Your Proposal Is Answering the Wrong Question

A research director I once worked with had a proposal rejected twice before she called for help. Both times, the written feedback was thin and polite - "does not sufficiently align with current priorities," "insufficient differentiation from existing portfolio." Both times, she was furious, because the proposal was, by any scientific measure, excellent. Novel hypothesis, rigorous methodology, a genuinely strong preliminary dataset. She had written the best case a scientist could make for why the work was true and interesting.

Nobody on the funding panel was actually asking whether the work was true and interesting. They were asking something else entirely, and the proposal had never once addressed it.

Pitching investors rather than a funding panel? See our guide to why a pitch deck is an argument, not a story.

Two different reviewers, two different questions

Most researchers write a proposal the way they'd write a journal submission: build the case for scientific merit, establish novelty, demonstrate methodological rigor, cite the gap in the literature. That instinct is not wrong. It is simply answering the wrong reviewer's question, because the person with actual authority to fund or kill a corporate or cross-institutional proposal is rarely evaluating scientific merit as a standalone criterion. They are doing something closer to portfolio risk allocation, comparing this specific ask against every other proposal competing for the same finite capital, asking what happens to the organization if the core assumption turns out to be wrong, and checking whether the work fits a set of strategic bets that were decided somewhere else, in a meeting the applicant never saw.

A proposal can clear every bar a peer reviewer would set and still fail completely at this second, unstated evaluation, because scientific merit was never the actual gate. It's necessary. It's not sufficient. The panel reading for portfolio fit is asking: if we fund ten proposals like this and three fail outright, does the organization survive that, and does this one specific project meaningfully change our position if it succeeds? A proposal that never engages with that question - that simply asserts its own excellence and assumes funding will follow - reads, to that specific reader, like an answer to a question nobody in the room was actually asking.

Where the written case actually breaks down

This mismatch shows up in a specific, recognizable pattern once you know to look for it. The proposal spends its strongest pages establishing novelty and rigor, and then treats risk, cost, and strategic fit as an afterthought - a single paragraph near the end, if it appears at all, gesturing vaguely at "potential applications" without ever pricing out what failure actually costs the funder or naming which existing initiative this displaces.

Compare that to a proposal built around the funder's actual evaluation criteria from the first page. It states, early and explicitly, what happens if the central hypothesis doesn't hold, not defensively, but as evidence that the applicant has already done the risk analysis the panel would otherwise have to do themselves. It names, directly, which existing line of work this either extends or replaces, rather than leaving the panel to guess whether they're being asked to fund something entirely new or something that duplicates a bet they've already made. It quantifies the downside as precisely as it quantifies the potential upside, because a panel doing risk allocation needs both numbers to make the comparison the proposal is actually being judged against.

This is not a matter of adding a "risks" section as a formality. It's a structural reordering of what the proposal is actually arguing. A scientifically excellent case for novelty, without an equally rigorous case for why this specific allocation of capital is the right one relative to every competing option, is an incomplete argument in the only venue where the decision actually gets made.

Writing to the reader who actually decides

The practical fix starts with a blunt exercise most applicants skip: identify who is actually in the room making the final call, and what their job requires them to optimize for, independent of what the science requires. A university ethics board, a corporate R&D steering committee, and a private foundation's grant panel are not evaluating the same criteria, even when they're reviewing the identical proposal, because their institutional incentives are different. A proposal written once and sent to all three, unchanged, is very likely answering at most one of their actual questions correctly.

For a corporate or cross-institutional funder specifically, this usually means writing the risk and strategic-fit case with the same rigor as the scientific case, pricing the downside explicitly, naming the opportunity cost against the organization's other live priorities, and stating plainly what the funder gets if the work succeeds, in terms that map onto their existing strategic language rather than the applicant's own academic framing. A proposal that does this isn't compromising its scientific integrity to court funding. It's simply answering the actual question being asked, in addition to the one it already answered well.

The uncomfortable diagnosis

Most rejected proposals were never scientifically weak. That's the uncomfortable part, and it's also the useful part, because it means the fix isn't more rigorous research, it's a more honest accounting of who is actually deciding, and what decision they're actually authorized to make. A proposal that reads like the best case for a fellow scientist, submitted to a panel doing risk allocation across a portfolio, was never going to be evaluated on the terms it was written for.

The proposal that gets funded is rarely the one that proves it's the most scientifically interesting idea in the stack. It's the one that proves, convincingly, that it survives the specific question the funder was actually in the room to ask.