three filters, not one.

gratul starts every match with a hard filter pass. before anything else happens, each grant is checked against your geography, entity type, budget tier, and required certifications. if the funder only gives to health nonprofits in california and you are an arts nonprofit in maine, that grant never makes it into your weekly email.

this filter is strict by design. we would rather miss an edge case than send you a list of grants where most are instantly disqualifiable. in practice, hard eligibility rules eliminate roughly 90 percent of the grants that come in through our daily ingestion pipeline.

grants that pass the eligibility check get a second, smarter read. gratul compares the language and focus areas of the grant description to your project description, focus areas, and target populations. this step catches the grants that look right on the surface but fund a completely different kind of work underneath.

topical relevance is scored, not binary. a grant that is a near-perfect fit scores high and bubbles to the top. a grant that is only tangentially related scores lower and makes the weekly email only if nothing better is available.

the final layer looks at who the funder actually gives money to. gratul pulls each funder's past grantees from their recent 990 filings and compares those organizations to you. size, geography, program focus, and grantee history all feed into this score.

this is the filter that separates grants you can realistically win from grants that just look good on paper. a funder might technically accept applications from your category, but if every grantee in the last three years has had a budget ten times larger than yours, the match score reflects that.