Solutions / Grant Matching Software

A ranked list of funders isn’t the same as knowing which ones actually fit.

A search feed tells you a grant exists. Matching software tells you, before you spend an hour reading the full notice, how closely it actually aligns with your mission — scored two ways: an instant keyword-overlap pass against a shared foundation directory, and a Claude-scored semantic pass with a stated reason for every funder already on your own list.

See your mission scored against real funders See how scoring works

The actual Funder Matching results

Not a mockup — a real mission statement (“emergency housing, case management, and workforce training for veterans and their families experiencing homelessness”) submitted against the live foundation directory, showing real ranked results with match percentages and the specific shared keyword behind each one.

The live Funder Matching page, showing ranked foundation results with match percentages and the shared keyword behind each score

Two scoring passes, for two different questions

INSTANT KEYWORD MATCH

“Which new funders should we look at?”

A plain keyword-overlap comparison between your mission text and each foundation’s stated focus areas across a shared, 130,000-plus profile directory. No AI call, returns instantly, and every result shows the specific keyword that produced the score.

SEMANTIC FUNDER SCORING

“Of the funders we already track, which are worth pursuing first?”

Claude reads your mission, programs, and service area against each funder’s recorded priorities and geographic focus, returning a 0-100 score with a one-sentence reason. Funders scoring below 40 are filtered out rather than padded onto the list.

Why the real results above include low scores

The screenshot above shows genuine results for a real mission statement, including entries scored as low as 0% and 5% match. Most of a foundation directory is not a fit for any single organization’s mission, and showing the full honest range, not just the top hits, is what makes a match score something you can trust rather than a list quietly curated to look better than it is.

Stated limitations

Keyword-overlap matching only sees what a foundation’s own directory listing states. A real fit whose listing doesn’t spell out its focus areas in matching language will score lower than it should.

Semantic scoring is Claude’s reasoning about likely alignment from text, not a funder’s official statement of intent. Treat it as a research aid for prioritizing outreach, not a substitute for reading the funder’s own current guidelines.

FAQ

What's the difference between grant discovery and grant matching?

Discovery finds candidate opportunities across sources you might not have checked. Matching takes a mission statement or a funder list you already have and scores how well each one actually fits, either instantly by keyword overlap or, for funders in your own list, by Claude-scored semantic alignment with a stated reason.

How is the match percentage calculated?

The instant Funder Matching pass scores by keyword overlap between your mission statement and a foundation's stated focus-area text, a plain, explainable calculation, not a black-box model. The semantic pass, run per-funder on funders already in your organization's list, has Claude score 0-100 with a one-sentence reason and filters out anything scoring below 40.

Does a high match score mean we'll win the grant?

No. Both scores measure stated alignment between your mission and a funder's stated priorities, not competitive strength, deadline timing, or funding availability that cycle. Treat a high score as a reason to read the full opportunity, not a guarantee.

Can we match against funders we've already researched, not just a shared directory?

Yes. The semantic pass scores against your organization's own funder list. The instant keyword pass searches a shared foundation directory of over 130,000 indexed profiles, useful for finding new candidates outside what you've already researched.

Does running a match cost extra AI usage every time?

The keyword-overlap pass is a plain-text comparison with no AI call, so it's effectively free to re-run. The semantic pass calls Claude per funder scored, and is metered the same way other AI-driven agent runs are on your plan.

HAVEN’T FOUND FUNDERS TO MATCH AGAINST YET?

Start with discovery

Grant Discovery Software builds the candidate list this matching pass scores against.

See Grant Discovery Software

Score your mission against real funder data

A live demo runs both scoring passes against your own mission statement and funder list.

Book a demo Create an account