Bothy

Picture a development director at a nonprofit running on $400,000 a year. She is one person wearing four hats, and grant-seeking is the hat she wears on Thursdays — working from a funder list in a spreadsheet inherited two years ago that nobody has verified since.

The failure is never dramatic

A report deadline slides past on a Friday spent doing payroll, and a renewal she was counting on quietly does not renew. Nobody sends a rejection letter for that — the money just isn't there next year. That's the loss Bothy was built to remove.

The next thing due is already on the screen

That's the whole product idea. When she opens Bothy, there's no dashboard to interpret and no search box to start from — just the report due Friday, the funder she owes a reply to, the application closing in nine days.

Somebody actually reads the 990s

The part we're proudest of isn't software. Every workspace starts from a funder list built for that specific organization out of IRS filings and what funders publish themselves — twenty-five to fifty funders verified by hand, not scraped and guessed at.

It finds money that was already out there

The list isn't a directory to go searching in — it's funders whose actual giving matches the organization, most of whom she'd never heard of. Four organizations have already applied to grants they found this way and won them: tens of thousands of dollars that was sitting in reach the whole time.

65 organizations use it today

A two-person studio can live with not scaling

Bothy is built for people the studio knows and likes. The reason nothing like this existed: the only thing that makes it work is a person doing research, and that doesn't scale the way software companies need things to scale — so they don't sell it. We can live with that trade, and it's the whole reason the thing is worth anything.

It won't write the grant

That's still on her. It won't fix underfunding, and if an organization runs fewer than about ten active funders, a spreadsheet is genuinely fine. It isn't an AI inventing grant opportunities either — which is a feature, given how many funders are actively wary of AI-written applications.