Arvacore logoArvacore

Arvacore Getting Started Guide | ERC

ERC Starting Grant: Where to Start in 2027

A practical ERC Starting Grant starting point: eligibility, host, official materials, scientific idea, Part I, Part II, support and next steps.

By Alessandro Brunetti, DPhil (University of Oxford), Founder of Arvacore | Reviewed 20 September 2026

If you have decided that an ERC Starting Grant may be relevant to you, the first challenge is often surprisingly practical: where do you begin?

The official ERC material gives the rules, templates and submission process. What it cannot do for you is decide whether your scientific idea is ready, how to turn it into a five-year programme, or what you should write first when the document is still blank.

This page is the shortest route through the Arvacore ERC resources. It is designed for a researcher who is interested in applying but has not yet built the full proposal.

Quick answer

Start in this order:

  1. Check that you are eligible and can identify a Host Institution.
  2. Read the current ERC call documents and official templates.
  3. Do not begin by filling the template. Begin with one blank page that defines the scientific problem, unknown, central question, proposition or hypothesis, decisive evidence and possible outcomes.
  4. Turn that scientific architecture into Part I / B1.
  5. Build Part II / B2 only after the scientific vision is clear.
  6. Use your Host Institution, NCP, peers and specialist review support for the parts they are best placed to challenge.
  7. Stress-test the proposal before submission rather than simply proofreading it.

The official ERC Starting Grant page confirms that applicants should identify a Host Institution, contact their National Contact Point, read the call documents, start writing and allow time for review by peers and other scientists.

Official source: ERC Starting Grant

Step 1 - confirm the basic conditions

Before investing heavily in the proposal, confirm the current eligibility window, any valid extensions, the call deadline, and whether you can secure an eligible Host Institution.

For the 2027 Starting Grant call, the ERC states that applicants may be within 0-10 years of successful PhD defence, subject to the detailed rules and possible extensions. The Host Institution must be an eligible legal entity and must provide suitable conditions for the PI to lead the research independently.

You do not need to already hold a permanent position at the Host Institution when you submit. You do need a Host Institution that is prepared to support the application and engage you if the proposal is funded.

Read next:

Step 2 - collect the official material before you write

Use the current call documents, not an old template found in a shared folder or an application from a previous year.

The official ERC Starting Grant page currently links to the 2027 Work Programme, Information for Applicants, submission and evaluation rules, and the Guide for Peer Reviewers. The current proposal structure separates the Scientific Proposal into a five-page Part I and a seven-page Part II for Starting Grants.

Part A contains the online administrative information and budget/resources sections. Part B1 contains Part I of the Scientific Proposal plus the CV and Track Record. Part B2 contains Part II plus the Funding ID.

Read next: ERC Starting Grant templates and official materials

Step 3 - start from a blank page, not the template

The template is useful once you know what the proposal is trying to say. It is a poor place to discover the scientific idea.

Before writing Part I, take one page and answer:

  1. What is the important scientific problem?
  2. Why does it matter to the field?
  3. What is genuinely unknown?
  4. What is the central question?
  5. What proposition or hypothesis will you test?
  6. What evidence would discriminate between plausible outcomes?
  7. What does each plausible result teach us?
  8. What changes in the field if the programme succeeds at its highest level?

The output should be a compact scientific architecture, not polished grant prose.

Read next: How to write an ERC Starting Grant proposal from scratch

Tool: ERC Blank-Page Builder

Step 4 - decide whether the idea is ERC-shaped

A technically strong project is not automatically an ERC project.

The question should support a five-year research programme and contain genuine scientific uncertainty. The strongest outcome should matter beyond the immediate experiment. The programme should establish a direction that is recognisably yours.

My own stress test is simple:

Could the strongest version of this programme create important work for the field for the next 10-15 years?

This is not a requirement in the ERC rules. It is an Arvacore test for the scale of the scientific vision.

Read next:

Step 5 - write Part I as the scientific argument

Part I is limited to five pages under the current structure. At Step 1, the panel sees Part I together with the CV and Track Record.

This means Part I must stand on its own as the scientific case. It should make the major open question, originality, objectives and overall research strategy easy to understand.

Do not try to compress the entire methods section into five pages. Use the space to make the scientific logic visible.

A useful sequence is:

problem -> unresolved limitation -> central question -> new proposition -> objectives -> overall strategy -> possible field-level consequence

Read next: ERC Part I / B1 guide

Step 6 - use Part II to make the programme executable

Part II is where the detailed implementation case becomes visible: methodology, work plan, risk assessment and mitigating measures.

The question is no longer only whether the idea is exciting. It is whether the programme can generate interpretable evidence within the proposed resources and team structure.

Read next: ERC Part II / B2 guide

Step 7 - build the PI, team and host case around the science

The ERC proposal is PI-led. The team and Host Institution should make the research programme possible without diluting the PI's intellectual ownership.

Think about:

  • which expertise belongs in the core group;
  • how many PhD and postdoctoral roles the scientific programme genuinely needs;
  • whether the Host Institution can provide the relevant facilities and independence;
  • how your CV demonstrates that you are ready to lead the proposed direction.

Read next:

Step 8 - get the right kind of help

Different people are useful for different problems.

  • Host Institution research office: submission requirements, institutional commitments, budget process and internal deadlines.
  • ERC National Contact Point: rules, call interpretation and national support.
  • Peers and senior scientists: scientific challenge and field-specific judgement.
  • Previous applicants or grantees: lived experience of the process.
  • External proposal support: independent review of scientific framing, structure, figures, risk, reviewer logic and proposal coherence.
  • ERCEA: specific questions about the call that require an official answer.

The science must remain yours. Good support helps the scientific idea survive the proposal format and independent evaluation.

Read next: Who can help with an ERC Starting Grant proposal?

Step 9 - use AI as a thinking aid, not a substitute for authorship

The ERC Scientific Council recognises that researchers may use AI for tasks such as brainstorming, literature search, revision, translation or summarisation. The applicant remains fully responsible for authorship, accuracy, plagiarism and scientific conduct.

Use AI to challenge assumptions, identify ambiguity, simulate an informed sceptical reader or structure an outcome tree. Avoid asking it to invent scientific claims, citations or a project that you do not intellectually own.

Be especially careful with confidential or unpublished material.

Read next: Useful AI prompts for ERC proposal preparation

Step 10 - stress-test before you submit

A final review should do more than correct language.

Ask:

  • Can a reviewer state the central question after one read?
  • Is the field-level consequence visible early?
  • Can the central proposition genuinely fail?
  • Are different outcomes scientifically meaningful?
  • Does Part I create a reason to continue to Part II?
  • Does Part II demonstrate a credible route to evidence?
  • Is the programme clearly owned by the PI?
  • Does every major resource have a scientific reason to exist?

Read next:

What Arvacore can do

No support provider can guarantee an ERC award. Evaluation is competitive and scientific judgement remains central.

Arvacore works on the parts that can be improved systematically: the clarity of the scientific question, ambition, Part I architecture, Part II coherence, figures, scientific risk, PI independence narrative, reviewer readability and interview preparation.

The goal is to reduce avoidable weaknesses and increase the chance that a strong scientific idea is understood as intended during evaluation.

Explore ERC proposal support ->

Check your proposal readiness ->

Official sources

Request a fit check

If you are deciding whether an idea is ready for ERC or another European research call, share the programme stage and central scientific question. Arvacore will confirm the smallest useful next step before any paid work starts.

Request a fit check →