CharitySense Data

Inside American Muslim nonprofits

We followed one faith label through IRS filings and captured websites. Religion-related causes dominate the count; a much smaller international-affairs group carries a large share of reported revenue.

By the CharitySense Data team

Published Updated

1,934
nonprofit records currently labeled Islam
1,300
with an available full Form 990 revenue record
$324K
median revenue within those full returns
830
with a current captured website record
148
with captured evidence of accepting zakat

01/label not a census

A label, not a census

We found 1,934 nonprofit records currently classified with an Islam affiliation. This label helps us ask a focused question, but it is not a count of all American Muslim institutions. The classification may miss organizations or include an occasional mistake. Houses of worship can also be exempt from annual IRS return filing. The IRS explains that filing exception.

The denominator changes with the question. We find an available full Form 990 record for 1,300 of these organizations. That is the group in which we can compare detailed revenue and program spending. Another 830 have a captured website record, which lets us read some public statements about faith-specific giving. Those two groups overlap, but neither represents the entire labeled cohort. We keep them separate throughout.

1,934
Islam-labeled records
1,300
with a full-return context
830
with a captured website
1,806
with an IRS cause category

We also have more than one filing context for some organizations because an organization can file a different form over time. A latest available full return can therefore be older than its latest short-form notice. This matters especially when we sum dollars: the result is a cross-year view of the records we can inspect, not a single-year estimate of Muslim charitable activity.

02/count and the money

The count and the money point in different directions

The IRS cause categories classify 1,142 of the 1,934 labeled organizations as religion-related, about 59%. Education accounts for 181 more. The international-affairs category has just 57. Yet those 57 are associated with $846 million, or 36.4% of the $2.32 billion we can measure in available full returns. The religion-related group accounts for $424 million, or 18.3% of the same revenue total.

A different picture when we count organizations and dollars

Blue shows a cause’s share of all 1,934 labeled organizations. Green shows its share of $2.32 billion in the 1,300 full-return revenue records. Select a cause for the underlying counts.

Source: CharitySense analysis of current faith classifications, IRS cause categories and latest available full Form 990 revenue, read September 25, 2026. Units: percentage of organizations and percentage of reported dollars. Hover or focus for exact values; select for detail.

That gap is the article’s central finding. A story told from organization counts centers religious and educational infrastructure. A story told from full-return dollars centers a much smaller international-aid tier. Neither view is a portrait of every Muslim congregation or every dollar given by American Muslims. We explore the broader cross-faith pattern in the geography of faith and giving.

03/long revenue tail with

A long revenue tail, with a filing-year caveat

Among the 1,300 organizations with an available full-return revenue figure, the median is $324,384. The 25th percentile is $127,551 and the 75th is $831,500. Some 279 report at least $1 million. These figures describe full-return filers only; smaller organizations that last filed short forms are not folded into the median.

We summed $2.32 billion across those latest available full returns. The top 20 account for 54.2% of that total. Concentration is real within this measured group, but the years vary. Of the 1,300 full-return records, 968 are for tax years 2024 or 2025; the remaining 332 have earlier latest full returns. We should not read the $2.32 billion as revenue earned by the cohort in 2025.

The typical full-return filer and the organizations moving the most money occupy very different parts of this picture.

04/Foreign grants run through

Foreign grants run through a narrow channel

Among the 1,257 full-return filers with positive total expenses, the filings report $708.6 million in grants for foreign organizations or individuals. Relative to their $2.05 billion in total expenses, that is 34.6%. Yet only 67 of the 1,257 report a positive foreign-grant amount, or 5.3%. Five filers account for 77.5% of the foreign-grant dollars. Participation and dollar share answer very different questions here.

What the foreign-grant line measures
The IRS includes some grants routed through a domestic intermediary for a designated foreign recipient. The amount is not a map of where every program operates, and it is not the share of all Muslim charitable giving sent abroad. The Schedule F instructions explain the reporting rules. The spending denominator includes full-return filers with no positive foreign-grant line.

Detailed foreign-grant schedules add 790 rows across 61 of these filers. Their region descriptions vary in wording and are often broad. We did not turn them into country totals. New relationship records also let us inspect counterparties for 486 of the 1,934 labeled organizations, with 1,892 current relationship entries in total. Those entries may describe a financial flow or a structural relationship; they are not 1,892 grants. Our money-flow essay looks more closely at why the distinction matters.

05/What captured websites disclose

What captured websites disclose about zakat

Zakat is a religious obligation for eligible Muslims to give part of their wealth. The IRS return does not have a dedicated question about how a charity handles it. For that, we look at captured public website evidence. Among the 830 labeled organizations with a current website record, 264 have a zakat-specific evidence entry. In those entries, 148 have affirmative evidence that they accept zakat. The other entries do not provide affirmative evidence of acceptance; we cannot treat them as a count of charities that decline zakat. These are observations about captured evidence, not a count of all zakat-accepting charities.

The detail becomes more revealing inside the 148 affirmative records. Twenty-three have evidence of a standalone published policy, 28 of a partial policy and 93 only of a mention. Three show no policy found; one lacks a policy status. On fund handling, 23 explicitly describe segregation and 28 use restricted-fund language. Ninety-two do not state a separation practice in the captured material, while five have no status. Administrative-fee treatment is unstated in 130 of the 148.

We do not take an unstated website policy to mean the organization lacks a practice. The finding is about what a donor can verify from the pages we captured. The evidence is strongest when acceptance is accompanied by a written explanation of eligible uses and fund handling. This is one example of how a useful diligence file distinguishes a missing public statement from a negative finding.

07/Behind the Data

Behind the Data

We read current records on September 25, 2026. The starting group is 1,934 organizations with an Islam affiliation label. IRS cause categories are available for 1,806; the chart’s “other or unclassified” row includes the 128 without a category. Financial figures use the most recent full Form 990 available for each of 1,300 organizations, which can predate a later short-form filing. Percentiles use all 1,300 numeric revenue values, including filed zeros. Revenue shares use their $2,321,792,888 sum as the denominator.

Foreign-grant share divides $708,637,788 in positive foreign grants by $2,047,474,001 in total expenses across 1,257 full-return filers with positive expenses. Participation is 67 of those 1,257. The top-five share is measured among filers with positive foreign grants. Relationship counts refer to source organizations with a current relationship record and entries within them; they are neither unique beneficiaries nor proof of a transfer. Website counts refer only to currently captured pages matched to the starting group. “Accepts zakat” requires affirmative captured evidence; policy figures use those 148 affirmative records. Missing or unstated fields remain unknown.

This is a cross-year administrative-record analysis, with incomplete website coverage and possible classification error. It does not estimate the number of American Muslim nonprofits, the full value of Muslim giving or compliance with religious rules. See our methodology for how filing records are assembled.

Sources

Ask the research team

What would you want us to pull from the filings? Tell us, and a future piece may answer it.

08/Final Thoughts

Final Thoughts

We started with a faith classification and found two scales of activity in the same records. Religion-related organizations dominate the headcount. A much smaller international-aid group carries a large share of the measurable full-return dollars, and foreign grants are concentrated in still fewer filers. The newer website evidence brings another question into focus: when an organization says it accepts zakat, how much of its policy can a donor actually read? The answer is uneven in the pages captured so far. That is an invitation to look more closely at each organization, not a verdict on a community.

About the author

The CharitySense Data team

Research team

We read public nonprofit filings and other sources to understand how the sector works. This blog shares what we find, along with the limits of the evidence. Each article explains how its figures were measured and cites the records or primary sources behind them.