EIC Accelerator: a step-by-step guide to the proposal and the pitfalls that sink it

The three stages, in order

The Accelerator uses a staged process. Each stage filters on different things, so writing the whole application in one voice — usually the technical one — is a structural mistake.

• Short application: a compact form, a slide deck, and a short video pitch. Continuously open, remotely evaluated. The output is a go / no-go to prepare the full application.
• Full application: a long business plan with technology, market, commercialisation, team, finance, and a funding request. Submitted to a fixed cut-off date and evaluated remotely by expert evaluators.
• Jury interview: shortlisted companies pitch in person or online to a jury of investors and entrepreneurs, followed by questions. The jury decides funding.

Step 1 — Qualify honestly before you write

• Check eligibility: a for-profit SME (or, in some windows, a single natural person intending to establish one) established in an eligible country.
• Check maturity: the technology should be validated in a relevant environment — roughly TRL 5/6 — with a credible route to TRL 9. Applications well below that maturity are filtered out early.
• Check ambition: the instrument funds market-creating innovation, not incremental improvement of an existing product line.
• Check the funding fit: grant for the pre-commercial part, equity for scale-up. Ask for what the plan actually needs, not the maximum available.

Step 2 — Write the short application to be understood in five minutes

The short application is read fast. Lead with what the innovation is, who buys it, and why it is not possible today. Anything an evaluator has to reconstruct works against you.

• Open with one sentence a non-specialist can repeat: what it does, for whom, and what changes.
• Quantify the problem with a source, not with adjectives.
• Show validation evidence already achieved — test results, pilot customers, letters of intent — rather than promises.
• Make the deck self-explanatory. It is read without you in the room.
• In the video, have the actual founders speak. Juries look at the team from the very first stage.

Step 3 — Build the full application around a defensible business case

• Excellence: what is genuinely novel, what is the state of the art, and where the technical risk sits. Name the risk instead of hiding it, and show the mitigation.
• Impact: the market you can actually reach, the go-to-market route, pricing, unit economics, and the scale-up path. Bottom-up market sizing beats a top-down TAM slide.
• Implementation: a work plan with milestones tied to decision points, a realistic budget, and a team with the gaps identified and a hiring plan.
• Finance: a coherent multi-year model. The numbers in the narrative, the tables, and the deck must match exactly.
• Freedom to operate: patents, licences, and a short statement of what protects the position beyond IP — data, integration, regulatory approvals.

Step 4 — Prepare the interview as a separate exercise

• The jury tests the team, not the document. Rehearse with hard questions from outside your field.
• Assign who answers what, and keep answers short. Long answers read as evasive.
• Have the three weakest points of the plan pre-answered — the jury will find them.
• Know your numbers cold: burn rate, runway, cost per unit, sales cycle, and what the requested funding buys.

Common evaluator pitfalls

• Technology-heavy, market-light: pages of architecture and half a page on customers.
• Unsupported market figures: a large global number with no route from it to your first ten customers.
• No clear state of the art: claiming novelty without naming who is closest and why they fall short.
• Risk denial: a plan with no failure mode described reads as inexperienced, not confident.
• Inconsistent numbers between narrative, budget, and deck — a fast credibility loss.
• Team gaps ignored: no commercial profile in a market-creating proposal.
• Recycling a collaborative-project proposal: the Accelerator is an investment case, not a research work plan.
• Requesting funding without justification: the amount must map to specific activities and milestones.
• Ignoring regulatory or certification timelines in health, energy, and mobility markets.
• Rewriting a rejected application without addressing the evaluation summary report point by point.

If you are resubmitting

Treat the evaluation summary report as the specification for the next version. Map each criticism to a concrete change, and state in the new application what has materially changed since the last submission — new validation, new customers, new team members. A resubmission that reads as a reformatted version of the previous one usually gets the same outcome.