Erasmus+ Application Assessment Criteria Explained

A strong Erasmus+ idea does not compensate for a proposal that leaves an evaluator to infer the delivery model, partner contribution or route to impact. The Erasmus+ application assessment criteria reward evidence that is explicit, proportionate and traceable through the form. You only get one chance to submit. The assessment must therefore begin before the deadline, when a significant weakness can still be corrected.

For most cooperation and partnership actions, evaluators assess a defined set of award criteria and sub-criteria set out in the call documentation and application form. The precise labels, weightings and thresholds vary by action, sector and call year. Do not rely on a generic checklist from a previous Erasmus+ round. Use the evaluation form published for the action you are applying to, and work from that version.

Start with admissibility, eligibility and exclusion

Award criteria are not the first test. A proposal can be well written and still fail before quality is scored if it is inadmissible or ineligible. Page limits, compulsory annexes, correct submission route, eligible countries, organisation type, partnership composition, activity duration and financial rules are not editorial details. They are conditions of participation.

Eligibility checks are usually administrative, but they affect proposal strategy. A consortium may have an impressive additional partner whose role is not eligible under the action, or a work plan may include activities outside the permitted scope. Neither issue is repaired by a persuasive impact section.

Treat this as a separate control step. Ask whether every formal requirement is met, whether the application contains the required declarations and whether the activity described is actually the activity funded by that call. Only then should the team argue about quality scores.

Erasmus+ application assessment criteria: what receives points

The familiar structure for many Erasmus+ cooperation proposals covers four broad areas: relevance, quality of project design and implementation, quality of the partnership and cooperation arrangements, and impact. Their point allocations and minimum thresholds differ between action types. Some calls also frame the criteria differently or place particular emphasis on priorities, inclusion, digital practice, green transition, participation or sector-specific outcomes.

The evaluator does not award points for mentioning these themes. They award points where the proposal shows that the theme changes the design, the participants, the methods, the outputs or the measurement plan.

Relevance

Relevance is often where a proposal establishes whether it belongs in the call at all. Evaluators test the fit between the identified need, target groups, objectives, priorities and expected results. The weakness is rarely an absence of policy language. It is an unsupported chain of reasoning.

A proposal may state that young people need better digital skills, for example, but provide no evidence that its named target group lacks them, no explanation of the barrier faced by disadvantaged participants, and no reason why the proposed transnational activity is necessary. That reads as a plausible premise, not a demonstrated need.

The stronger version identifies the local or sectoral problem, distinguishes evidence from assumption, explains why the selected countries and organisations are relevant, and converts broad priorities into specific objectives. If inclusion is central, specify who may be excluded, at which stage, by what mechanism and how participation will be supported.

Quality of project design and implementation

This criterion examines whether the project can be delivered as described. Evaluators read across objectives, work packages, activities, outputs, milestones, resources, risk controls and monitoring. They look for internal consistency rather than decorative detail.

A common scoring loss occurs when objectives are measurable but the work plan does not generate the evidence needed to measure them. Another occurs when a task list exists but ownership, sequencing and acceptance criteria do not. An output called a toolkit is not a delivery plan. Who develops it, who validates it, what makes it usable, when it is tested and what happens if testing fails are the operational facts an evaluator needs.

Be cautious with KPI tables. A large number of indicators can create the appearance of control while obscuring the few measures that matter. Each indicator should have a baseline where relevant, a target, a data source, collection responsibility, timing and a decision it will inform. If a consortium cannot say what it would change after an adverse monitoring result, it has described reporting rather than management.

Quality of partnership and cooperation arrangements

An eligible partnership is not automatically a credible one. This criterion asks whether the consortium has the capacity and structure to carry out the work, and whether each organisation has a justified role.

Evaluators notice when partner profiles repeat generic claims about experience while work package assignments concentrate substantive responsibility in one organisation. They also notice the opposite problem: every partner is assigned a vague contribution, such as dissemination or support, without a defined deliverable.

Show the connection between capability and task. Name the relevant expertise, prior access to the target group, facilities, networks or institutional authority that makes each assignment credible. Then explain how the consortium will make decisions, resolve disagreement, monitor quality and handle underperformance. A governance chart without escalation routes, meeting cadence or decision rights is an organisational diagram, not a governance system.

Impact

Impact is not the final dissemination paragraph. It is the test of whether the project’s results will reach the intended users, be taken up and continue where continuation is plausible.

Weak impact sections frequently overstate reach. Claims about European uptake, policy influence or sustained use are not credible merely because materials will be posted online or partners have social-media accounts. Evaluators want to see the route from output to user: who adopts it, why they would do so, what adaptation is required, who maintains it and how use will be evidenced.

Dissemination, exploitation and sustainability should be separated even where the form discusses them together. Dissemination concerns visibility and access. Exploitation concerns uptake and use. Sustainability concerns what remains after grant funding. A plan can be strong in one and weak in the others. Be honest about the boundary: some outputs will be sustained through partner operations; others may require further funding or a host institution to continue.

How evaluators find a significant weakness

Evaluators work against the published criterion and score-band descriptors, not against the effort the consortium has invested. A section can be polished, technically informed and still score below expectation if its key claim cannot be verified in the application.

The most costly weaknesses are usually cross-document inconsistencies. The narrative promises co-creation, but participant involvement begins after the design is fixed. The impact section claims scale, but the budget has no resource for translation, adaptation or outreach. The risk register names staff turnover, but the implementation plan has no deputy role or handover process. These faults matter because they reduce confidence in feasibility.

A practical internal review should therefore follow a disciplined sequence:

1. Read each criterion and sub-criterion against the exact wording in the call’s evaluation form.
2. Mark the proposal passage that provides evidence for every scoring claim. If no passage can be marked, the claim is unsupported.
3. Test consistency across objectives, activities, budget, work packages, governance, risk and impact.
4. Distinguish a minor omission from a significant weakness: would the missing evidence materially change confidence in delivery or results?
5. Prioritise amendments that improve a low-scoring criterion or protect a threshold, rather than making cosmetic edits everywhere.

This is not a request to make every paragraph longer. In a page-limited application, specificity must replace repetition. One clear sentence assigning responsibility and acceptance criteria is worth more than a page restating ambition.

Use an independent reading before submission

The writing team is poorly placed to judge what is obvious. It knows the partner discussions, the unwritten assumptions and the evidence sitting in internal files. An evaluator sees only the submitted document, usually under time pressure and alongside competing applications.

An independent assessment should reproduce that constraint as far as possible: criterion-specific scoring, cited evidence, an explicit account of unsupported claims and visible disagreement where different readers interpret the same section differently. Averaging disagreement away can conceal the exact passage that needs revision.

BidShark applies the published marking scheme for the relevant call type, with independent specialist readings and adjudication where scores materially diverge. The automated report is not a human panel verdict; its purpose is to expose defensible weaknesses before the actual panel sees the file. Where the stakes warrant it, the separate expert Q&A package adds responses from a practising EU evaluator.

Before pressing submit, ask a less comfortable question than whether the proposal reads well: can an evaluator award every point you expect without supplying missing logic on your behalf? Where the answer is no, that is the next amendment.