Who this is for:This playbook is built to run yourself, with the free checker below and a self-serve plan. Prefer's done-for-you Managed service fits larger national nonprofits and foundations with real marketing budgets; a smaller, volunteer-run nonprofit is best served running these plays in-house.
Why AI search decides nonprofits outcomes
$617.2B
US charitable giving in 2025
Giving USA 2026 (IU Lilly Family School of Philanthropy) put total US giving at $617.2 billion, the first year above $600 billion. It sizes the pool nonprofits compete for.
245,000+
charities rated by Charity Navigator
Charity Navigator's own About page reports more than 245,000 charities rated. It is the most-cited evaluator in our pull and the reference AI uses for 'is this charity legitimate'.
$394.2B
of giving came from individuals
Giving USA 2026 estimated individual giving at $394.2 billion, about 64% of the total. Individuals are exactly the donors who now ask an AI where to give.
3 of 5
top sources were evaluators, not charities
Across our four 'best charities' prompts, three of the five most-cited domains were charity evaluators, not any nonprofit's own site (Prefer, Sept 2026).
The buyer prompts that decide nonprofits
| Prompt family | An example buyer asks | What wins the citation |
|---|---|---|
| Cause and categoryDonor wants options for a cause | best charities for veterans | Strong, current evaluator profiles plus a cause page AI can lift |
| LegitimacyDonor is checking before they give | is [charity] legitimate | High evaluator scores and a transparent financials and impact page |
| EfficiencyDonor is comparing overhead | how much of my donation goes to programs | Program-ratio and impact data stated plainly and confirmed by evaluators |
| ComparisonDonor is choosing between two | [Charity A] vs [Charity B] for [cause] | Honest impact and efficiency data, corroborated by evaluators |
| Discovery and taxDonor wants a deductible, timely option | best charities to donate to for a tax deduction | 501(c)(3) status stated plainly and current cause pages |
| RatingDonor wants the number | what is [charity]'s Charity Navigator rating | Claim and complete the evaluator profile so the rating is right |
Example prompts are illustrative of each family; run your own category, rivals and personas to build the real set.
Which engines matter for nonprofits, and why
- Primary
ChatGPTThe donor-research conversation ('who should I give to, and are they legit') is a natural ChatGPT flow, and our pull was run on ChatGPT and came back dense with evaluator sources. - Secondary
Google AI Overviews'Best charities for [cause]' and 'is X a good charity' trigger AI Overviews built on the same evaluators, so the groundwork carries over from Google. - Minor
PerplexityCites evaluators cleanly for the research-minded donor who wants to click through and verify a rating. - Minor
GeminiRides Google's index of the same evaluator ecosystem, so a strong evaluator and entity record shows up here too.
Who AI reads for nonprofits answers
- charitynavigator.orgEvaluatorThe most-cited source in our pull. Its rating and financial breakdown are what AI quotes on legitimacy and efficiency prompts.
en.wikipedia.orgReferenceCarries charity entities, history and any controversy. A clean, well-sourced entry is how a model learns what your nonprofit is.
reddit.comCommunity'Where should I donate for [cause]' threads. A nonprofit people vouch for on Reddit gives AI a corroborating signal.- charitywatch.orgEvaluatorGrades charities on efficiency and governance. A strong grade is a citation asset on comparison and efficiency prompts.
- givewell.orgEvaluatorEffectiveness-focused and cause-specific. It carries weight on 'most effective charity for [cause]' prompts.
- give.orgStandardsBBB Wise Giving Alliance accreditation. Meeting its standards is a trust signal AI can cite.
- Government charity registriesAuthorityState attorney-general charity registries and the IRS Tax Exempt Organization Search corroborate that you are real and in good standing.
These source patterns trace to our four-engine AI Citation Study; re-check them as the category moves.
nonprofits-specific moves
- Own your evaluator profiles firstCharity Navigator, CharityWatch, GiveWell and give.org decide charity discovery. Claim each profile, submit current financials, and get your rating as high as your real numbers allow.
- Publish a transparent financials and impact pageState your program ratio, your outcomes and where the money goes, in plain text a model can quote for efficiency and legitimacy prompts.
- Keep your Wikipedia entity and 501(c)(3) status clearA clean, well-sourced Wikipedia entry and a plainly stated tax-exempt status are how AI confirms you are real and deductible.
- Show up in the cause threads and roundups AI readsReddit cause threads and 'top charities' roundups are cited for cause prompts. A genuine presence there enters answers your own site cannot win.
Charity discovery has moved to AI, and the answer to “where should I donate for [cause]” names a few nonprofits with a reason each. Prefer (our product) shows whether you are one of them: the free checker below gives you a baseline, and a self-serve plan tracks your cause prompts on ChatGPT, Gemini, Perplexity, Google AI Overviews and AI Mode. The sections above show where those reasons come from, and the pattern is unusual for this segment: AI leans on charity evaluators and community sources far more than on any nonprofit’s own website. Your visibility is downstream of your evaluator profiles, so most of the work below happens off your domain. The plays follow, in the order a nonprofit starting from low visibility should run them.
The plays that win charity recommendations#
Claim and strengthen your evaluator profiles
Claim and complete your Charity Navigator, CharityWatch, GiveWell and give.org profiles, and submit current financials, because these evaluators are the sources AI reads first for charity answers.
AI trusts the evaluator, not the nonprofit. In our 'best charities' pull, three of the five most-cited domains were Charity Navigator, CharityWatch and GiveWell, and Charity Navigator rates more than 245,000 charities. Your score and financial breakdown there are what a model quotes when a donor asks if you are legitimate or efficient. (Charity Navigator, About, checked 2026-09-05)
- Claim your profile on Charity Navigator, CharityWatch and give.org, and apply to GiveWell where your cause fits its criteria.
- Submit your most recent audited financials and Form 990 so the ratings reflect current numbers.
- Complete every field: mission, programs, leadership, governance and impact metrics.
- Fix anything dragging your score, then keep the profiles current each year.
Done when: Your profiles on the major evaluators are claimed, complete and showing current financials.
Verify it worked: Ask 'is [your nonprofit] a good charity' in ChatGPT and Perplexity and check whether the rating it reports matches your evaluator profiles.
Common failure mode: An unclaimed or stale profile with old financials. Engines quote the evaluator literally, so an out-of-date score becomes the answer about you.
Publish a transparent financials and impact page
Publish a financials and impact page that states your program ratio, outcomes and governance in plain, quotable text, so efficiency and legitimacy prompts have something to cite.
Donors ask AI how much of a gift reaches programs, and the answer is lifted from whatever states it plainly. Evaluators corroborate the number, but a clear impact page on your own site gives the model a first-party source to quote alongside them, and transparency is the signal both evaluators and AI reward.
- State your program, administrative and fundraising ratios in plain sentences, not only in a chart.
- Report outcomes with real numbers (people served, results delivered), not adjectives.
- Explain your governance and where the money goes, in text a crawler can read.
- Add FAQPage schema to the efficiency and impact questions donors actually ask.
Done when: Your program ratio, outcomes and governance are stated in crawlable text with a schema-marked FAQ.
Verify it worked: Ask 'how much of a donation to [your nonprofit] goes to programs' and check whether the answer can state your ratio.
Common failure mode: Burying the numbers in an annual-report PDF. Crawlers skip most PDFs, so the model uses the evaluator's figure or none at all.
Fix your Wikipedia entity and legitimacy record
Keep a clean, well-sourced Wikipedia entity and state your 501(c)(3) status plainly, so AI can confirm you are real, legitimate and tax-deductible.
Wikipedia was the second most-cited source in our pull, and legitimacy prompts pull from registries and the entity record. A clear Wikipedia entry, a plainly stated tax-exempt status, and a listing in the IRS Tax Exempt Organization Search give AI the corroboration it needs to name you without hedging. (Prefer AI Citation Study, checked 2026-09-05)
- Make sure your nonprofit has an accurate, well-sourced Wikipedia entry with independent citations.
- State your 501(c)(3) status and EIN plainly on your site so 'is [nonprofit] tax deductible' can be answered.
- Confirm your listing in the IRS Tax Exempt Organization Search and your state charity registry is correct.
- Correct any outdated controversy or merger information across these sources.
Done when: Your Wikipedia entity, tax-exempt status and registry listings are accurate and agree with each other.
Verify it worked: Ask 'is [your nonprofit] legitimate and tax deductible' and check whether the answer is right and confident.
Common failure mode: Letting a thin or contradicted Wikipedia entry stand. AI leans on it heavily, so an inaccurate entry becomes an inaccurate answer.
Show up in the cause threads and 'top charities' roundups
Earn a genuine presence in the Reddit cause threads and editorial roundups AI cites, the sources that enter cause-and-category answers your own site cannot win.
For 'best charities for [cause]' prompts, AI leans on Reddit cause threads and editorial roundups more than any single nonprofit site. Reddit was the #1 source ChatGPT cited across the queries in our Reddit study and the #3 domain in our charities pull, so a nonprofit people genuinely recommend there gets a corroborating signal. (Prefer Reddit study, checked 2026-08-19)
- Run your cause and category prompts and note which Reddit threads and roundups get cited.
- Take part honestly in the cause subreddits where donors and volunteers ask for recommendations; be useful, not promotional.
- Pitch the editors behind the 'top charities for [cause]' roundups with your evaluator scores and impact data.
- Where a roundup omits or misstates you, offer the correction and a checkable reason to include you.
Done when: You are accurately present in the cause threads and roundups AI cites for your cause.
Verify it worked: Ask 'best charities for [your cause]' in ChatGPT and Google and check whether the sources cited now include you.
Common failure mode: Planted or promotional posts in cause threads. They get removed, and a recommendation that reads as astroturfed convinces neither donors nor the model.
How this fits your existing fundraising and SEO#
None of this replaces your fundraising or your existing SEO. The transparency, clear entity and real reviews that build donor trust are the same signals AI reads, so a nonprofit that is already open about its finances has a head start. The difference is where the answer is decided: AI corroborates across evaluators and community sources far more than it reads your homepage, and it rewards an accurate, up-to-date record everywhere over website copy. Because “best charities for [cause]” never had a rank, judge this work by whether it names your nonprofit.
Start with what AEO is for the full method, or see the AEO for local businesses playbook, which shares the “your own site is rarely the cited source” pattern. You can also browse every AEO-by-business-model playbook.
A worked example
Holds high evaluator scores and surfaced as an entity in our pull, the pattern of a nonprofit AI trusts.
Built its brand on transparency, exactly the signal evaluators and AI reward.
A category-leading cause brand that ranking roundups and evaluators both cite.
Named brands are public, illustrative examples of the category, not customers or endorsements.
Sources
- Giving USA 2026, US charitable giving reached $617.2 billion in 2025 (IU Lilly Family School of Philanthropy)
- Charity Navigator, About page (more than 245,000 charities rated)
- Prefer Reddit study, Reddit is the most-cited source ChatGPT names
- Prefer AI Citation Study, who gets cited in AI search (four engines)
People also ask
- How do charities get recommended by ChatGPT?
- Do charity evaluators affect AI recommendations?
- Which sources does AI use to recommend a nonprofit?
- How do donors find charities using AI?
- How is AEO different from SEO for a nonprofit?