Co-Funder Networks: Who Else Funds Your Grantees

Ask a program officer which other foundations fund their grantees and you will usually get three or four names — the ones that come up in conversation, sit on the same convenings, or appear in a grantee's annual report. The real number is typically an order of magnitude larger.

This is not a failure of attention. Nobody has the information. A foundation sees its own grants clearly and everyone else's not at all, because the only complete record of who funds whom is spread across hundreds of thousands of separate tax filings. Reconstructing it means parsing all of them at once and resolving several million recipient names to actual organizations.

Once that is done, something useful appears. Grants stop being a list and become a network: funders on one side, nonprofits on the other, every grant an edge between them. And networks support questions that lists cannot answer — about position, overlap, fragility, and where the money you did not give is coming from.

What a co-funder network actually is

The structure is simple. Take every grant reported in Part XV of a Form 990-PF, and every grant reported on Schedule I of a public charity's Form 990. Each record names a funder, a recipient, an amount, and usually a purpose. Treat funders and recipients as nodes and grants as edges, and you have a bipartite graph covering essentially all institutional grantmaking in the United States.

From that graph, two funders are "co-funders" when they support at least one organization in common. The strength of that relationship can be measured several ways — the count of shared grantees, the share of either portfolio that overlaps, or dollar-weighted overlap. Each gives a different and legitimate answer, which is why any serious analysis states its measure.

The construction is unglamorous and the difficulty is entirely in the details. Recipient names in filings are free text: the same organization appears as "Boys & Girls Club of Metro Denver," "Boys and Girls Clubs of Metro Denver Inc," and half a dozen other variants. Without careful entity resolution, one organization becomes six and the network is nonsense. This is why the graph is rarely built well, and why "we have the grants data" and "we have the funding network" are very different claims.

The five things the network shows you

1. Who your actual peer funders are. Not the ones you meet at conferences — the ones whose portfolios overlap with yours. These lists routinely surprise. Foundations discover regular co-funders they have never spoken to, and discover that a foundation they consider a close peer shares almost no grantees with them.

2. Where you are the only one. For some grantees, no other institutional funder appears anywhere in the data. This is the sole-funder count, and it is the single most operationally useful output of the whole exercise. It tells you which organizations depend on you entirely — a strategic fact for you and an existential one for them.

3. Where the crowd already is. The inverse: grantees with a dozen institutional funders. Not a problem, but worth knowing. A grant that is one of fifteen plays a different role than a grant that is one of two.

4. What you have not funded. Organizations tightly connected to your grantees' funders — that is, backed by the same foundations that back you — but never funded by you. This is the most immediately actionable output, because it is a concrete, evidenced list rather than an abstraction.

5. Clusters and coalitions. Groups of funders who repeatedly show up together around the same set of organizations. Sometimes these are formal collaboratives. More often they are emergent — a de facto coalition nobody convened.

Why sole-funder analysis matters more than it sounds

Of all these, sole-funder status deserves separate treatment, because it is where the network view produces something genuinely new rather than merely convenient.

An organization with one institutional funder has a specific fragility. A change in your program strategy, a leadership transition, or a bad market year at your foundation is, for them, an extinction-level event. They usually know this. You may not, because from your side the grant looks like one of many you make.

Knowing the number changes several decisions:

• Exit planning. Ending support for an organization with eleven other funders is a different act than ending support for one with none. Both may be right; they warrant different notice periods and different conversations.
• Grant type. Sole-funder relationships are where multi-year and general operating support do the most good, because the recipient has no ability to smooth across funders.
• Introductions. If you are the only institutional money, the most valuable thing you can offer may not be more money — it may be an introduction to two of your co-funders.
• Risk concentration. A portfolio in which most grantees are sole-funded means your foundation is carrying more of the sector's stability than your grant total suggests.

None of this is visible from your own records. Your filing shows what you gave. It cannot show that nobody else did.

How network position differs from size

A common assumption is that a foundation's importance tracks its assets. In network terms, that is only partly true, and the divergence is where the analysis earns its keep.

A $20m local foundation can be the anchor of an entire county's nonprofit infrastructure while a $2bn national funder is one of twenty backers for each of its grantees. Neither is better. But the local funder's withdrawal would be catastrophic and the national funder's would be absorbed, and only the network shows that.

The "discoverer" row deserves a caution. Being first money in is measurable — you can establish that a grant preceded other institutional funding. What you cannot establish is that your grant caused the later funding. The organization may have been on an upward path you correctly spotted. That is a real skill and worth knowing about, but it is an association, not a causal claim, and it should always be stated that way.

Where the data comes from and what it cannot do

Everything here derives from public filings. Private foundations itemize grants in Part XV of Form 990-PF; public charities that make grants report them on Schedule I of Form 990. Since mandatory electronic filing took effect under the Taxpayer First Act, these are published by the IRS as machine-readable XML, which is what makes national-scale reconstruction possible.

Plinth builds this graph across the full e-filing universe — 17,896,418 grants from 205,036 grantmakers, fiscal years 2017 to 2025 — with the entity resolution required to make recipient matching hold up. Funder and nonprofit pages are free to search at data.useplinth.com with no sign-up.

The limits are real and worth stating plainly:

Filing lag. Data describes fiscal years 12 to 24 months in the past. A co-funder relationship that started last year may not appear yet.

Early years are thinner. Mandatory e-filing applied to private foundations for tax years ending on or after July 31, 2020. Coverage is effectively complete from then on; before that some foundations still filed on paper, so network density in the earliest years understates reality slightly.

Individual and anonymous giving is invisible. The graph covers institutional grantmaking. A grantee funded largely by individuals will look under-supported when it is not.

Donor-advised funds obscure the source. A DAF grant appears as coming from the sponsoring organization, not the donor who advised it. For grantees with significant DAF support, the true funder set is partly hidden.

Government funding sits outside the grants graph. Federal and state dollars often exceed foundation dollars for the same organization. Plinth matches USASpending and state checkbook data to recipients separately for this reason, but it is a different dataset joined in, not part of the filing record.

Matching is imperfect. Good entity resolution handles the large majority of name variation. Nothing handles all of it.

What the network looks like from the nonprofit's side

The same graph read from the other direction answers a different and equally useful question: for any given nonprofit, who funds it, and who funds organizations like it?

This is the basis of the reverse lookup that most fundraisers actually want. Rather than searching a directory by keyword and hoping the categorization matches how you describe yourself, you identify three or four organizations genuinely comparable to yours and read off every institutional funder behind them. The resulting list is derived from funding behavior rather than from stated priorities, which is why it tends to be better qualified.

It also surfaces three things a nonprofit rarely knows about itself:

• How its funder count compares to similar organizations in the same field and geography. An organization with two institutional funders where comparable peers average seven is not necessarily doing anything wrong, but it is a fact worth knowing.
• Which funders back peers but not them. The gap list. This is the practical output.
• Which of its funders rarely co-fund with anyone. A funder that is usually the sole backer of its grantees behaves differently from one that habitually joins syndicates, and that changes how you approach it.

Foundations that share this view with their grantees tend to find it lands well. For an organization where you are the only institutional money, showing them the eight funders that back comparable organizations elsewhere in the state is genuinely more valuable than a small increase in the grant.

Using this without being creepy about it

Network data is dual-use, and the sector is right to be a little wary of it. A few principles keep it useful and proportionate.

Analyze institutions, not individuals. Funding relationships between organizations are legitimately public. Building profiles of named individuals from the same filings is a different activity with a different ethical weight.

Do not use it to rank grantees. "This organization has few funders" is not evidence of quality. It may reflect being new, being rural, or working on something unfashionable.

Share it with grantees. Most nonprofits have never seen their own funder network and find it genuinely useful. A foundation that shows a grantee the eight other funders backing comparable organizations has given them something more valuable than a small grant.

Treat overlap as an invitation, not a threat. Discovering that four foundations fund your grantees is an argument for a phone call, not for territorial behavior.

Where software fits

The external network is only half the picture. It is most useful joined to what you know internally — which grantees are struggling, which are scaling, what they asked for and did not get.

Tools like Plinth hold that internal record, so external network data can be read against live portfolio context rather than in isolation. Portfolio insights surfaces patterns recurring across grantee reports, and due diligence runs checks on applicants where knowing who else already funds an organization is directly relevant.

For foundations exploring collaborative funding, the federated giving and community foundation pages cover the operational side of shared grantmaking.

Getting the entity resolution right

It is worth understanding why this analysis is rarer than it should be, because the reason is unglamorous and it determines whether any of the above can be trusted.

Recipient names in tax filings are typed by humans into free-text fields, once a year, at hundreds of thousands of organizations. The same grantee routinely appears with and without "Inc", with ampersands and with "and", with and without a chapter or affiliate designation, abbreviated, misspelled, or recorded under a former name. Addresses are sometimes the grantee's, sometimes a fiscal sponsor's, sometimes a national office's.

Get this wrong in the permissive direction and distinct organizations merge — every "Community Action Agency" in the country collapses into one node, and the resulting network is fiction. Get it wrong in the strict direction and single organizations shatter into several, so real multi-year relationships disappear and co-funder overlap is systematically understated.

Neither failure is visible in the output. A network built on bad matching looks exactly like a network built on good matching; it is just wrong. This is why the practical question to ask of any co-funder analysis is not "how many grants does it cover?" but "how are recipients matched, and can I see the underlying filings?" An analysis that links every figure back to the specific return it came from lets you check. One that does not is asking for trust it has not earned.

Plinth's approach is to resolve recipients against the IRS exempt-organization master file where an EIN match is possible, fall back to normalized name and location matching where it is not, and cite the source filing for every displayed figure. Some residual error survives that, as it does in any system operating at this scale — which is why the sole-funder list is best treated as a strong signal to verify rather than a certainty.

Frequently asked questions

What is a co-funder?

Another grantmaker that funds at least one of the same organizations you do. Strength of the relationship is usually measured by number of shared grantees or by the share of either portfolio that overlaps.

How would we find our co-funders?

From public filings — every grant is itemized in Form 990-PF Part XV or Form 990 Schedule I. In practice this requires the whole dataset and reliable recipient matching, which is why it is usually done through a platform rather than by hand.

What does sole-funder status mean?

That no other institutional funder appears in the public data for that grantee. It indicates dependency and fragility, though it can also reflect strong individual or government support that the grants data does not capture.

Does a grantee with many funders need us less?

Not necessarily, and the inference is risky. Many funders can mean an organization is well-supported, or that it is patching together small restricted grants and would benefit most from one unrestricted commitment.

Can this show who funded an organization first?

Yes, within the filing record — you can establish that one grant preceded another. It cannot establish that the earlier grant caused the later ones.

Why do donor-advised fund grants complicate this?

Because the grant appears as coming from the sponsoring organization rather than the individual donor who recommended it. The advising donor is not identifiable from the filing.

Is this data free?

The underlying IRS filings are public and free. Plinth's funder and nonprofit pages are free to search without an account; deeper analysis, bulk API access, and people or asset data sit on paid tiers.

Recommended next pages

• Foundation Portfolio Analysis — reading your own portfolio first
• What Your Form 990-PF Reveals — the filings the network is built from
• How to Benchmark Your Foundation — position relative to peers
• How to Research a Foundation — the grantseeker's use of the same data
• Grant Management for Community Foundations — collaborative and pooled funding

Last updated: August 2026