Frequently Asked Questions

Everything you need to know about how we evaluate charities.

The Questions That Matter Most

Why don’t you just use overhead ratios like other evaluators?

Because overhead ratios can be misleading. An organization can have a 95% program expense ratio while doing something completely ineffective. Conversely, a legal advocacy organization might have higher administrative costs because lawyers are expensive — but win cases protecting millions of Muslims. A food bank with 92% program spending might distribute food recipients don’t need, while a civil rights organization with 75% program spending might achieve far greater impact. We measure whether programs actually work, not just how money is allocated.

Why do you need a specific evaluator for Muslim charities?

General charity evaluators often miss nuances important to our community. They don’t check for Zakat compliance, may unfairly penalize organizations working in conflict zones like Gaza or Syria, and often overlook smaller grassroots organizations that serve Muslim communities. We built this to fill that gap — and to ask: would YOUR donation make more difference here than elsewhere?

General

About Good Measure Giving

Why do you need a specific evaluator for Muslim charities?

Is Good Measure Giving affiliated with any charity?

No. We are completely independent. We do not accept payments from charities to be listed or to influence their scores. Our goal is to serve donors, not organizations.

How is this different from Charity Navigator or Candid?

Those are excellent resources that we actually use as data sources. But they focus primarily on financial metrics like overhead ratios. We go further by evaluating two dimensions: Impact (how effectively does this charity use donations to create change?) and Alignment (is this the right charity for Muslim donors?). We also provide Zakat/Sadaqah classification and a Data Confidence signal that other evaluators don’t offer.

Can I request a charity to be evaluated?

Yes! Use the ‘Suggest a Charity’ option in our feedback form, available in the footer or on the browse page. Include the organization’s name and EIN if available. You can also email us at hello@goodmeasuregiving.org. Priority is given to registered 501(c)(3) organizations serving Muslim communities.

Do you evaluate mosques and Islamic centers?

We primarily evaluate 501(c)(3) charitable organizations with standard public filings, but we do include some Islamic centers and mosque-like organizations when enough reliable data is available. Coverage is still strongest where Form 990 and third-party profile data are robust, so some congregational organizations remain harder to assess confidently.

Our Methodology

How we score charities

What is the GMG Score?

The GMG Score is our 0-100 rating built from two dimensions, each worth 50 points: Impact (how effectively does this charity use donations to create measurable change?) and Alignment (is this the right charity for Muslim donors?). We also apply risk deductions (up to -10 points) for serious concerns like very low program spending or governance problems. Alongside the score, a Data Confidence signal (High, Medium, or Low) tells you how much public data supports the evaluation. Scores above 75 indicate exceptional performance, while many organizations cluster in the middle score bands.

Why do I see qualitative labels instead of a big number?

We intentionally prioritize qualitative signals to avoid false precision. A charity now shows an archetype (what kind of organization it is), an evidence stage (Verified, Established, Building, or Limited Evidence), four signal states (Evidence, Financial Health, Donor Fit, Risk), and a recommendation cue (Maximum Alignment, Strong Alignment, Mixed Signals, or Needs Verification). The numeric score still exists and is available in collapsed methodology details, but it is no longer the primary cue for browsing.

What does “Sovereignty Builder” mean?

Sovereignty Builder is an archetype for organizations focused on Muslim civic power and representation. These groups typically work on voter engagement, policy influence, legal rights, or institution-building so communities can shape decisions that affect their lives.

What does ‘Impact’ measure?

Impact (50 points) assesses organizational health indicators that research associates with effective programs. We score the same seven components for every charity (cost per beneficiary, directness, financial health, program ratio, evidence/outcomes, theory of change, governance), but the exact weights are archetype-adjusted by charity type. Most sub-components (financial health, governance, program ratio) are organizational health indicators rather than direct outcome measurements. Where charities provide verified outcome data through independent evaluation, we weight it more heavily. Evidence quality is assessed on a five-level scale: Verified (independent third-party evaluation), Tracked (3+ years of outcome data), Measured (1-2 years of data), Reported (basic output tracking only), and Unverified (no structured tracking).

What does ‘Alignment’ measure?

Alignment (50 points) measures whether this charity is the right match for Muslim donors. The largest component is Muslim donor fit (19 points), followed by cause urgency (13 points). We also evaluate underserved space (7 points) — is this need overlooked by mainstream philanthropy? — track record and organizational history (6 points), and funding gap (5 points) — would your donation make more difference here than elsewhere? Muslim-focused charities often score higher because they serve communities overlooked by mainstream funders.

Why don’t you just use overhead ratios like other evaluators?

Because overhead ratios can be misleading. An organization can have a 95% program expense ratio while doing something ineffective. Conversely, a legal advocacy organization might have higher administrative costs because lawyers are expensive — but still deliver major impact. We include program ratio as one part of Impact, but we also score cost-effectiveness, outcomes evidence, financial health, theory of change, and governance. In short: we evaluate whether programs work, not just how spending is labeled.

How do you handle charities working in conflict zones?

We explicitly account for the higher costs of operating in places like Gaza, Syria, Yemen, or other difficult environments. Security, logistics, and compliance costs are legitimately higher in these areas. Our cost-per-beneficiary benchmarks are cause-adjusted and include conflict-zone adjustments, so organizations aren’t penalized for necessary operating conditions.

AI & Technology

How we use AI responsibly

Do you use AI to evaluate charities? How can I trust that?

Yes, we use AI to process data consistently. Here’s the division of labor: AI reads and extracts structured data from IRS 990 filings, rating agencies, and websites. Deterministic code calculates all scores — the AI never assigns point values. AI generates narrative summaries that cite specific sources. Automated validators flag conflicts between sources. What AI does NOT do: make scoring decisions (code handles that), issue religious rulings (our Zakat classifications are informational only), or invent data (every claim must cite a verifiable source).

How do you discover information about charities?

We use a technique called ‘Grounded Search’ — instead of manually crawling thousands of web pages, we search for specific answers: ‘Does this charity accept zakat?’, ‘Has this charity been independently evaluated?’, ‘What outcomes does this charity track?’ This finds information from news articles, foundation databases, and academic papers that we’d never find by just scraping the charity’s website. It’s much faster than traditional web crawling, and discovers things the charity might not advertise on their homepage.

How do you prevent AI hallucinations?

Multiple safeguards work together: (1) Every factual claim must cite a source you can verify yourself (IRS filings, Charity Navigator, etc.). (2) We maintain a ‘denylist’ of high-risk fields (zakat eligibility, cost-per-beneficiary) that get extra scrutiny. (3) After narratives are generated, specialized ‘judge’ AIs audit them — checking citations exist and support claims, testing that URLs work, verifying facts against source data. (4) If validation fails, the evaluation fails — we don’t publish fallback narratives. (5) All scoring uses deterministic code, never AI judgment.

What if the AI makes a mistake?

We have several safeguards: every factual claim cites a source you can verify yourself (IRS filings, Charity Navigator profiles, etc.), automated validators check for conflicts between sources, all scoring uses deterministic code (same data = same score every time), and community members can report errors for investigation. If you find an error, please email us at hello@goodmeasuregiving.org with specifics — we’ll investigate and update the evaluation if warranted. Our goal is accuracy, not defending AI-generated content.

Why not just have humans do all the evaluations?

Human evaluators have biases and limited capacity. A single human evaluator might be harsher in the afternoon, unconsciously favor certain types of organizations, or simply lack time to research 100+ charities thoroughly. Our approach uses AI to extract data consistently, then deterministic code to score every charity against identical criteria. This produces more consistent evaluations than human-only approaches. We’re transparent that this is an automated system — the tradeoff is scale and consistency versus the nuanced judgment a human expert might provide. That’s why we cite sources: so you can verify claims yourself.

Can I see the actual prompts you use?

Yes. We publish our core prompts and prompt annotations at /prompts — including data extraction, narrative generation, and quality validation flows. You can inspect how we instruct models and where additional prompt coverage is still being added. We currently list 14 active prompts and 16 planned category-specific calibration prompts.

Zakat & Sadaqah

Religious classifications

How do you determine if a charity is Zakat-eligible?

We use a self-assertion model: if a charity explicitly claims Zakat eligibility on their website, we classify them as ‘Zakat Eligible.’ We verify this claim using search tools to find evidence on the charity’s official website. We also note which of the eight asnaf categories (Zakat recipients) their work serves. We don’t make our own independent rulings on Zakat eligibility — that’s for scholars and the charity itself to determine.

Can I give Zakat to non-Muslim beneficiaries?

This is a matter of scholarly debate. The majority of classical scholars restrict Zakat to Muslim beneficiaries, though some contemporary scholars permit giving to non-Muslims in certain categories (like ‘those whose hearts are to be reconciled’). Our platform labels charities based on their stated policies, but we recommend consulting with a scholar for your specific madhab’s ruling.

How do you verify a charity’s Zakat fund segregation?

We check for explicit statements on the charity’s website about Zakat fund handling. A strong Zakat policy should include: (1) separate accounting for Zakat funds, (2) clear statement about which programs receive Zakat, (3) commitment to 100% Zakat reaching eligible recipients. If we can’t verify these elements, we classify the charity as Sadaqah to err on the side of caution.

Is your Zakat classification a religious ruling (fatwa)?

Absolutely not. Our classifications are informational only, based on publicly available data about the organization’s policies and programs. We report what charities claim about themselves and provide context for donors. For definitive guidance on your specific situation, please consult a qualified Islamic scholar who can consider your madhab and circumstances.

What scholarly frameworks inform your Zakat approach?

We don’t issue religious rulings—we classify based on what charities publicly claim. Our understanding of the eight asnaf categories draws from classical sources (Quran 9:60, major tafsir works) and contemporary scholarship including guidelines from the Assembly of Muslim Jurists of America (AMJA), the European Council for Fatwa and Research, and individual scholars like Dr. Yusuf al-Qaradawi’s ‘Fiqh al-Zakah.’ Where madhabs differ—such as on giving to non-Muslim beneficiaries or the scope of ‘fi sabilillah’—we note the existence of legitimate scholarly disagreement rather than adopting one position. Our role is to help you find charities that claim Zakat eligibility; your scholar’s role is to confirm whether that claim aligns with your madhab.

What are the eight Zakat categories (asnaf)?

Islamic jurisprudence identifies eight categories eligible for Zakat (Quran 9:60): Al-Fuqara (the poor), Al-Masakin (the destitute), Al-Amileen (Zakat administrators), Al-Muallafatul Quloob (those being brought closer to Islam), Ar-Riqab (freeing captives — modernly interpreted as refugees and trafficking victims), Al-Gharimeen (those in debt), Fi Sabilillah (in Allah’s path — Islamic education, humanitarian work), and Ibnus-Sabil (stranded travelers, displaced persons). When a charity claims Zakat eligibility, we note which categories their work serves.

Why is a high-scoring charity classified as ‘Sadaqah’ instead of ‘Zakat’?

The GMG Score measures overall strength across impact and alignment — while wallet classification is about religious compliance. A medical research organization or civil rights group might score very high but not fit traditional Zakat categories. This doesn’t make them less worthy — it just means you should use your Sadaqah funds rather than Zakat. Both are important forms of giving.

Should I consult a scholar before giving Zakat?

We recommend it, especially for: large Zakat payments where you want extra assurance, specific madhab requirements (Hanafi, Shafi’i, etc. may differ on details), edge cases involving mixed-purpose organizations, and family situations like giving to relatives or calculating nisab. Our classifications help you narrow down which charities to consider, but a qualified scholar can address your specific situation.

What if a charity accepts Zakat but you classified them as Sadaqah?

Our classification is based on the data we have. If an organization has a Zakat fund that we missed, please email us at hello@goodmeasuregiving.org and we’ll update our records. However, some organizations accept Zakat without proper segregation or clear policies — in those cases, we err on the side of caution and classify as Sadaqah until we can verify compliance.

Data & Accuracy

Sources and updates

Where does your data come from?

We aggregate data from 7 sources: IRS Form 990 filings (via ProPublica’s API — official legal filings), Charity Navigator (scores, accountability), Candid/GuideStar (transparency seals, outcomes), BBB Wise Giving Alliance (standards compliance), charity websites (programs, mission, Zakat policies), Form 990 grant data (who they fund), and ‘discovered’ information via Grounded Search (finding zakat claims, evaluations, and awards from across the web). When sources conflict, official IRS filings take precedence over rating agencies, which take precedence over self-reported website content.

How do you ensure evaluations are reproducible?

All our charity data lives in DoltDB — a database with Git-like version control. Every pipeline run creates a ‘commit’ with a timestamp and description of what changed. You can diff any two pipeline runs to see exactly what data changed and why. This means we can always explain why a score changed: was it new 990 data? A Charity Navigator rating update? A website change? We never lose history, and we can even ‘time travel’ to see what our data looked like at any previous point.

How often do you update evaluations?

We re-evaluate organizations when new 990 filings become available (typically annually) or when significant changes occur. If you notice outdated information, please email us at hello@goodmeasuregiving.org and we’ll prioritize an update.

What if I think an evaluation is wrong?

We welcome feedback. Use the Report Issue button on any charity page to flag specific errors, or the feedback button to share general concerns. You can also email us at hello@goodmeasuregiving.org with specifics. We’ll review the data and update our evaluation if warranted. Our goal is accuracy, not defending our initial assessments.

What if I represent a charity that’s been evaluated?

We welcome organization feedback. Use the ‘Tell us more’ link on your charity’s page or the Report Issue button to share corrections, context, or updated information. Our process: we receive your submission, review it against our data sources, and update the evaluation when warranted. Organization submissions may be reviewed before publication.

Why is a charity I know listed as ‘Insufficient Data’?

Some organizations, particularly newer or smaller ones, don’t have enough publicly available information for us to make a confident assessment. This isn’t a negative judgment — it just means we need more data. Often this improves as organizations file more 990s or achieve transparency seals from Candid.

Still have questions?

We’re here to help. Reach out and we’ll get back to you as soon as we can.