Marici: frequently asked questions.

Is Marici a registered charity?

Yes, in the United States. Marici is a 501(c)(3) tax-exempt organisation registered under EIN 82-1536804, based in San Francisco, California, with an IRS ruling date of 1 September 2017 (ProPublica Nonprofit Explorer). Marici's own FAQ also states: “Marici Canada is registered with the Canada Revenue Agency (Charity Registration Number: 82732 5150 RR0001).” (archived 11 June 2026). This review has not independently retrieved that Canadian registration; readers can check it in the Canada Revenue Agency list of charities. Being registered is a statement about legal status. It does not substantiate any of the impact claims reviewed on this site.

When was Marici founded?

Marici's relaunched website states “ESTABLISHED 2018” (home page) and “FOUNDED 2018” (about page), both archived 11 June 2026. The IRS record gives a ruling date of 1 September 2017, and the organisation's first Form 990, for tax year 2017, was filed under the name “Take Her Back” (2017 Form 990), reporting $198,110 of revenue that year. Later filings use the name Marici. See the timeline for the name change and website history.

Is Marici the same as the "Marici Foundation"?

This site reviews the US 501(c)(3) charity registered under EIN 82-1536804, which it calls Marici (marici.org). The records cited on this site give its name as Marici, previously “Take Her Back” (2017 Form 990), and this site does not use “Marici Foundation” as its name. If you searched for the “Marici Foundation” and mean this organisation, confirm it by EIN rather than by name, because a name alone does not identify a registered charity. The registration details are in the registry information on the regulators page.

How many staff does Marici have?

Marici's home page refers to “Marici's 350+ interdisciplinary operatives” (archived 11 June 2026). Its IRS Form 990 filings report the following total number of individuals employed (Part I, line 5): 2 for 2024 (filing), 2 for 2023 (filing), 4 for 2022 (filing) and 1 for 2021 (filing). That figure counts only people employed by the US organisation. It does not count contractors, staff of partner organisations or volunteers, and the filings do not say who employs the operatives Marici describes. For 2024 the filing reports $248,332 of salaries, other compensation and employee benefits, and a separate expense line labelled “Consultants” of $1,042,324. Details: Claim 12.

Where does Marici operate?

Marici's website describes work across “five massive interconnected cities” and “32 cities across 21 countries” (about page), while its solution page lists “QUEUED 42 CITIES · 23 COUNTRIES” (solution page); both archived 11 June 2026. Its Form 990 filings for 2022, 2023 and 2024 answer “No” to the Part IV questions asking whether the organisation maintained an office, employees or agents outside the United States (line 14a) and whether it had more than $10,000 of activities outside the United States (line 14b). The same 2024 filing's supplemental schedule (Schedule O) itemises expense lines labelled with two countries: “COMBODIA” (six lines totalling $34,445) and “KENYA” (four lines totalling $11,133) (2024 filing). No other country is named in the expense lines. A public explanation of how the overseas work described on the website is organised and funded, and through which entities, would reconcile these records. Related: Claim 09.

Who runs Marici?

The officers and directors listed in Part VII of the 2023 and 2024 Form 990s (2024 filing) are: Divya Joshi, President (10 hours a week, $0 reportable compensation); Geetika Tewari, Vice President and principal officer (48 hours a week, $120,000 reportable compensation); Jethen Pandian, Treasurer (5 hours a week, $0); and Martin Sahakyan, Secretary (5 hours a week, $0). These are the names in the public filings. This review makes no claim about any individual beyond what the filings record.

Where does the money go?

For 2024 Marici reported $4,147,171 of revenue and $5,478,264 of expenses, a shortfall of $1,331,093, of which $4,626,511 was classified as program service expenses (2024 filing). The largest itemised expense line was “CONSULTANTS” ($1,042,324). Seven expense lines are labelled as grants, including “GRANT FOR PROG-CHILD PROT” ($1,018,459), “GRANT FOR PROG-PROGM EXEC” ($594,057), “GRANT FOR PROG-COMM STREN” ($557,375) and “GRANT GOR PROG-CARE &EMPO” ($542,297), totalling $2,884,100. The same filing answers “No” to the Part IV questions on grants of more than $5,000 to foreign organisations (line 15) and to domestic organisations (line 21), and names no grant recipients. The 2023 filing itemises “SUBGRANTS- SUPPORT” of $1,960,000 and “SUBGRANTS” of $245,000 (2023 filing). The full revenue and expense history since 2017 is on the claims page.

What are the main criticisms of Marici?

This site reviews specific public claims made by Marici and asks for the evidence that would substantiate them. It does not allege fraud. The current list of reviewed claims, each with its status, is below.

• Claim 01: Evidence needed: named models, vendors, benchmarks, or technical staff. Evidence located: none.
• Claim 02: Evidence needed: a tool inventory, demos, or accuracy benchmarks. Evidence located: three named example tools without descriptions.
• Claim 03: Evidence needed: city names, partner agencies, baselines, or third-party confirmation. Evidence located: none in the public record reviewed.
• Claim 04: Evidence needed: a published definition of "saved" or "rescued". Evidence located: cost-per-outcome arithmetic that does not reconcile with the audited budget.
• Claim 05: Evidence needed: an audited cost-per-outcome methodology backing the donor tiers. Evidence located: none; substantiation request unanswered.
• Claim 06: Evidence needed: a named Stanford department, principal investigator, IRB number, or Stanford communication. Evidence located: none in Stanford's public materials at the date last checked.
• Claim 07: Status: an archived capture of the relaunched donate.html (11 June 2026) shows both EINs on the same page — 85-3455940 in the trust strip, 82-1536804 in the footer tax-deductibility notice. The live site, behind a Cloudflare challenge, has not been re-checked since. Evidence needed: a public statement reconciling the two numbers.
• Claim 08: Evidence needed: audited statements, audit committee, board minutes, or annual report on marici.org. Evidence located: third-party rating already flags the same disclosure gaps.
• Claim 09: Evidence needed: a single published list naming the cities and a consistent count across pages. Evidence located: three marici.org pages, captured the same day, giving five, six, 32, 41, and 42 as counts for what is described as the same programme.
• Claim 10: Evidence needed: a published methodology reconciling this FAQ total with the figures documented in Claim 04. Evidence located: two different lifetime totals, both attributed to Marici's own materials, describing what appears to be the same programme.
• Claim 11: Evidence needed: cohort sizes, follow-up methodology and duration, and independent evaluation. Evidence located: a 0% figure with no published cohort data.
• Claim 12: Evidence needed: who employs the 350+ operatives, in what country, and under what compensation. Evidence located: an IRS filing reporting 2 W-2 employees for 2024 (2 in 2023, 4 in 2022, 1 in 2021) alongside a $1,042,324 "Consultants" expense line; the filings do not say who the operatives are or how they are engaged.

Has Marici responded?

No response from Marici has been recorded on any claim as of 29 September 2026. The open letter sets out the questions put to the organisation. If Marici responds, the response will be published on the relevant claim page.

About these answers

Figures come from Marici's IRS Form 990 filings as published on ProPublica Nonprofit Explorer and in the IRS e-file data, and quotations come from Wayback Machine captures of marici.org, all retrieved on 29 September 2026. Quotations are reproduced exactly as they appear, including capitalisation. Where Marici's statements and its filings differ, both are shown and no conclusion is drawn beyond the difference itself. Corrections are welcome through the tips page.