Having managed delivery on proposals across Horizon Europe, EIC and Erasmus+, I've sat on the other side of enough scoring outcomes to see where applicants misjudge the process. The mechanics are public, but few applicants plan their drafting time around them — and that's usually the difference between a proposal that clears threshold and one that doesn't.
Standard Horizon Europe proposals are scored on three criteria, each on a 0-5 scale in half-point increments:
| Criterion | What it actually evaluates |
|---|---|
| Excellence | Clarity of objectives, soundness of concept and methodology, ambition relative to state of the art |
| Impact | Credibility of the pathway to outcomes, quality of dissemination/exploitation/communication planning |
| Quality & Efficiency of Implementation | Work plan coherence, resource allocation, consortium capability and complementarity |
Each criterion typically needs to clear an individual threshold (commonly 3/5) and the sum needs to clear an overall threshold (commonly 10/15) — but clearing threshold is not the same as being funded. Calls are almost always oversubscribed relative to budget, so proposals are ranked by total score once threshold is cleared, and funding goes down the ranked list until the budget runs out. A proposal that passes every threshold can still be unfunded if it sits below the funding line.
Where applicants misallocate drafting time
Excellence is where most applicants — understandably — spend the bulk of their effort, since it's the section closest to their own expertise. But Implementation is scored independently and just as heavily, and it's the section most often finished in the final days before deadline: a rushed Gantt chart, a risk table copied from a previous proposal, a management structure diagram with no explanation of the decision-making process behind it. A brilliant technical concept with a thin Implementation section still loses points it didn't need to lose.
What actually strengthens an Implementation score
- Work packages with genuinely distinct, measurable deliverables — not descriptions that could apply to any similar project
- A resource table (person-months, budget) that visibly matches the ambition described in Excellence, rather than being built separately by whoever handles the budget
- A governance structure that names who resolves a disagreement between partners, not just who chairs meetings
How the evaluation actually happens
Proposals are typically scored first by individual remote experts, then discussed in a consensus meeting where those experts align on a single score and written justification, sometimes followed by a panel review across proposals in the same call. This matters for how you write: a section that makes sense to you as the author, but that individual reviewers might read differently before comparing notes, is a risk. Ambiguity that survives an individual read can get resolved against you in consensus, since the evaluator who scored it lower has to justify raising it, not the other way around.
None of this replaces strong science. It just means the strongest science still needs an Implementation section written with the same care as Excellence, and a draft tight enough that it reads the same way to every reviewer who opens it.