How to Write a Grant Application That Doesn't Get Binned in Round One

Most grant applications fail in the first sift. Not because the business is bad, but because the application does not answer the questions the assessor is actually scoring against.

Grant bodies publish their assessment criteria. They tell you exactly what they are looking for and how many points each section is worth. The founders who win grants treat the application like an exam with a mark scheme – because that is exactly what it is.

It’s a scoring exercise, not a pitch

Grant assessors are not investors. They are not looking for excitement or upside. They are civil servants, academics, or industry experts working through a stack of 50-200 applications with a rubric in front of them.

Each section of your application gets a score. The scores are added up. The top-scoring applications get funded. Everything else gets a polite rejection.

We score every application against published criteria. If an applicant doesn’t address a criterion, they get zero for that section – no matter how good the rest of the application is. – Innovate UK assessor guidance

This means a brilliant application that misses one criterion can score lower than a mediocre application that addresses all of them. Completeness beats brilliance.

Read the mark scheme before you write a word

Every major grant programme publishes its assessment criteria. For Innovate UK Smart Grants, the criteria are:

1. Innovation (weighted) – What is genuinely new? Not ‘we’re using AI’ but what specific technical or commercial advance are you making?

2. Market awareness (weighted) – Who will buy this, how big is the market, what’s your route to market?

3. Project plan (weighted) – What will you do, when, with what resources? Is the timeline realistic?

4. Team capability (weighted) – Can this team actually deliver? What evidence supports that?

5. Value for money (weighted) – Is the budget justified? Are costs reasonable for the proposed work?

For US programmes like SBIR/STTR, the criteria shift to: scientific/technical merit, commercial potential, team qualifications, and budget justification.

The structure of your application should mirror the structure of the criteria. If they score five things, your application has five clearly labelled sections addressing those five things. Do not make the assessor hunt.

The language problem

Grant applications have their own vocabulary. Using the wrong register signals you have not read the guidance – or worse, that you have copy-pasted from an investor pitch.

Words that work in grant applications:

• “Novel” (not “disruptive”)
• “Demonstrable market need” (not “massive TAM”)
• “De-risking” (not “scaling fast”)
• “Knowledge transfer” (not “competitive moat”)
• “Measurable outcomes” (not “hockey-stick growth”)

The easiest way to calibrate your language is to read what has already won. The Innovate UK funded projects database lets you search by programme and read descriptions of what got funded. For US programmes, SBIR.gov publishes abstracts of every awarded project. Three examples from your target programme will give you a feel for the register, the level of detail, and the structure assessors expect. Your application should sound like it belongs in that stack – not like it wandered in from a pitch competition.

The applications that score highest read like they were written by someone who has read 20 other successful applications from the same programme. They match the register, the structure, and the level of detail the assessors expect. – Grant writing consultant, Research Professional

What funded applications got right

Patterns from published successful applications across Innovate UK, SBIR, and EIC programmes:

They quantified everything. Not “significant market opportunity” but “addressable market of $340M based on 12,000 target companies spending average $28,000/year on this problem (source: IBISWorld 2023).” Every claim has a number. Every number has a source.

They showed the work plan as a Gantt chart or milestone table. Not prose describing what they will do. A visual timeline with: work packages, deliverables, decision gates, and dependencies. Assessors can see at a glance whether the plan is realistic.

They addressed risk explicitly. A table: risk, likelihood, impact, mitigation. Not buried in prose – a standalone section that shows you have thought about what could go wrong and have a plan for each scenario.

They matched the budget to the work plan. Every cost line traces back to a specific work package. If the plan says “Month 3-6: user testing with 50 participants” then the budget shows the recruitment cost, incentive payments, and analysis time. No unexplained lump sums.

They included letters of support from potential customers or partners. Not generic “we think this is a good idea” letters. Specific commitments: “We would trial this product with our 200-person sales team if it achieves the stated performance targets.” Named organisations, named individuals, specific conditions.

The single biggest differentiator between funded and unfunded applications at the same quality level is specificity. Funded applications are concrete. Unfunded applications are abstract. – UKRI feedback analysis

Common mistakes that cost you points

Innovate UK Smart Grants score each criterion 1-10. Applications need 7+ on every section to be fundable. In competitive rounds, the difference between a 7 and a 6 on a single criterion is the difference between funded and rejected – there is no averaging across sections. (Innovate UK scoring guidance)

These are the most common point-leakers:

• Exceeding word counts. Assessors stop reading at the limit. Everything after is invisible.
• Answering a different question. The criterion says “market awareness” and you write about your technology. Zero marks for that section.
• Vague timelines. “Phase 1: months 1-6” with no detail. What happens in month 1 vs month 6?
• No evidence of customer demand. “We believe there is strong demand” with no LOIs, no survey data, no pilot results.
• Budget without justification. “$50,000 for marketing” – what marketing? Which channels? What conversion assumptions?
• Ignoring the “so what” question. You describe what you will build but not why it matters. What changes in the world if this succeeds?

Grant applications are scored, not judged. The assessor has a rubric, a stack of applications, and limited time. The founders who win grants write for that reality – criteria-matched structure, quantified claims, explicit risk management, and budgets that trace back to work plans. Treat it like an exam with a published mark scheme, because that is exactly what it is.