On 30 April 2025 the European Commission published the interim evaluation of Horizon Europe as Commission Staff Working Document SWD(2025) 110 final[1]. The document is neither a commentary nor an outside view — it is the Commission's own official assessment of the running framework programme, based on a targeted survey of 17,254 participants and on the internal CORDA monitoring database. Two figures from it now shape the political debate about the successor programme: 32 per cent of respondents find applying to Horizon Europe more burdensome than to Horizon 2020, and 48 per cent of recipients state that they spend more than 10 per cent of their project budget on administrative work. Both figures appear as a central finding in the European Parliament's official resolution of February 2025 on the interim evaluation[2]. This piece places them in their original sources, separates indicator from interpretation, and names the structural causes the Commission itself documents.
What the Commission measures itself
Horizon Europe is the European Union's ninth research framework programme, anchored in Regulation (EU) 2021/695 of 28 April 2021 and originally endowed with a total budget of €95.5bn for 2021 to 2027; after the revision of the multiannual financial framework (MFF) in February 2024, the envelope was reduced to €93.5bn, still making it the largest research and innovation funding programme in the world[3]. Article 52 of the regulation obliges the Commission to carry out an interim evaluation by 31 December 2024 and then submit it to the European Parliament and the Council. The interim evaluation is therefore not an academic exercise — it is a legally prescribed reporting point, and one to which the successor framework programme (FP10) connects in substance.
Methodologically the assessment rests on three sources that check one another. First, the CORDA monitoring database, which records administrative metrics for every single signed contract — from time to inform through time to sign to the final payment. Second, a targeted survey of 17,254 respondents, split into 64 per cent unsuccessful applicants (11,028) and 36 per cent successful ones (6,226), with a sub-dataset of 5,161 recipients who commented on their project's administrative costs[1]. Third, a public consultation and external evaluation studies of the individual parts of the programme. The Commission thus combines operational metrics from its own system with participants' subjective feedback and tests plausibility against the OECD's Frascati framework[11].
The evidence base for the evaluation was set out in the earlier working document SWD(2023) 132 final[7]. The indicators defined there — on administrative burden, application costs, the duration of the procedure and the share of SMEs and first-time participants, among others — were carried through consistently into the interim evaluation. The figures published in SWD(2025) 110 final are therefore not a snapshot but the result of three years of observation within a defined measurement framework.
32 per cent — the burden indicator
The interim evaluation's most prominent finding is the answer to the question of whether access to Horizon Europe has become easier than to Horizon 2020. The European Parliament's resolution A10-0021/2025 sums the finding up in paragraph 24: "administrative simplification stagnated under Horizon Europe given that 32 % of participants consider applying to Horizon Europe to be more burdensome than Horizon 2020, while nearly half of participants report no difference"[2]. The 32 per cent is therefore not the result of an external lobbying survey but the average from the Commission's own targeted survey.
The counterpart figure matters just as much: almost half of respondents report no noticeable difference, and only around a fifth report a perceptible simplification. In the Commission's official wording, "applicants have not experienced any substantial change in the effort involved in proposal preparation and submission compared to Horizon 2020", on a base of 9,668 responses to that specific question[1]. The share who perceived relief (around 31 per cent) does sit slightly above the share who perceived the opposite (around 18 per cent), but it falls well short of the expectations Regulation 2021/695 itself set out: the recitals and Article 6 make it an explicit objective that "the Commission shall continue to aim for administrative simplification and a reduction of the burden for the applicants and beneficiaries"[3]. Measured against that legally fixed aspiration, the interim evaluation is not a celebration but a diagnosis.
Notable is how the survey figures validate against the earlier public consultation: there, 74 per cent of respondents reported an effort "similar" to Horizon 2020 and 17 per cent a "higher" one[1]. The qualitative assessment of the two waves is thus congruent — and on very different samples. The Commission itself calls the result a "strong finding", particularly against the background that Horizon Europe started with a larger budget and that success rates have risen compared with Horizon 2020.
48 per cent above 10 per cent admin costs
The interim evaluation's second lead indicator measures money rather than feeling. In the targeted survey, 5,161 recipients were asked what percentage of their project budget they spend on administration, project reporting and project-related financial management. The answer categories ran from "less than 1 per cent" to "more than 20 per cent". The median and the most frequent value (the mode) lie, in almost all parts of the programme, in a corridor of 6 to 10 per cent of the project budget[1].
"on average beneficiaries reported spending 6-10 % of their project budget on administrative costs, with 48 % reporting administrative costs of more than 10 %, including a 10 % share of beneficiaries reporting administrative costs of more than 20 %"— European Parliament, resolution A10-0021/2025, paragraph 24[2]
Translate that into absolute figures and the scale becomes clear. In SWD(2025) 110 final the Commission itself estimates that all Horizon Europe projects signed up to the cut-off date will produce, across their full term, between €4.7bn and €6.5bn in administrative costs — a corridor corresponding to 9 to over 12 per cent of the project costs contracted so far[1]. These are not individual budget lines but a collective heading for activity that is not research itself: contract negotiation, periodic reporting, time records, internal consistency checks, answering queries from the disbursing agencies, preparing for audits.
The distribution across types of organisation is interesting: 16 per cent of research funding organisations, 12 per cent of non-governmental organisations and 11 per cent of universities report administrative costs above 20 per cent; among SMEs, startups, large companies and private research institutions that share is considerably lower, at around 7 per cent[1]. The median values, though, shift significantly upwards neither for coordinators (who report 11 to 15 per cent on average) nor across consortium sizes — an empirical finding that runs against the intuitive narrative that "large consortia automatically generate more overhead", and one the Commission confirms in its own regression analysis.
Where the extra burden comes from
The interim evaluation does not look only qualitatively at why a framework programme marketed as simplified is stagnating in participants' perception. It breaks the process down into the four procedural stages for which the Commission keeps its own targets: time to inform (TTI), time to sign (TTS), the two combined as time to grant (TTG), and time to pay (TTP)[1]. For each of these steps, the monitoring shows where delays relative to Horizon 2020 have taken hold.
The target for time to grant in Horizon Europe is 245 days, that is eight months — the upper limit laid down as the "maximum time period" by Article 31 of Regulation 2021/695[3]. The average actually achieved across the projects registered in CORDA up to January 2025 is 240 days, with only 77 per cent of grants concluded within the limit[1]. A direct comparison: average time to sign is 95 days against 76 days under Horizon 2020; time to inform 130 days against 112 days. Grant agreement preparation — the stage in which budget, partners, timing and reporting obligations are negotiated — has become noticeably longer and more laborious. The Commission puts it cautiously but unambiguously: "the grant agreement preparation phase remains the more challenging period of the two phases of the time-to-grant period"[1].
In substance, the extra burden concentrates in three places. First, the contractual apparatus itself: the Annotated Grant Agreement (AGA) in version 2.0 of 1 April 2025 runs to 810 pages, covering Articles 1 to 59 of the model grant agreement plus 20 annexes[4]. Every substantial cost item — personnel, travel, materials, depreciation, subcontracting, third parties — has its own eligibility criteria, its own evidence and its own accounting logic. Second, the digital portal: the EU Funding & Tenders platform forces applicants into standardised forms whose field semantics do not always match the AGA's requirements. In the targeted survey, 45 of 189 qualifying open responses (around 24 per cent) explicitly criticised the disproportion between the effort of applying and the chance of success[1]. Third, the reporting architecture: alongside classic cost reporting for real-cost grants, Horizon Europe also runs lump sum grants (flat funding per work package), personnel unit costs and hybrid constructions. Having these funding types side by side is meant as a simplification, but it creates extra rounds of deliberation in applying and in delivery.
The lever marketed as having the greatest simplification potential — lump sum grants — does indeed reduce financial reporting effort considerably: 65 per cent of lump sum beneficiaries surveyed agree that this form of funding lowers administrative effort[8]. By 1 January 2025, 1,582 lump sum grants had been signed under Horizon Europe, 706 of them ERC Proof of Concept grants[1]. That is still less than a tenth of the 15,148 Horizon Europe grants signed in the programme's first four years, and it explains why the overall perception of simplification is stagnating in the surveys.
A structural comparison: H2020 vs Horizon Europe
The interim evaluation is not the only document comparing Horizon Europe with its predecessor. The Annual Report on the European Union's research and technological development activities and monitoring of Horizon Europe and Horizon 2020 in 2024, published as COM(2025) 298 final on 6 June 2025, sets out the operational metrics for both programmes side by side[6]. Three structural differences are central to the burden picture.
First, the programme architecture. Horizon 2020 was built on three priorities (Excellent Science, Industrial Leadership, Societal Challenges); Horizon Europe is ordered in three pillars, with a newly cut Pillar II ("Global Challenges and European Industrial Competitiveness") that is itself divided into six thematic clusters (Health; Culture, Creativity, Inclusive Society; Civil Security; Digital, Industry, Space; Climate, Energy, Mobility; Food, Bioeconomy, Natural Resources). Five EU missions and a considerably expanded partnership landscape were established on top. The number of legal and programmatic interfaces at which a project has to be located has therefore risen against Horizon 2020[3].
Second, the funding architecture. Regulation 2021/695 explicitly simplified cost reimbursement: the split between basic and additional remuneration for staff costs was abolished, the Horizon 2020 cap on additional remuneration removed, and unit costs for personnel extended[3]. In parallel, though, new cost rules were introduced for internationalised projects, for the "associated country" regime (the UK, Switzerland and Israel in particular) and for the partnership formats. The effect: selective simplifications coexist with new questions of delimitation. In its 2024 annual report the European Court of Auditors finds that despite these simplifications the error rate in checking cost eligibility shows "no significant difference" between Horizon 2020 and Horizon Europe — 26 of 99 audited projects in the representative sample contained quantifiable errors.
Third, reporting frequency. Horizon 2020 had established stable, standardised periodic reporting; Horizon Europe additionally spreads reporting obligations across requirements specific to parts of the programme (the EIC Accelerator with pitch decks and business plans, missions with impact reporting, EIT KICs with cascade funding conditions). In terms of time that means: 41 per cent of respondents in the beneficiary survey neither agree nor disagree with the statement that project management and delivery have become easier[1]. That neutrality is itself a finding — it means the rhetoric of simplification has not visibly landed in delivery.
The average consortium size has grown from 5.0 to 7.1 partners, and the average project duration from just under 39 to around 42 months. Both add up in proposal preparation: the median coordinator reports 36 to 45 person-days per application, and the mode is higher still. In consortia with more than 30 partners the effort per individual partner does not fall linearly, because the coordination layer needs extra rounds of agreement[1].
What this means for SMEs in particular
Horizon Europe has visibly raised the SME share of the participant base compared with Horizon 2020. The Directorate-General for Research and Innovation's report SME participation in Horizon Europe — Key figures and key issues in the first three years shows that around a third of all Horizon Europe participants are SMEs and that more than half of SME participants are newcomers to the framework programme[5]. Total flows to SMEs stand at around €6.6bn over the first three years, over 20 per cent of the funding contracted by then. The success rate for SMEs has risen markedly, to 19.9 per cent from 12 per cent under Horizon 2020[5].
At first glance that speaks for a programme that has made access easier for mid-sized companies and startups. A second look is more nuanced. The cost distribution shows that SMEs and startups are not structurally better off on administration: 7 per cent of SME respondents put their administrative costs above 20 per cent of the project budget — a share lower than at universities, but unchanged in absolute terms[1]. Because SME projects in Horizon Europe typically have smaller overall volumes, a percentage cost burden hits them harder: 10 per cent of €2m on a Cluster 4 collaborative project is €200,000 that does not go into the actual development.
The Federal Ministry of Research, Technology and Space (BMFTR) points to the national contact points (NKS), which provide advice and coordination for German participants on the federal government's behalf[9]. The network is organised into ten thematic contact points plus an office at the DLR Project Management Agency[10]. The NKS themselves mark clearly in their guides where the greatest effort lies: delimiting the project, choosing consortium partners, deriving the budget under the AGA, checking for state aid, the data management plan, the reporting plan. For newcomers — the more than 50 per cent of SMEs that have never taken part in a framework programme — the entry hurdle into the AGA's system is felt most sharply. The European Parliament puts this in A10-0021/2025 as "administrative burden, especially for newcomers and small organisations"[2].
The EIC instruments — the EIC Accelerator in particular — show the divergence between the chance of success and the effort especially clearly. 40 per cent of successful Accelerator applicants report an application effort above 65 person-days — the most frequent value in this programme line[1]. With rejection rates that can exceed 80 per cent, the total cost sum comes out on average well above what a single SME can financially absorb. The Commission itself rates this a "strong finding" and names it as a central starting point for the successor framework programme.
Taken together, then, the interim evaluation's figures are not evidence of a failed programme but evidence that Horizon Europe has not achieved across the board the simplification Regulation 2021/695 expressly promised. 32 per cent finding it more burdensome, 48 per cent with administrative costs above 10 per cent of the project budget, a time to grant that meets the statutory eight-month frame only 77 per cent of the time — these are the metrics against which the successor programme FP10 will have to be measured. For applicants they mean that working with Horizon Europe remains a fine-grained, documentation-intensive procedure that has become substantially easier neither through the digital portal nor through individual simplifications such as lump sum grants. The procedural structure this requires — delimiting the project, consortium responsibility, AGA compliance, the reporting plan, the liquidity forecast, audit preparation — we build with you, and we document every step so that it stays auditable across the whole programme term.
- [1]Commission Staff Working Document — Interim Evaluation of the Horizon Europe Framework Programme for Research and Innovation (2021–2024), SWD(2025) 110 final, Brussels, 30 April 2025European Commission, Publications Office of the EU · 2025Open source
- [2]European Parliament — REPORT on the assessment of the implementation of Horizon Europe in view of its interim evaluation and recommendations for the 10th Research Framework Programme, A10-0021/2025European Parliament, Committee on Industry, Research and Energy (ITRE) · 2025Open source
- [3]Regulation (EU) 2021/695 of the European Parliament and of the Council of 28 April 2021 establishing Horizon Europe — the framework programme for research and innovationEuropean Union, EUR-Lex · 2021Open source
- [4]EU Grants — Annotated Grant Agreement (AGA), version 2.0 of 1 April 2025European Commission, DG Research and Innovation · 2025Open source
- [5]SME participation in Horizon Europe — Key figures (and key issues) in the first three years, July 2024European Commission, Directorate-General for Research and Innovation · 2024Open source
- [6]Annual Report on the European Union's research and technological development activities and monitoring of Horizon Europe and Horizon 2020 in 2024, COM(2025) 298 final, 6 June 2025European Commission · 2025Open source
- [7]Commission Staff Working Document — Evidence Framework for the monitoring and evaluation of Horizon Europe, SWD(2023) 132 final, 27 April 2023European Commission · 2023Open source
- [8]Assessment of lump sum funding in Horizon 2020 and Horizon Europe, September 2024European Commission, Directorate-General for Research and Innovation · 2024Open source
- [9]Horizon Europe — portal page of the Federal Ministry of Research, Technology and SpaceFederal Ministry of Research, Technology and Space (BMFTR) · 2025Open source
- [10]The German network of national contact points (NKS) — Horizon Europe, office at the DLR Project Management AgencyBMFTR Horizon Europe / DLR Project Management Agency · 2025Open source
- [11]Frascati Manual 2015 — Guidelines for Collecting and Reporting Data on Research and Experimental DevelopmentOECD Publishing, Paris · 2015Open source
Horizon Europe — what the 17% success rate leaves out
Only 6% get through the EIC Accelerator. 240 days average time to grant, only 77% within the 245-day limit. A year of your life per application — with the outcome open.
The 20/80 asymmetry: why writing the application is the smallest part of the work
The FDP Faculty Burden Survey, the EC's Horizon Europe interim evaluation and the ECA error statistics show where the effort actually lies — and where classic consultants structurally cannot reach.
The ten-year rule: audit obligations after the grant notice
In the ECA's sample of audited research transactions, one in four shows quantifiable errors. Staff costs are the number one source of error — and the biggest block of the budget.
From the analysis into the application.
We show the platform on a real case.