Finding grants is a research job, not a search job, and treating it as a search is why offices apply to funders that were never going to say yes. The work has three parts: build a list of funders whose published behaviour matches what you do, read enough of each one to know whether you fit, and then order the list by the deadline calendar so the applications are spread across a year rather than stacked in March. Everything after that is writing.
Where the list comes from
Federal opportunities are published on grants.gov with their notices, eligibility and deadlines. Private and family foundations file a Form 990-PF that lists every grant they made, which tells you sizes, recipients and patterns rather than intentions. Your community foundation publishes its cycles. And the least used source is the best: look at organisations doing work like yours and see who funds them, which is public on their own filings and often on their website.
What to read before you apply, and the order to read it in
First the eligibility, because it is binary and it is where most wasted applications die. Then two years of actual grants, which tell you the real size and the real subject better than the priorities page. Then the priorities and the notice. If the average grant is $15,000 and you are asking for $150,000, you have learned that before writing rather than after.
Turning research into a calendar
A funder list is only useful once it has dates on it. Put every candidate's deadline on one calendar, then look at the shape: if six deadlines fall in the same fortnight you will submit three of them badly. Deciding in November which of March's deadlines you are actually going to meet is the single highest-value hour in the grant year, and it needs the deadlines to be visible somewhere other than in a folder.
Keeping what you find
Most offices research the same funders repeatedly because nothing kept the last search. Keep the funder, what it gives, to whom, in what range, on what cycle, and what happened when you last approached it. That record is worth more each year, and it is the thing this product exists to hold.
Questions people ask about how to find grants for nonprofits
Are grant databases worth paying for?
They save time on discovery and none on judgement. The reading you have to do about each funder is the same whether the name came from a database or a 990.
How many funders should be on the list?
Enough to fill the year at the rate you can actually write. Twelve well-matched funders beat sixty names nobody has read.
Should I apply to a funder that has never funded work like mine?
Rarely. A funder's grant list is a better description of its behaviour than its priorities page, and a first-of-its-kind application is a long shot you should take knowingly.