Most Horizon Europe proposals that fail don't fail because the idea is weak. As an Independent Expert Evaluator for the European Commission, I read a lot of technically strong projects that still score below threshold. The reason is almost always the same: the proposal doesn't do the one job an evaluator needs it to do, which is to make the excellence, the impact and the delivery logic impossible to miss in a limited reading time.
That's good news, because it means the gap is fixable. It's a writing and structuring problem, not a science problem, and it can be closed before you submit rather than after a rejection.
What evaluators are actually scoring
Every Horizon Europe proposal is scored against three criteria: Excellence, Impact, and Quality and efficiency of the implementation. Evaluators work from the call text and the evaluation form, not from your enthusiasm for the technology. If a point the form asks about isn't clearly answered on the page, it doesn't count, however obvious it seems to you.
Here are the patterns I see most often, criterion by criterion.
1. Excellence: a strong idea with a weak structure
The Excellence section should let an evaluator answer three questions quickly:
What exactly are you building, and what is new about it compared with the state of the art?
Why does it matter for the call's expected outcomes?
How will you prove it worked, with measurable objectives and a credible methodology?
Strong engineering teams often answer these implicitly, spread across pages of technical description. The evaluator then has to reconstruct the argument, and that costs points. I covered this in more detail in How to Structure a Winning Horizon Europe Excellence Section.
2. Impact: describing outputs instead of change
This is the most common reason a good proposal scores low. The Impact section describes what the project will build, not what changes because it exists. Typical signs:
Vague claims about "societal benefit" with no numbers behind them.
No named end users or early adopters outside the consortium.
A dissemination plan that amounts to "conferences and publications".
Evaluators aren't scoring your idea here. They're scoring how convincingly you've shown that someone outside your consortium will actually use what you build. A useful test: could a stranger reading only your Impact section tell exactly who benefits, how, and how you'll prove it?
3. Implementation: a to-do list instead of a delivery plan
The work plan often reads like a list of tasks rather than a risk-managed plan. Evaluators look for clear dependencies between work packages, realistic milestones, a risk table with real mitigation measures, and a consortium where each partner's role is obvious.
The budget is part of this too. If the person-months and costs don't visibly map back to the deliverables being scored, evaluators notice. A budget that can't be traced to the work plan reads as padding, even when it isn't.
Fix it before you submit
None of these are technical problems. They're proposal-writing problems, and they're far cheaper to fix before submission than after a rejection and a six-month wait for the next cut-off.
If you're preparing a Horizon Europe, Portugal 2030 or Interreg application, you can start with our free 2-minute Funding Fit Check. If you already have a draft, our Evaluator's-Eye Review gives you an independent, scored read of it from someone who evaluates these applications for the European Commission.

