risk second-guessing the panelist instead of just answering the question. Keep it simple, keep it factual, and above all, show genuine enthusiasm for your own research. If you’re not visibly excited about your ideas, why should they be?
Finally, if you don’t get it on the first try, don’t despair. Read the feedback carefully, give it some rest, and re-read again down the line. Take the feedback seriously - the reviewers are generally trying to help, even when it stings. And then try again.
Good luck!