Nonprofits get found in AI answers the same way brands do: by being clearly associated with a specific cause, place and approach, on their own site and in the sources AI assistants trust for questions about charities. The difference is that four audiences ask about you: donors, volunteers, people who need your services and grantmakers. This playbook covers the prompts each group uses, the pages to build, the trust sources that matter in the sector, and how to measure progress.
The prompts that matter for nonprofits
Nonprofit prompts split by audience, and each audience needs different answers from you.
| Audience | Example prompt | What they need |
|---|---|---|
| Donor (cause) | "Best charities fighting childhood hunger in Ohio" | Your cause and place, stated plainly |
| Donor (trust) | "Is [charity] legit?" / "How much of my donation goes to programs at [charity]?" | Financial transparency, evaluator profiles |
| Donor (method) | "Can I donate stock to [charity]?" | Giving options in text |
| Volunteer | "Volunteer opportunities with animals near Austin on weekends" | Roles, times, requirements |
| Service seeker | "Free legal help for tenants facing eviction in Chicago" | Eligibility, how to apply, contact |
| Grantmaker or partner | "Nonprofits working on rural broadband access" | Programs, outcomes, geography |
| Corporate giving | "Charities that accept employee matching gifts for STEM education" | Matching and partnership info |
Weight your prompt set toward unbranded cause prompts and service-seeker prompts, then add trust prompts for your own name. See how to build a prompt set worth tracking.
The pages to build
1. Mission and impact page
One page that says, in the first two sentences, what you do, for whom, where, and how. Follow it with outcomes in numbers you can stand behind, the year they cover, and how you measure them. Vague mission language ("empowering communities") gives AI nothing to repeat.
2. Financials and transparency page
Donors ask AI about overhead and accountability. Publish your annual report and audited financials as HTML summaries, not only PDFs, with program spending, revenue sources, leadership and board. In the US, link your Form 990. State your registration number and legal name.
3. Program pages
One page per program with who it serves, eligibility, how to access it, location and hours. These answer service-seeker prompts, which are often the most urgent. See use-case pages that match the situations people describe.
4. "Get help" page
A single, plain page that routes people to the right program, with eligibility rules, languages offered and contact details in text. If you don't serve a need, say so and point to who does. AI assistants are more likely to send people to an organization whose page answers the eligibility question directly.
5. "Ways to give" and "Volunteer" pages
List every giving method (one-time, monthly, stock, donor-advised funds, matching gifts, legacy gifts, in-kind) and every volunteer role with time commitment and requirements. These are short, factual pages that answer method prompts exactly.
Technical basics
- Allow AI search crawlers; many nonprofit sites sit behind security plugins that block unfamiliar bots. See AI crawlers and robots.txt.
- Make sure donation and program information isn't locked inside a third-party widget or iframe that crawlers can't read.
- Add Organization markup, or schema.org's more specific NGO type. US 501(c)(3) organizations can also use the
nonprofitStatusproperty, which schema.org lists with values such asNonprofit501c3.
{
"@context": "https://schema.org",
"@type": "NGO",
"name": "Riverside Food Network",
"legalName": "Riverside Food Network Inc.",
"url": "https://example.org",
"nonprofitStatus": "Nonprofit501c3",
"areaServed": "Riverside County, California",
"sameAs": ["https://example.org/your-evaluator-profile", "https://example.org/your-register-entry"]
}
See Organization schema and your entity home page.
The third-party sources AI trusts for charities
For trust prompts, AI assistants lean on independent sources. Prioritize these:
- Charity evaluators and databases. In the US, Candid (formed in 2019 when Foundation Center and GuideStar merged) and Charity Navigator are the obvious starting points. Claim and complete your Candid profile and check what Charity Navigator shows. Charity Navigator's own methodology page describes ratings built from four "beacons," including accountability and finance and impact and results. Cause-specific evaluators matter in some sectors.
- Official registers. The IRS tax-exempt organization search in the US, the Charity Commission register in England and Wales, and equivalents elsewhere. Make sure your legal name and registered details match your website.
- Local and trade news. Coverage of your programs and outcomes is often the source AI cites for cause and place prompts. See digital PR for AEO.
- Partner and funder pages. Foundations, corporate partners and coalitions that list you as a grantee or member.
- Volunteer platforms. Listings on the volunteer-matching sites people actually use in your area.
- Wikipedia, only if you meet its notability rules and never written by your own staff. See Wikipedia and Wikidata.
Run the trust prompts yourself and record which sources get cited. Fix inaccurate data at the source rather than adding more content to your own site.
Cautions specific to nonprofits
- Name confusion. Many charities share similar names. State your legal name, location and registration number consistently so AI doesn't merge you with another organization or a scam.
- Outdated programs. A closed program that AI still recommends sends vulnerable people to a dead end. Remove or clearly mark closed programs and update the sources that mention them. See fixing outdated information in AI answers.
- Eligibility errors. Check that AI describes who you serve correctly, including income limits, age ranges and geography.
- Overstated impact. Only publish outcome figures you can document. AI repeats claims, and inflated ones damage trust with donors and evaluators.
Measurement
| Metric | What good looks like |
|---|---|
| Mention rate on cause and place prompts | Rising against peer organizations |
| Accuracy on trust prompts (legitimacy, overhead, registration) | Near 100% |
| Accuracy on eligibility and service prompts | Near 100% |
| Citation share for your impact and program pages | Rising |
| Donations and volunteer sign-ups from AI referrals | Tracked in analytics |
| "How did you hear about us?" answers mentioning AI | Tracked on donation and intake forms |
Add an "AI assistant" option to your donation and volunteer forms. See self-reported attribution for AI search.
Example: a regional food bank
This example is illustrative. A food bank serving three counties runs 40 prompts across donor, volunteer and service-seeker groups. It finds that AI names it for "donate to food bank in [county]" but not for "free groceries near me" prompts, where a national directory with an outdated pantry list is cited instead. One partner pantry it lists closed a year ago.
Its plan: publish a "Find food" page with every current pantry, hours and eligibility in HTML; ask the directory to update the listing; complete its Candid profile; and add a volunteer page with shift types. Two months later it re-runs the prompts, checks whether the closed pantry has dropped out of answers, and reviews the new "AI assistant" responses on its intake form.
The bottom line
Nonprofit AEO is about being described accurately and associated clearly with your cause and community. Build plain pages for mission, impact, financials, programs, getting help, giving and volunteering; keep evaluator and registry profiles complete and consistent; correct outdated information at the source; and measure accuracy on trust and eligibility prompts as closely as visibility on cause prompts.
