Evaluation (application review)

Three criteria are assessed, and your proposal's structure should mirror them.

Excellence – clarity and pertinence of the objectives, how far the work is ambitious and goes beyond the state of the art, soundness of the methodology including the gender dimension and open science practices.

Impact – credibility of the pathways to the outcomes expected in the work programme, and the quality of the measures to maximise impact including dissemination and communication.

Quality and efficiency of the implementation – the work plan, risk assessment, appropriateness of effort and resources, and the capacity of each participant and of the consortium as a whole.

Scale and thresholds

Each criterion scores 0 to 5, and half marks are allowed (tenths are not – that is a common invention).

0 – fails to address the criterion or cannot be assessed. 1 – serious inherent weaknesses. 2 – addressed broadly, significant weaknesses. 3 – good, with a number of shortcomings. 4 – very good, few shortcomings. 5 – excellent, shortcomings minor.

Note that even a 5 allows shortcomings, provided they are minor. The difference between 4 and 5 is their number and weight, not their absence.

The threshold per criterion is 3, the overall threshold 10 out of 15. Proposals passing both are considered for funding within the available call budget; the rest are rejected.

Weighting: for ranking only

Scores are normally not weighted. One exception: in innovation actions Impact carries a weight of 1.5 – but only for the ranking, never for passing thresholds.

Two-stage calls

Stage one assesses only Excellence and Impact, and only the aspects shown in bold. The threshold is higher – 4 per criterion. The overall threshold floats: it is set so the combined budget of proposals invited to stage two is close to three times the available budget and not less than twice. In practice this is usually 8 or 8.5.

Who assesses, and how

First an individual evaluation: each expert scores with comments and writes their own report. A proposal is normally read by at least three experts, sometimes two, often five.

Then a consensus group: the same experts agree a common position, moderated by a Commission official. Where no agreement is reached, a minority opinion is recorded with the dissenting names.

Finally the panel: it checks consistency across proposals, may revise scores, resolves disagreements and produces the ranked list. In Horizon Europe the panel usually consists exclusively of external experts.

What happens on equal scores

The priority order is fixed: first proposals covering call aspects not covered by others; then the higher Excellence score (for innovation actions, Impact first); then gender balance among researchers in leading roles; then geographic diversity – countries not yet represented higher in the ranking; and only then other panel considerations.

An inflated budget lowers the score

Worth knowing separately. Experts are forbidden to recommend modifications – they may not suggest cutting the budget or changing the consortium. Shortcomings must instead be reflected in a lower score. So an inflated budget is not "fixed at grant preparation" – it costs you points under the implementation criterion.