Candid Foundation Directory Alternatives: An Honest Comparison (2026)

Candid's Foundation Directory is the default. For most American nonprofits, "doing funder research" means searching it, and for good reason: it is the most comprehensive curated directory of US grantmakers that exists, it has been built over decades, and Candid is a nonprofit doing genuine public-good work with the data.

So the useful version of this article is not "here is why Candid is bad." It is not. The useful version is narrower: Foundation Directory is one kind of tool — a curated, searchable directory with editorial enrichment — and several distinct kinds of tool now exist alongside it. If what you need is a directory, it is probably still the right answer. If what you need is bulk analysis, network structure, or programmatic access, you are paying for the wrong shape of product.

This compares the realistic options, states prices with the date they were checked, and is explicit about which category each tool belongs to.

What Foundation Directory actually gives you

Worth being precise about, because the alternatives differ on exactly these points.

Candid Search states over 304,000 funder profiles — private foundations, community foundations, corporate giving programs, and public charities that regrant — alongside 1.9 million+ nonprofit profiles and 29 million+ grants, with recipients and IRS filings, updated with new grants and RFPs. On every measure that counts records, that is a larger corpus than any of the alternatives below, including Plinth's.

The important part is the enrichment. Candid adds things the filings do not contain:

• Application guidelines and deadlines, gathered from funders directly
• RFPs and open opportunities, which do not appear in tax returns at all
• Contact details and program staff, curated rather than derived
• Editorial categorization of funder interests, applied consistently by humans

That last point is the real product. A tax filing tells you a foundation gave $50,000 to an organization for "general support." It does not tell you the foundation accepts unsolicited proposals in March. Candid's Foundation Maps geographic visualization is also genuinely strong — arguably the best in the sector.

Candid also runs a substantial access program: organizations with under $1 million in revenue or operating expenses can receive Candid Premium free by earning a Gold Seal of Transparency, and Foundation Directory is available free at hundreds of Funding Information Network partner libraries. Anyone writing about cost should say so.

Pricing, as of August 2026

Prices change, and most vendors in this market revise them without announcement. The figures below are as published or reported in mid-2026 and should be verified directly with each vendor before you buy.

Two notes. Candid's Foundation Directory and GuideStar are being consolidated into a unified "Candid search" platform, so tier names may shift. And Instrumentl's published plans vary by billing term and project count more than a single figure conveys — check current plans directly.

The three categories, and which question each answers

Comparing these tools on price alone produces bad decisions, because they are not the same kind of thing.

Category 1 — Free 990 lookups. ProPublica's Nonprofit Explorer and Grantmakers.io index IRS e-filed returns and let you pull any organization's filings quickly. Excellent for "show me this foundation's 990." Not built for discovery, and they add nothing beyond what the IRS published.

Category 2 — Curated directories and prospecting platforms. Candid Foundation Directory, GrantStation, and Instrumentl. These add human-gathered information — deadlines, guidelines, RFPs, contacts — and, in Instrumentl's case, workflow for tracking applications. You are paying for editorial work and, increasingly, for matching.

Category 3 — Full-universe analytical datasets. Cause IQ and Plinth. These parse the entire e-filed corpus and expose it for analysis rather than browsing. You are paying for completeness, structure, and query access rather than curation.

Most people asking "what's a cheaper Candid?" are in category 2 and looking for a category 2 answer. Some are actually doing category 3 work — cohort analysis, network mapping, bulk exports — and have been trying to do it in a directory, which is why it felt slow.

Where a curated directory is the right answer

Be honest about this before considering anything else. Foundation Directory or a comparable directory is the better tool when:

• You are a fundraiser building a prospect list and need application deadlines and guidelines
• You want RFPs and open opportunities, which do not exist in tax filings
• You need contact names for program staff
• You want consistent human categorization of funder interests
• Your organization qualifies for free Premium access, in which case the cost question largely disappears
• You have a nearby Funding Information Network library

If most of those apply, the rest of this article is not for you. Renew.

Where a directory is the wrong shape of tool

The limits are structural rather than deficiencies. A curated directory is optimized for looking up one funder at a time.

Curation means selection. 304,000 curated funder profiles is a large number, and it is drawn differently from the full e-filing universe. Plinth's dataset covers 209,000 grantmakers that actually made traceable grants, drawn from a broader base of roughly 793,000 organizations in the filing corpus. These are different counts of different things — profiles versus active grantmakers — and neither is "more." But a curated set is by definition a chosen subset, and small or newly active funders can lag.

Search is not analysis. Directories answer "which funders match these criteria?" They are not built to answer "what share of housing funding in this metro went to organizations under $1m revenue, by year?" That is a query, not a search.

Relationships between funders are mostly absent. Which grantmakers repeatedly fund the same organizations, which nonprofits have exactly one institutional funder, which funder clusters exist — these require the whole graph. See Co-Funder Networks.

Programmatic access is limited or expensive. If you want to pull data into your own analysis, a directory UI is the wrong interface.

Cost scales badly for occasional use. A consultant who needs deep data on twelve foundations a year is paying subscription rates for burst usage.

How Plinth's funder analysis differs

Plinth's public dataset is category 3, and it is worth being clear about what that means — including where it is worse.

It reads the full e-filed universe: roughly 19.2 million grants across 209,000 grantmakers, fiscal years 2017 to 2025, plus federal spending from USASpending and nine state checkbooks matched to recipient organizations, and grants to overseas entities with no US EIN. Of those grants, 12.9 million are resolved to a named recipient organization; the rest carry a recipient name the resolver will not stake an EIN on, and are published as such rather than quietly dropped. Every figure links to the filing it came from.

What it does that a directory does not:

• Reconstructs the funding graph. Co-funder relationships, sole-funder counts, funder clusters, and look-alike organizations — computed across all filers rather than looked up per funder.
• Exposes a real API, and one shaped like Candid's. Request parameters and response field names follow Candid's Grants API vocabulary, so a Candid client is close to drop-in: transactions, funders, recipients and summary all map across, POST /essentials matches Candid's Essentials search rather than being collapsed into a single-record fetch, and Plinth's own extras ride in an x_plinth object so they never collide with the contract. Nothing is dropped in silence — a parameter Plinth cannot honor comes back as a warning, and strict=1 turns it into a 400 so a migration test suite fails in CI rather than in a board pack. A published OpenAPI 3.1 spec is generated from the service's own routes.
• Screens compliance. IRS Business Master File status, Pub 78 deductibility, automatic revocation, OFAC sanctions (both the SDN and Consolidated lists, matched on alias names as well as primary), IRS Internal Revenue Bulletin §170 deletions, and California registration against both the Attorney General's Registry and the FTB revoked list. Bulk screening takes up to 100 EINs per call. Where a source cannot be reached the answer is "not screened" rather than "clear".
• Answers questions in natural language, and takes raw SQL. An AI analysis surface writes its own SQL against the dataset and shows the query behind every figure, so answers are checkable rather than trusted; a SQL endpoint lets you skip the chat and query the corpus directly. There is also an MCP connector, so Claude can query the dataset as a tool inside a normal conversation.
• Includes government funding. Federal and state dollars flowing to the same nonprofits, which frequently exceed foundation grants and change how you read a funder's contribution.
• Reads governance and assets. Officers, trustees and the board interlocks between organizations, plus investment holdings and managers, all from the same filings.
• Is free to search. Funder and nonprofit pages require no account — around 209,000 funder pages and 327,000+ nonprofit pages.

What it does not do, and where Candid is straightforwardly better:

• No RFPs or open opportunities. Filings are historical. If you need to know what is open now, this is the wrong tool.
• No application deadlines or guidelines, except where they appear in Part XV of a filing.
• No curated contact lists.
• Filings lag 12 to 24 months. For an active fundraiser working this quarter's deadlines, that is a serious limitation.
• Less editorial categorization, since classification is derived rather than hand-applied.

The honest summary: if you are a fundraiser chasing deadlines, a directory is the better buy. If you are doing analysis — benchmarking, portfolio work, network mapping, market sizing — a full-universe dataset does things a directory cannot, and the free tier means you can establish which camp you are in before spending anything.

Questions to ask before switching anything

Subscription decisions in this market are made badly because the trial period rarely tests the thing that matters. A few questions cut through it.

What is the actual question you ask most often? Write down the last ten research questions you had. If eight begin "which funders...", you need a directory. If eight begin "how much..." or "who else...", you need an analytical dataset.

How much of the subscription do you use? Directory subscriptions are frequently renewed on the strength of two weeks of heavy use during proposal season. That may still be worth it — but price it against burst usage, not against the calendar.

Do you qualify for free access? Between Candid's Gold Seal program for organizations under $1 million and the Funding Information Network library partnerships, a meaningful share of small nonprofits are paying for something they could have at no cost.

Can you export? If your workflow ends in a spreadsheet regardless, export limits matter more than search features. Check the cap before you buy, not after.

Who else will use it? Per-seat pricing changes the comparison substantially for teams, and the tools differ in how they handle multiple users.

What happens to your saved work if you cancel? Prospect lists, tags, and notes built inside a platform are often the real switching cost, well ahead of the data itself.

A decision path

1. Do you need open opportunities and deadlines? Yes → curated directory. Check whether you qualify for free Candid Premium first, or whether a Funding Information Network library is nearby.
2. Do you just need to read specific 990s? → ProPublica Nonprofit Explorer or Grantmakers.io, free.
3. Do you need to track applications through a pipeline? → a prospecting platform with workflow, or grants management software if you are the one giving.
4. Are your questions about patterns, cohorts, or relationships across many funders? → a full-universe dataset with query access.
5. Are you a foundation rather than a grantseeker? → most of this category is built for the other side of the table; see Grant Management Software for US Foundations.

Many organizations end up with two tools: a free 990 lookup or analytical dataset for research depth, and one paid directory subscription for deadlines. That is usually cheaper and better than forcing one tool to do both jobs.

Why "the data is free" is true and misleading

A recurring argument in this market is that paid grant databases are unnecessary because IRS filings are public. The premise is correct and the conclusion usually is not.

The filings genuinely are free. Anyone can download the raw XML from the IRS. What that gets you is several hundred gigabytes of structured tax returns in which recipient names are free text, purposes are inconsistently described, and the same organization appears under a dozen spellings. Turning that into something answerable requires parsing every schema version the IRS has published, resolving millions of recipient names to actual organizations, classifying purposes, and joining the result to the exempt-organization master file.

That work is the product. Paid tools charge for it in one of two ways:

• Editorial interpretation — humans reading, categorizing, contacting funders, gathering guidelines and deadlines. This is Candid's model, and it produces information that does not exist in the filings at all.
• Computational interpretation — parsing, entity resolution, and graph construction at scale, exposed through search, query, or API. This is what analytical datasets charge for.

Both are legitimate answers to the same underlying problem, and they produce different things. The editorial model gives you information the filings do not contain. The computational model gives you completeness and the ability to ask questions across the whole population.

What neither can do is make the filings current. Every tool in this market, paid or free, is working from returns that describe fiscal years 12 to 24 months in the past. A vendor implying otherwise is describing its RFP feed, not its filing data.

Frequently asked questions

Is there a free alternative to Candid Foundation Directory?

For raw filings, yes — ProPublica's Nonprofit Explorer and Grantmakers.io are free. Plinth's funder and nonprofit pages are free to read with no account, and its grants API is free with a key at 50 calls a day. None of these include RFPs, deadlines, or curated contacts, which is the main thing Candid's subscription buys. Candid also has free routes of its own: free registration, two of its APIs at no fee, and a free year of Premium for nonprofits under $1 million in revenue that earn a Gold Seal of Transparency.

Can I get Candid Premium for free?

Organizations with under $1 million in revenue or operating expenses can receive Candid Premium free by earning a Gold Seal of Transparency. Foundation Directory is also available free at Funding Information Network partner libraries.

How much does Candid Foundation Directory cost?

As published on Candid's own pricing page and checked in August 2026: Premium at $219/month, or $183/month billed as $2,199 annually; Ultimate at $208/month billed as $2,499 annually. No free tier and no Enterprise price is listed on that page. Third-party write-ups circulating lower annual figures are roughly $1,000 out of date on each tier. Verify directly before purchasing, as Candid is consolidating its platforms.

Is Instrumentl a Candid alternative?

Partly. Instrumentl overlaps on funder discovery and adds application tracking, and third-party reviews generally place it at the higher end of the market and best suited to larger nonprofits and grant consultants. It is a different balance of discovery and workflow rather than a cheaper directory.

Why do grant databases cost money if 990 data is free?

Because raw filings are unusable at scale without parsing, entity resolution, and enrichment. Paid tools charge for interpretation — either editorial curation, as Candid does, or computational structure, as analytical datasets do.

Which tool is best for prospect research consultants?

Usually two: a directory for deadlines and guidelines, plus an analytical dataset for depth on specific funders. Consultants doing client-facing work also tend to need bulk export or API access, which directories rarely provide affordably.

Is there a cheaper alternative to the Candid Grants API?

Candid publishes ten APIs, priced per product — the Grants API from around $6,000 a year, Charity Check from around $2,750, Premier from around $9,900 — with a 30-day trial on request, and its Demographics and Taxonomy APIs listed at no fee. Plinth's grants API is free with a key at 50 calls a day, $75/month for the organization and compliance endpoints plus ad-hoc SQL, and $250/month for 10,000 calls a month. It deliberately uses Candid's parameter and response vocabulary, so migrating is mostly a hostname change, and strict=1 will fail your test suite on any filter it cannot honor rather than returning a plausible wrong answer. What Candid's Charity Check sells that Plinth does not is the filable PDF artifact and the PCS taxonomy; if you need either for a regulated workflow, that is the product built for it.

Can Plinth do charity check and compliance screening?

Yes, from the same public sources: IRS Business Master File status, Pub 78 deductibility, the automatic-revocation list, OFAC sanctions screening across both the SDN and Consolidated lists including alias names, IRS Internal Revenue Bulletin §170 deletions, and California registration against the Attorney General's Registry and the FTB revoked list. Bulk screening takes up to 100 EINs in a call, against Candid's 25. It is on the $75/month tier. The one deliberate difference: if a source cannot be reached, the screen reports "not screened" rather than "clear".

Does Plinth compete with Candid?

Only partially, and it is fair to say they are different products. Candid is a curated, human-enriched directory you search. Plinth is a computed, full-universe dataset you query. Neither replaces the other, and Candid's geographic mapping remains the strongest in the sector.

Recommended next pages

• How to Research a Foundation — the process the tools support
• Co-Funder Networks — what a directory structurally cannot show
• What Your Form 990-PF Reveals — what the underlying filings contain
• Grant Management Software for US Foundations — for the funder side of the table
• Best Free Grant Management Tools — zero-budget options

Last updated: 16 August 2026. Candid figures are as published on Candid's own pages and read on 12 August 2026.