Peer Search IS the Geography Filter

The feature we get asked for most is a geographic filter. "Show me foundations in my state." It sounds so reasonable that for a long time I had trouble explaining why we haven't built the one you're asking for.

Here is the honest answer: the request is a habit. It's a habit learned in directory tools, it's a perfectly rational habit inside those tools, and it is quietly costing you time and money.

The only motion you were ever taught

Every grant research tool you have ever used trains the same reflex. Candid trains it. Instrumentl trains it. The paper foundation directories that came before them trained it for decades. The motion is always the same: start with every foundation in America, apply facets, and narrow. Geography, giving range, program area, keyword. Shrink the haystack until the list fits on a screen.

That is the only research motion most development directors have ever been taught. It isn't wrong; inside a directory, it's the only motion available. A directory is a phone book, and the only thing you can do with a phone book is cut it down.

GrantSnag runs the other direction

GrantSnag doesn't narrow. It seeds.

You name one organization that does work like yours, and the peer search walks backwards through the grant records: which foundations already wrote checks to that organization, and to others like it. The funders who are already paying for your kind of work come back to you.

Those are opposite directions, not different features. Narrowing starts from everything and deletes. Seeding starts from one true example and expands. And someone running the narrowing habit inside a seeding tool will reach the same conclusion every time: the filters must be missing. They aren't missing. The search is pointed the other way.

Here's the part I find genuinely elegant. The seed is the one thing you already know cold, and the one thing no other tool ever asks you for. Ask a development director to describe their ideal funder and you'll get a hesitant paragraph. Ask them to name three organizations doing similar work and they'll answer in four seconds without looking anything up. You already carry the key. Directory tools just never had a lock it fit.

Headquarters is not where the money lands

Now the specific request: "foundations in my state." A headquarters filter has a defect that isn't obvious until you've read a lot of 990-PFs.

Where a foundation is headquartered and where its money lands are two different facts, and they disagree constantly. A foundation in Wisconsin can fund a Texas organization five years running. A family foundation registered in Delaware for legal reasons may give exclusively in Oregon. Filter by headquarters and you delete that Wisconsin funder from your Texas prospect list, along with every other out-of-state foundation that has already proven it will write checks where you live.

We looked at this pattern across our data in an earlier post: whole states export billions more in foundation money than they receive. Texas alone exports roughly $4 billion a year. That money lands somewhere, and the somewhere is not the headquarters.

What matters is recipient geography, not funder geography. The question isn't "which foundations live near me." It's "which foundations send money to where I am." Those lists overlap far less than you'd hope.

Filters are delete operations

There's a bigger principle underneath, and it's worth saying plainly: a filter is a delete operation. Every checkbox you tick removes rows.

Your job as a development director is to find every funder worth an ask. Against that job, what does "hide the national funders" improve? Nothing; national funders write checks to local organizations every day. What does "hide the out-of-state funders" improve? Nothing; that's where much of the money comes from. There is no version of those checkboxes that makes your prospect list better. A filter that only ever removes qualified prospects isn't a feature. It's an anti-feature wearing a feature's clothes.

That said, the request isn't coming from nowhere, and I don't want to wave it off. Underneath "give me a geography filter" is a real and legitimate need: a priority signal. A foundation that gives only in your metro has a better hit rate for you than a foundation that gives in all fifty states, and you should be able to see that before you click. That belongs on the funder card as a label: this funder concentrates here, that one gives nationally. Information you can rank by, not a checkbox that deletes rows. One tells you where to start; the other decides, on your behalf, who you never see.

Four months of doing this by hand

I know the backwards search works because it's how I found funders before GrantSnag existed, at the cost of a small piece of my sanity.

I was the development director at a nonprofit serving children with disabilities in Virginia. I started the way I had been taught: search foundations by mission alignment, read the guidelines, build the list. It was slow, and worse, it was slow in a way that hid its own failure. I ended up with a long list of foundations that each needed more research, and most of that research ended the same way: they don't fund what we do. Mission statements are written to sound universal. Matching ours against theirs produced prospects that looked good on paper and went nowhere.

What worked was flipping it. I picked organizations doing similar work or serving the same families and dug through their funders' 990-PF filings by hand to find out who was already writing them checks. It took four months of manual 990 reading. And it produced so many qualified prospects that I burned out on the applications before I ran out of funders to apply to.

That's the origin of GrantSnag in one sentence: the backwards search was the thing that worked, and it should not take four months.

The logic hasn't changed since. A foundation that has already written a check to an organization like yours, in your region, has told you something no guidelines page can: it funds your category, in your geography, with actual money. A mission statement is a hope. A grant record is a receipt.

The practical version

So here's the whole method, and it's shorter than the argument for it:

1. Name a peer. An organization doing work like yours, or serving the same people. You already know three.
2. Seed the search with it and let the funders who already pay for that work come back to you.
3. Then filter by where the money landed, not where the foundation gets its mail.

You've been trained for years to start with everyone and cut. Start with someone instead. The check has already been written; you're just following it back.