The Horizon Europe Part B template, section by section

A Horizon Europe application has three parts, and only one of them is the document evaluators actually read and score. Part A is the structured administrative form you complete in the Funding & Tenders Portal (participants, budget, declarations). Part B is the technical description: a single PDF built on a fixed template, and the part where your proposal is won or lost. Part C and the annexes hold the supporting documents, such as the ethics self-assessment, letters of support, and anything else relevant to the Commission and the evaluators.

This page covers the template for collaborative Research and Innovation Actions and Innovation Actions (RIA and IA), the Pillar 2 cluster calls that make up most of Horizon Europe. Missions and Joint Undertakings use essentially the same structure. Other action types and programmes (CSA, EIC, Digital Europe, COFUND, ERC, MSCA) use genuinely different templates, and two-stage calls use a separate, shorter stage 1 template. There is more on that at the end.

This guide goes through the RIA/IA Part B subsection by subsection: what to write in each, roughly how much space it should take, and what evaluators are looking for. It is the same guidance GrantForge gives writers inside each section of the template. If you already have a draft, from a previous call or written elsewhere, you can import it into GrantForge and it is reformatted into the exact, compliant template for your call, so you can use this page to improve content rather than rebuild structure.

How Part B is organised

Part B is not arranged the way you would tell the story of your project. It is arranged the way it is scored. The three numbered sections match the three award criteria, each marked out of 5 with a threshold to pass:

Excellence: is the science sound and ambitious?
Impact: if it works, does it matter, and will the results be used?
Quality and efficiency of the implementation: can this consortium actually deliver it?

Write in this order, under these headings. When you merge subsections or answer out of sequence, the evaluator has to search for the points they are required to award, and anything harder to find scores lower. How the marks and thresholds work is covered in how Horizon Europe proposals are scored.

Part B is arranged the way it is scored: three sections, three criteria.

Section 1: Excellence

1.1 Objectives and ambition

Problems and needs. The opening sets the tone, so make the first impression count. Describe the problem you address, quantify it, and show it is a genuine pain, ideally an urgent one. Explain the current solutions and their limits, and how your project improves on what is already known. Name the stakeholders affected by the challenge in the call topic (these are your target groups), and the benefits they will gain. Then make the case for why this proposal, and this team, is the right one to solve that specific challenge. A concept figure here helps the evaluator picture the project at a glance.

Goal and objectives. State one overarching goal in a single clear sentence. Below it, define your specific objectives: 3 to 5 is the usual range for a RIA or an IA. Before detailing them, add a short paragraph linking them to the work programme. Each objective must be measurable and verifiable, so give it a KPI and a means of verification. Present each objective as its own small table: the objective shortcode and name, a description, the call's specific challenges and expected outcomes it covers (written out by full name), one row per KPI, and the key results attached to it. One table per objective is what tells the evaluator, at a glance, that each is complete.

Ambition. Show how the project goes beyond the state of the art, and how ambitious it is. Be explicit about the novelty. The bar depends on the action type: a RIA brings new scientific insight, technologies, or proofs of concept; an IA brings first adoption, a new product, service, business model, or a large-scale demonstration. Where relevant, compare with products and services already on the market, and refer to the patent or publication search you carried out.

R&I maturity. For each key result, give a start TRL and an end TRL, with a justification for how the project reaches the end level. This has to stay coherent with the ambition you just described.

1.2 Methodology

This is the long narrative of the document, usually 4 to 10 pages. Describe the concepts, techniques, models and assumptions that underpin your work, and explain how they let you deliver the objectives, in other words why the project is feasible. It is not the work plan, which belongs in section 3. Set out the key research questions and how you will tackle them.

Section 1.2 also carries the cross-cutting content that first-time applicants tend to underweight. Each has a rough page budget, and each has an escape hatch: if it is genuinely not relevant to your project, you may say so with a justification rather than pad it.

Building on other projects (around 1 page): relevant EU projects (search CORDIS), plus national and international ones, and how you build on them. Note which partners were involved.
Interdisciplinarity (around a quarter page): how methods from different disciplines are integrated, or a justification if an interdisciplinary approach is unnecessary.
Social sciences and humanities (around half a page): how SSH is engaged from the start, which also strengthens societal impact.
Gender dimension (around 1 page): how the project addresses gender equality and inclusion in its content. A 50/50 balance is not required. Justify if not relevant.
Open science: how open science practices are built into the methodology, adapted to the nature of the work.
Research data management (around 1 page): an early data management plan showing how data and other outputs follow the FAIR principles (findable, accessible, interoperable, reusable).

Section 2: Impact

Impact carries a ranking weight of 1.5 for Innovation Actions at the panel stage, so on close calls it is the criterion that separates funded from rejected.

2.1 Project pathways towards impact

Start from your target groups. Then set out your contribution to the expected outcomes of the topic: aim to cover most of the outcomes listed, over the medium term, each with a realistic KPI and timeframe. Then your contribution to the wider expected impacts of the destination, over the longer term, with the benefit for each target group. This subsection is scored against the exact wording in your call topic, so quote and mirror that language rather than paraphrasing it.

Close with a barriers table: the requirements and obstacles on the path to impact, each with its mitigation.

2.2 Measures to maximise impact

Two things here. First, the plan for dissemination, exploitation and communication (the DEC plan and its tasks, each with a KPI). Second, your intellectual property strategy, around half a page: how partners handle existing knowledge and IPR, and how results will be protected, followed by a table of key exploitable results with their planned ownership and protection measures. The full treatment is in the dissemination and exploitation plan.

2.3 Summary

A one-page canvas linking, in six cells, your problems, target groups, DEC measures, outcomes and impacts. Use it to check the coherence of your whole impact pathway. As of 2026 this section is optional, and it is the only optional part of the template.

Section 3: Quality and efficiency of the implementation

This section is mostly tables, and it is where a vague project shows itself. Numbers and dates that do not add up are visible here in a way narrative can hide.

3.1 Work plan and resources

Open with a PERT diagram and a short explanation of why the work divides into these work packages and how they relate. Add a Gantt chart timing the work packages, tasks, deliverables and milestones. Then come the imposed Horizon Europe tables, whose formats are fixed:

Work packages (table 3.1a, 7 columns), plus a detail card per work package with objectives, work described, tasks, leader, contributors and duration.
Deliverables (8 columns) and milestones (6 columns).
Critical risks (table 3.1d, 3 columns), with likelihood and severity.
Summary of staff effort in person-months, per beneficiary and per work package.
Subcontracting costs and purchase costs. Note that since 2026 purchase costs are limited to major equipment only.

3.2 Capacity of participants and consortium as a whole

Describe the excellence and expertise of each partner against the objectives, including affiliated and associated partners (around half a page). Show how the consortium will access any critical infrastructure. Add a complementarity diagram with partners as rows and fields of expertise as columns, so the coverage is visible at a glance. Where relevant, describe the industrial or commercial involvement that will secure exploitation. Finally, justify the participation of any partner based in a country not automatically eligible for EU funding.

This is the RIA/IA template, not the only one

Everything above describes the Part B template for collaborative RIA and IA calls, with a page limit of 40 (45 under lump-sum funding) in the current 2026 application form. It is the most common Horizon Europe template, but it is not universal, and using the wrong one is an admissibility risk rather than a style choice:

Missions and Joint Undertakings (CleanH2, Chips, EU-Rail and others) use essentially this same structure, with minor programme-specific additions and different page limits.
CSA (Coordination and Support Actions) is a support action, not research: its section 1 is built differently, with no ambition or state of the art, and it is shorter.
Two-stage calls use a separate, short stage 1 template, a pre-proposal rather than a trimmed version of this one.
Other programmes, EIC, Digital Europe, COFUND, ERC and MSCA, have genuinely different Part B templates, sometimes with a different logic entirely (Digital Europe, for example, leads with relevance and has no Excellence section).

GrantForge holds the exact, current template for each of these in the database, so when you import a draft or start a proposal it is built on the right structure for your specific call, not a generic one. Which action type fits your project is covered in RIA vs IA vs CSA.

Lump-sum or actual-cost: the budget modality reshapes section 3

Independently of action type, a full call runs on one of two funding modalities, and it changes how section 3 and your budget work. Settle which one applies before you draft the work plan:

Actual-cost (the classic modality): you are reimbursed for the eligible costs you actually incur, reported by cost category and subject to financial audit. The budget lives mainly in the Part A tables; section 3 carries the staff-effort breakdown.
Lump-sum: the grant is a fixed amount agreed up front, and you are paid per completed work package, not against real costs, with no cost audit afterwards. Each work package therefore has to be a self-contained, verifiable payment unit, so lump-sum proposals carry more, and tighter, work packages and tasks, plus a detailed budget table splitting the estimated cost per work package and per beneficiary. That extra structure is why the page limit is higher.

Before you submit: the compliance checklist

Once the content is right, the mechanical rules still reject proposals. Check them last, but do check them:

Page limit: 40 pages for RIA/IA in the current 2026 application form (45 under lump-sum funding). Anything beyond it is not read: after the deadline the excess pages are made invisible and are not taken into consideration by the experts.
Formatting: minimum font size 11 points, Times New Roman for the body, A4 with margins of at least 15 mm. The Portal runs a technical check, and reducing the font to fit more text is a known move that evaluators are told to watch for.
AI use declaration: you must declare the use of generative AI in preparing the proposal. It does not disqualify you, but it must be stated.

From template to submitted proposal

Getting the structure right is the start. GrantForge takes you the rest of the way on the same document:

Import an existing draft and it is reformatted into the exact, compliant template for your call, so a previous proposal or an outside draft becomes a correct starting point in minutes.
Write with the Grant Coach, which gives inline advice on each section and subsection, drawn from the guidance above.
Work from the decoded call, to check your eligibility and confirm your objectives and outcomes actually cover what the topic asks for.
Run a pre-submission evaluation that mirrors how real evaluators score, so you see the weak sections while you can still fix them, not after the deadline.

Start your Part B in GrantForge →

Part of our guide on how to write a winning Horizon Europe proposal.