How to Demonstrate Novelty in Your Horizon Europe Proposal

Novelty Is Not Optional

Under the Excellence criterion, evaluators explicitly assess whether your proposal goes beyond the current state of the art. A proposal that describes competent research but offers nothing new will struggle to score above 3 — the minimum threshold.

Yet many proposals fail to demonstrate novelty convincingly. Not because the research lacks originality, but because the proposal doesn't articulate it clearly enough for evaluators to recognise.

This guide explains how to structure your novelty argument, based on the evaluation framework in the Horizon Europe Programme Guide.

What "Beyond the State of the Art" Means

Evaluators are looking for a clear advancement over what currently exists. This can take several forms:

1. New knowledge — discovering something previously unknown
2. New methodology — applying a fundamentally different approach to an existing problem
3. New combination — integrating existing methods in a novel way that produces better results
4. New application — applying known technology to a new domain where it hasn't been used
5. Significant improvement — achieving measurably better performance than existing solutions

All of these are valid forms of novelty. The key is making the advancement explicit and evidence-based.

Step 1: Map the State of the Art

Before you can argue novelty, you need to establish the baseline. A strong state-of-the-art section should:

Cover the Full Landscape

Don't cherry-pick references that make your approach look unique. Include:

• The leading approaches and methods in your field
• Key competing projects (including those funded by Horizon Europe or H2020)
• Recent publications from other research groups
• Relevant commercial products or technologies already on the market

Identify Specific Limitations

For each existing approach, state what it cannot do:

• "Method X achieves high accuracy but requires labelled training data that is expensive to obtain"
• "Technology Y works well at laboratory scale but has not been demonstrated above TRL 4"
• "Approach Z addresses problem A but does not account for factor B, which limits its applicability to real-world conditions"

Be Honest About Your Own Prior Work

If you have published or developed earlier versions of the proposed approach, acknowledge this. Evaluators can check. Position the proposal as a clear step forward from your previous results, not a re-packaging.

Step 2: Articulate Your Novelty

Once the baseline is established, explain exactly what is new. Use a structure like:

Current state: [Existing approaches] achieve [results] but are limited by [specific limitations].

Our approach: We propose [specific innovation] that addresses [specific limitation] by [mechanism]. This differs from existing work because [explicit comparison].

Expected advancement: This will enable [specific capability] that is not currently possible, advancing the state of the art from [TRL X] to [TRL Y] / from [performance baseline] to [target performance].

Be Specific

Weak novelty claim: "Our approach is more efficient than existing methods."

Strong novelty claim: "Current graph neural networks for molecular property prediction require pre-training on millions of labelled molecules (ref). Our self-supervised approach eliminates this requirement by leveraging 3D conformational data, reducing training data needs while maintaining prediction accuracy within 5% of supervised baselines."

Use a Comparison Table

A table comparing your approach to 3–5 existing methods on key dimensions (accuracy, scalability, cost, TRL) is one of the most effective ways to demonstrate novelty. Evaluators can see the advancement at a glance.

Feature	Method A	Method B	Our approach
Dimension 1	Limited	Good	Improved
Dimension 2	Not addressed	Partial	Fully addressed
Dimension 3	High cost	Medium	Low cost

Step 3: Provide Evidence

Novelty claims need backing. Depending on the type of novelty, provide:

• For new methodology: Cite preliminary results, feasibility studies, or theoretical analysis showing the approach is viable
• For performance improvements: Reference your own pilot data or published benchmarks showing the gap between current and proposed performance
• For new applications: Explain why existing solutions don't work in the target domain and what adaptation is required

If you have preliminary results, include them. A figure showing proof-of-concept data is worth more than a paragraph of promises.

Common Mistakes in Demonstrating Novelty

1. Assuming the Evaluator Knows Your Field

Evaluators are experts, but they may not be specialists in your exact sub-field. Don't assume they know why your approach is novel. Spell it out. If the novelty is in a subtle methodological difference, explain why that difference matters in terms of outcomes.

2. Claiming Novelty by Omission

Ignoring competing approaches doesn't make your proposal novel — it makes it look poorly researched. Evaluators will notice major omissions and may conclude that you are unaware of the competition.

3. Confusing "New to Us" with "New to the Field"

Adopting a method that is well-established in another field is not inherently novel. If you're applying machine learning to a new domain, the novelty is in the adaptation, not in using machine learning itself. Explain what makes the application non-trivial.

4. Over-Claiming

Evaluators are sceptical of proposals that claim to "revolutionise" or "disrupt" without evidence. Use precise, defensible language. "We expect to improve X by Y%" is stronger than "We will transform the field."

5. Novelty Buried in the Text

If evaluators have to search for your novelty argument, it is not prominent enough. Consider:

• A dedicated "Beyond the State of the Art" subsection
• Bold text or a highlight box for the key novelty statement
• A comparison table (as described above)

Novelty Across Different Action Types

The nature of expected novelty varies by funding instrument:

Research and Innovation Actions (RIA) — evaluators expect scientific or technological novelty: new knowledge, new methods, proof that an approach works at a fundamental level.

Innovation Actions (IA) — novelty is more about application and integration: demonstrating that existing technologies can work together in a new way, at a new scale, or in a new context.

EIC Pathfinder — the bar is highest here. The EIC explicitly requires a "convincing long-term vision of a radically new technology" with "concrete, novel and ambitious science-towards-technology breakthrough." High-risk/high-gain is the expectation.

Benchmarking Against Funded Projects

One effective way to calibrate your novelty claim is to study what has been funded in your topic area. If your approach is similar to three funded projects from the last call, evaluators may question what is genuinely new.

Conversely, if no funded project addresses the specific gap you've identified, that strengthens your case — provided you can explain why the gap exists and why now is the right time to address it.

You can search funded Horizon Europe and H2020 projects by topic, keywords, and research area in our Project Explorer.

Key Takeaways

1. Novelty must be explicit — don't make evaluators guess what is new
2. Start with a thorough state-of-the-art that covers competitors honestly
3. Use comparison tables to show the advancement at a glance
4. Provide evidence — preliminary results, feasibility data, or theoretical justification
5. Match novelty to the action type — RIA expects scientific novelty, IA expects application novelty
6. Don't over-claim — precise, evidence-backed claims score higher than grand visions without support

Want to see how your proposal's novelty compares to funded projects? Try a free evaluation and get an AI-generated novelty assessment in under 30 seconds.