Gates Foundation Grand Challenge

AI That Helps Donors Give More, Give Sooner, and Give Again to Maternal Health

MamaMatch AI removes the friction between donor intent and maternal health impact — connecting everyday givers to the right partner faster, and bringing them back with personalized context that makes every next gift count.

$91M+
Raised to date
50+
Partner organizations
192K+
Donors reached
45
Donor countries
Grand Challenge Alignment

One integrated system addressing all three areas

Most proposals address 1-2 challenge areas. This addresses all three as interdependent layers — each strengthens the others.

🔍
Area 1 — Understand

AI-Powered Donor Matching

The gap: Donors want to give but feel paralyzed by 50+ organizations — most leave without giving.Helps donors navigate a complex, multi-layered cause. AI identifies donor preferences — country, type of support, and urgency — and matches them to 1-3 partners whose work resonates. Turns "I want to help mothers" into "Here's the organization doing exactly what you care about."

🤝
Area 2 — Convert

Personalized Pathways to Give

The gap: Donors on a partner page have intent but not enough conviction — last-mile hesitation is where donations die.Establishes a clear, simple pathway from motivation to meaningful contribution. AI-driven personalization across channels — seasonal campaigns, donor re-engagement, targeted outreach — makes the act of giving easier for everyday donors across different countries and communities.

📊
Area 3 — Infrastructure

Impact Data

The gap: Giving is often a one-time action. Donors lack clear pathways to deepen their impact, access information, and maintain an ongoing connection to the mothers they support.Based on data on where donations went, combined with public health data models, our AI solution will provide the donor with impact estimates and suggestions on how to make deeper, transformational change.

Data Sources

Three sources that power MamaMatch AI

Every Pregnancy Platform
Proven demand at scale
192,000+ donors across 45 countries, 15+ program countries, 4+ years of Ramadan giving baselines and conversion data.
The platform, donors, and giving history already exist. The grant funds the AI layer on top.
54+ Partner Organizations
Coalition knowledge base
Partner profiles, countries served, program types, and interview data Every Pregnancy has already collected — the coalition's own knowledge, ingested once and structured for AI retrieval.
Open Public Health Data
The evidence layer
IHME Global Burden of Disease — Impact Engine
Country-level cause-specific mortality rates that power all impact calculations and partner MMR rankings. The same source used by GiveWell, MSI Reproductive Choices, UNFPA, and the Gates Foundation. Free, annually updated, no license required.
LiST Methodology — Effectiveness Coefficients
The Lives Saved Tool (Johns Hopkins) is the field standard for estimating intervention impact — used by UNICEF, WHO, and governments. We apply its RCT-validated coefficients to what partners already report. No new data collection from partners required.
UNICEF + SESRIC — Donor-Facing Context
Human-readable health statistics for partner pages and chat. SESRIC (Statistical Centre for Islamic Countries) anchors credibility for the Muslim donor and Zakat audience at scale.
Also: UNFPA · World Bank · Lancet MNH series · Cochrane reviews · DCP3. All open access, no license required.
The Result

Donors give more, give sooner, and give again.

Pre-Donation · Give sooner

MamaMatch AI

Reduce decision fatigue and make giving easier and more intuitive. This will also bring attention to underfunded but critical interventions and smaller, lesser-known partners.

Post-Donation · Give more

Impact Estimate

Build trust through transparency and tailored pathways for personalization.

Return Visit · Give again

Intelligent Re-engagement

Motivate recurrent giving by providing donors with a clear sense of the impact made and offering tailored pathways for continued engagement.

All three moments are shown below.

Pre-Donation · MamaMatch AI

Solving decision paralysis — "Help me choose"

MamaMatch AI guides donors three ways: a Direct Match for donors with a partner in mind, a Guided Match (cause + region + urgency), or region-first browsing. Every path is designed so partner relationships are protected and donor intent is honored.

Scope note: the Year 1 chatbot is English-only. Multilingual support (Arabic, Urdu, French, Indonesian) requires cultural-register advisors and is out of scope for this grant — tracked as complementary-funding work.

MamaMatch AI

AI-powered matching
Select a demo scenario
Guided Match
I want to help mothers in Gaza what partner works in Somalia zakat for pregnant women Bangladesh
Guided Match · portfolio split
$100 monthly for the most urgent need
Direct Match
my friend donated to SRD
The problem we're solving

Helping donors decide — without diverting funds from the partner they came for

❯

Every Pregnancy's partners each work hard to acquire their own donor base. A donor who arrives knowing they want to support Partner X — because of a Ramadan campaign, a friend's share, or a trusted recommendation — represents committed intent that belongs to Partner X. The published evidence is unambiguous about what happens if MamaMatch AI pushes alternatives or splits at this moment: both total donations and the number of successfully-funded partners go down.

Intent override is documented to backfire
When donors with a chosen recipient are offered competing alternatives, total contributions and the number of successfully-funded projects both decrease. Corazzini et al., Journal of Public Economics — peer-reviewed (Tier B).
Recognition beats abstract sorting at entry
Asking donors to self-categorize their intent (“which best describes you today?”) forces an analytical decision before any emotional engagement. Concrete cause cards let donors recognize an outcome and tap. Subsequent forked decisions stay tight — 3 urgency options, top 3 matches with “See all” expansion. Iyengar/Lepper paradox-of-choice lineage — Tier B.
AI persuasion needs concrete framing
AI bots delivering data-driven framing (numbers, sources, specific outcomes) outperform the same bot delivering abstract appeals — p < 0.001 across 4 experiments, ~1,000 participants. Wang et al. 2025, MDPI — peer-reviewed (Tier A).

Our solution maps directly onto the evidence:

  • Multi-layered entry — Cause, Region/Country, Urgency. The matching engine narrows along the layers named in the proposal — type of support, country, urgency. Step 0 lets the donor enter through whichever layer matters most: four cause buttons plus Zakat-eligible giving (Cause / type of support), “Browse by region” (Region / Country), and “I have a partner in mind” (Direct Match shortcut). The remaining layers, including Urgency, are captured in follow-up questions. A passive donor can tap one option emotionally without first deciding what kind of donor they are.
  • Progressive specificity — the path narrows, never forks. Once a cause is set, the donor moves cause → region → urgency → top-3 matches → walkthrough → checkout. Friction is added only after the donor has committed to a topic.
  • G1 and G2 partner-intent guardrails. On Direct Match (donor arrived with a partner in mind) no portfolio or split UI ever appears (G1). The Match Portfolio CTA is available only on the Guided Match results screen, never to a donor who arrived with a specific partner in mind (G2). Eliminates Corazzini-style intent override.
  • Concrete data framing throughout. Every partner card cites IHME Global Burden of Disease MMR with the country source; every Zakat claim is verified against the partner’s profile; every modelled dollar-to-outcome estimate is labelled as a modelled estimate and cites the LiST methodology with its source RCT or Cochrane meta-analysis.

Full source list and verification tiers in docs/chatbot-flow/research-conversion-evidence.md.

AI reads what donors type. Rules do the matching.

The matching engine is deterministic — rule-based filters, mortality ranking, Zakat eligibility flags. The AI has one job: when a donor types instead of clicking. That first free-text message is where intent is fragile, phrasing is unpredictable, and keyword routers break. The AI extracts what it can, asks only for what's missing, and hands off to the rule-based engine. These examples show the full range of what it handles.

Donor use cases

seven typed entries, color-coded by the path each one triggers · illustrative — partner names and MMR figures are demo placeholders; production reads from the live partner dataset and IHME GBD
Guided Match region typed → 1 clarification → single partner
I want to help mothers in Gaza
EP
Got it — partners in Gaza. Any particular cause you’d like to focus on?
Prevent maternal deaths Emergency obstetric care Open to anything
Emergency obstetric care
EP
Healing for Gaza is your best match — they deliver emergency obstetric care and safe delivery in active conflict zones, fully Zakat-eligible. Would you like to proceed to donate?
Yes. I want to donate for $100.
EP
Opening secure checkout for $100 to Healing for Gaza…
Healing for Gaza · $100
Gaza · Zakat ✓ · Emergency obstetric care
Checkout →
Zakat path · Direct match full context typed → instant match, no questions
zakat for pregnant women Bangladesh
EP
Mercy Without Limits is your best match — they provide Zakat-eligible prenatal care and safe delivery support in Bangladesh, top-ranked for your region and giving type. Would you like to proceed to donate?
Yes. I want to donate $75.
EP
Opening secure checkout for $75 to Mercy Without Limits…
Mercy Without Limits · $75
Bangladesh · Zakat ✓ · Prenatal care
Checkout →
Match Portfolio undecided → 1 clarification → split gift
$100 monthly for the most urgent need
EP
$100/month · highest urgency — got it. Which region should this reach?
Middle East East Africa Anywhere critical
Anywhere critical
EP
Here are the 3 highest-need partners — your $100/month split by mortality urgency. Proceed with this portfolio?
Yes. Split it.
EP
Opening secure checkout for $100/month across SRD, MedGlobal & Amoud…
SRD · $38
MedGlobal · $35
Amoud · $27
Impact Portfolio · weighted by IHME data
Checkout →
Direct Match · partner in mind partner named → straight through, G1 guardrail
my friend donated to SRD, I want to do the same
EP
SRD is your match — they deliver emergency obstetric care and skilled birth attendant training in Somalia, one of the highest-need countries in East Africa. Would you like to proceed to donate?
Yes. $50.
EP
Opening secure checkout for $50 to SRD…
SRD · $50
Somalia · MMR 621 · Emergency obstetric care
Checkout →
Guided Match · region browser region typed → filtered partners → select → checkout
what partner works in Somalia
EP
3 partners in East Africa (Somalia) — ranked by maternal mortality urgency:
SRD · MMR 621 Amoud Foundation · MMR 621 Zamzam Foundation · MMR 621
SRD
EP
SRD is your best match — emergency obstetric care and skilled birth support in Somalia, highest maternal mortality in the region. Would you like to proceed to donate?
Yes. $75.
EP
Opening secure checkout for $75 to SRD…
SRD · $75
Somalia · MMR 621 · East Africa
Checkout →
Guided Match · vague intent emotional phrasing → 1 cause question → match
I want to do something meaningful for mothers
EP
That’s a great place to start. What would you like this gift to do?
Prevent maternal deaths Train midwives Open to anything
Train midwives
EP
Midwives for Mothers is your best match — they train skilled birth attendants in rural Sierra Leone. Would you like to proceed to donate?
Midwives for Mothers · West Africa
Sierra Leone · Midwife training
Checkout →
Knowledge Q&A · question mid-flow donor asks about a partner → grounded answer → checkout
Is SRD financially transparent? I want to know before I give.
EP
Yes — SRD is registered with the UK Charity Commission and publishes audited annual accounts. Administrative costs are 12% of expenditure. Would you like to proceed to donate?
Yes. $100 to SRD.
EP
Opening secure checkout for $100 to SRD…
SRD · $100
Somalia · Answer grounded in partner knowledge base
Checkout →

When a donor types instead of clicking, the AI is what prevents the "I don’t understand" dead end. It reads intent, fills in gaps, answers questions — then hands the rest to the rule-based engine. The structured flow handles the journey. The AI handles the door.

Post-Donation Experience

After the gift — impact and re-engagement

The donor journey doesn't end at checkout. On a donor's return visit (7+ days later), MamaMatch AI recognizes them via client-side localStorage and opens with a personalized follow-up — limited to the same device and browser without cleared history. Two post-donation features close the loop from barrier "Understanding impact" and "Knowing what to do next."

MamaMatch AI

AI-powered matching · return visit
EP
Welcome back. Thank you for your last donation…
— $50 to Mercy Without Limits in Bangladesh
— Your gift funded roughly a 2-safe-delivery share of their program, at their reported cost-per-delivery — modelled estimate, not a dedicated purchase.
MOCKUP · proportional attribution at partner's reported rate · LiST methodology + IHME baselines

Stored under mamamatch.donations — partner slug, amount, timestamp, impact string. No email, name, or device fingerprint.

Longitudinal partner outcome reporting — where partners submit updated field data months after a gift — is out of scope and requires complementary funding. Year 1 delivers three things: the immediate post-donation message in chat, the Ramadan return recognition via device memory, and the 3-month follow-up email — all generated from data captured at checkout, with no partner data submission required.

Post-donation email · 3 months later

An impact email — three months after the gift

Three months after donating, donors receive a personalised impact email grounded in donation-time data: the partner they supported, the country context, and the impact estimate generated at checkout. Content is built from donation data (amount, partner, timestamp from the secure payment processor), a conversation summary compressed into topic tags at session end, and country context from the partner selected — no name, device fingerprint, or browsing history is used, and no new partner data submission is required.

From Every Pregnancy <impact@everypregnancy.org>
To donor@example.com
Subject Three months on — a note from Mercy Without Limits
90-Day Dispatch
Jul 18, 2026
A quarterly check-in

Three months on. Your gift is still working.

A snapshot built from the data captured when you gave in Ramadan 2026. No new partner data was required.

Your gift · Ramadan 2026
$50 to Mercy Without Limits
Bangladesh · Apr 18, 2026 · Confirmed · Receipt EP-29481
Where it stands today
~2
safe-delivery share of Mercy Without Limits' Bangladesh program — a modelled attribution of your gift at their reported cost-per-service, translated to lives through LiST methodology and country baselines.
LiST (Lives Saved Tool, Johns Hopkins) · IHME Global Burden of Disease · Mercy Without Limits reported cost-per-service from Every Pregnancy partner data. Modelled estimate — not a measured outcome or dedicated-commodity claim.
In context — Sylhet, Bangladesh
Antenatal-care coverage in Sylhet is still below 50%. Yours is the kind of sustained funding this gap requires.

Ramadan 2027 is where your next gift will land hardest.

Give again to Mercy Without Limits→ Find another partner→
Every Pregnancy
A coalition of 54+ maternal-health partners across 15+ countries.
Unsubscribe View in browser Preferences
How this email is triggered

The email is scheduled at the moment of donation and sent automatically 90 days later. All content — the impact estimate, country context, and partner summary — is drawn from data captured at checkout. No partner data submission is required for this email to send.

Impact Translation & Reporting

Partner outputs → estimated lives saved

This engine powers two things
The chatbot's real-time impact estimate — shown immediately after checkout: "Your $50 ≈ a 2-safe-delivery share of the partner's program."
The 3-month follow-up email — draws on the same coefficients and country data so donors see the modelled impact of their gift months later.
Without this engine, we say "thank you for giving." With it, we say "here's what your money did."

Every estimate combines four inputs: (1) the partner's output count — how many PPH bundles delivered, CYP provided, attended births, etc., reported by the partner (captured once, usually from the existing Every Pregnancy dashboard; no longitudinal re-submission required); (2) the partner's reported cost-per-unit for the specific intervention, so each dollar can be attributed to a defensible output share; (3) a published effectiveness coefficient from a peer-reviewed RCT or Cochrane meta-analysis; (4) a country mortality baseline from IHME Global Burden of Disease (free, public, annually updated). The calculation method is the Lives Saved Tool (LiST) from Johns Hopkins — the same standard used by UNICEF, WHO, UNFPA and GiveWell. When the partner's output maps to a validated coefficient, MamaMatch AI shows both the modelled lives-saved estimate and the partner's reported activity. When it doesn't, the activity is shown on its own without a modelled estimate. Every modelled figure passes a four-condition validity gate and is clearly labeled as a projection, not a measured outcome.

How we attribute — donation to impact

A $50 gift is not a dedicated bundle purchase. It enters the partner's program and is allocated alongside all other funds received. The impact estimate is a proportional attribution — your gift's share of the partner's reported program work at their reported cost-per-unit. Industry standard: MSI Reproductive Choices, UNFPA and GiveWell use the same convention.

1. Donor gives $50 to Partner for Cause X
2. Partner reports cost-per-unit for that intervention
3. $50 ÷ cost-per-unit = proportional share of work
4. Share × LiST coefficient × IHME baseline = modelled impact

Every figure shown is labeled as a modelled estimate — never a measured outcome, never a dedicated-commodity claim. The chatbot passes the donor's selected cause to the donation record so the partner can allocate it appropriately within their program.

① PPH Bundle (E-MOTIVE)

Postpartum Hemorrhage Prevention

Uncontrolled bleeding after childbirth causes ~27% of all maternal deaths. The E-MOTIVE bundle (oxytocin + tranexamic acid + guided response) cuts severe PPH by 60%. Partners report: bundles delivered + country. Everything else is public data.

Illustrative example: a partner delivering ~560 PPH bundles in Nigeria — figures used here are demo placeholders; production uses each partner's reported output and cost-per-unit.
Evidence: E-MOTIVE Trial, Nature Medicine 2024 · RCT · N=10,000+ across Kenya, Nigeria, South Africa, Tanzania
Formula
deaths_averted
  = bundles × 0.60
  × PPH_share(country)
  × CFR_severe(country)
For each bundle delivered: 60% chance of preventing a severe bleed, multiplied by how often severe bleeds are fatal in that country.
Example calculation — MedGlobal · Nigeria
563 × 0.60 × 0.24 × 0.20
= 16–30 maternal deaths averted
Nigeria PPH share: 24% · CFR severe: ~20% · IHME GBD 2022
Donor view
$50 gift
$50 ÷ ~$7/bundle ≈ 7-bundle share of MedGlobal's PPH program
7 × 0.60 × 0.24 × 0.20 ≈ ~0.2 modelled deaths averted
Illustrative demo rate (sector range $3–$15/bundle). Production uses each partner's reported cost-per-unit. Your gift funds a proportional share of the program — not a dedicated-bundle purchase.
Country variables: IHME Global Burden of Disease (free, public) · Effectiveness: E-MOTIVE trial
Why IHME + LiST methodology?

Both were originally developed with Bill & Melinda Gates Foundation funding — IHME Global Burden of Disease (founded 2007 with a $105M Gates grant) and LiST (Lives Saved Tool, built at Johns Hopkins with Gates support as part of the Lancet Child Survival Series). LiST is now the standard used by UNICEF, WHO, UNFPA and GiveWell. When a Gates reviewer reads "we apply LiST methodology with IHME country data," they recognize it — it is the same standard their program officers use internally. This makes estimates auditable, not proprietary.

4 conditions before showing any estimate
  • The intervention has a published RCT or Cochrane coefficient
  • The country has IHME baseline mortality data
  • The partner has reported a verifiable output count
  • The partner has reported a cost-per-unit for that intervention (so $ can be attributed to output share)

If any condition fails, show the partner's reported activity count and program description — never a fabricated modelled number.

Intervention Library — Published Effectiveness Data
Intervention Evidence Effectiveness Primary Source
PPH Bundle (E-MOTIVE) RCT 60% reduction in severe PPH Nature Medicine 2024, E-MOTIVE trial · N=10,000+ across Kenya, Nigeria, South Africa, Tanzania
Antenatal Corticosteroids Cochrane 30% reduction in neonatal deaths Cochrane CDSR 2020, McGoldrick et al. — 27 RCTs
Kangaroo Mother Care Meta-analysis 20–41% reduction in neonatal deaths Cochrane 2023, Conde-Agudelo & Díaz-Rossello
Modern Contraception DCP3 Model 38/100K maternal deaths averted per 1% coverage gain Ahmed et al. 2012, Lancet · LiST methodology · IHME GBD cause fractions
Skilled Birth Attendant Multi-country 21% reduction in neonatal deaths Lee et al. 2011 · meta-analysis
Obstetric Fistula Repair Clinical series 65% quality-of-life improvement post-repair WHO 2018 clinical guideline; Arrowsmith et al. fistula outcomes

All effectiveness coefficients are from peer-reviewed published sources. Applied via LiST methodology with IHME GBD country-specific baselines. Estimates are labeled as modeled projections, not clinical claims.

What each source provides for Every Pregnancy
IHME Global Burden of Disease
Formula engine
  • Country-specific MMR (deaths/100K live births)
  • Cause-of-death fractions by intervention type
  • Free API — gates-funded, no license required
healthdata.org/gbd
UNICEF Data Warehouse
Donor-facing context
  • ANC coverage rates by country
  • Skilled birth attendant coverage
  • Contextual stats for post-donation impact cards
data.unicef.org
SESRIC (OIC Countries)
Muslim donor credibility
  • OIC member-state health indicators
  • Maternal health data for Zakat-relevant countries
  • Adds authority for Muslim-majority audience framing
sesric.org/stats
Technical Development

How we build MamaMatch AI

Six workstreams that turn a working prototype into a production AI system — from infrastructure to measurement. Each phase builds on the last; the knowledge base and partner data pipeline feed every downstream component.

Language scope: Year 1 build is English-only. Multilingual support (Arabic, Urdu, French, Indonesian) is treated as complementary-funding work — it requires faith-literate translation and cultural-register advisors, not just language-pack swaps.

  1. 1 Infrastructure
  2. 2 Knowledge Base
  3. 3 Core AI Development
  4. 4 Configuration & Calibration
  5. 5 Integration & QA
  6. 6 Measurement & Reporting
Responsible AI commitments

Data protection & accessibility

Data protection. The chat conversation never collects personally identifiable information — donors type causes, regions, urgency, and amounts; never names, emails, addresses, or payment details. Personally identifiable information is collected only at checkout under the existing Every Pregnancy privacy notice, stored in encrypted columns separate from conversation traces and analytics events. The language model never receives personally identifiable information — personalized follow-up emails are filled server-side from a deterministic template using checkout data, not generated by the model from donor records. Donors can request deletion via the existing Every Pregnancy data-rights endpoint. Country-level IP geolocation is used for analytics only (country only, no precise location, no linkage to donor identity) and is disclosed in the privacy notice.

Accessibility. MamaMatch AI is built to WCAG 2.1 AA — full keyboard navigation, screen-reader-compatible chat stream, AA color contrast, and visible focus indicators across every interactive element. The interface is mobile-first: chat layout, touch targets, and impact-estimate cards are designed for narrow viewports and low-bandwidth mobile connections, which is the actual context most global-south donors reach Every Pregnancy from.

Success Metrics & Measurement Plan

A practical plan to test, measure, and generate actionable learning

The design test would be three-layer validation (AI evals + A/B + phased rollout) paired with four-aspect measurement (volume, size, completion rate, impact accountability) to measure the success — and donor–AI interactions generate actionable learning for the philanthropic sector.

1
Primary metric
Donation volume lift
Total donation volume from chatbot-assisted donors during Ramadan 2027 vs. the Ramadan 2026 baseline. The clearest "give more" signal.
Collection: payment attribution tag (chatbot vs. direct) vs. prior year baseline
2
Completion rate
% of chatbot sessions that end in a completed donation. Measures whether the chatbot closes the gap between interest and action.
Collection: structured events log — session start → payment confirmed
3
Average gift size
Average donation amount from MamaMatch AI-routed donors vs. direct-traffic donors during the same Ramadan 2027 window. A same-period stepped-wedge comparison — no historical baseline required. The "give more per donor" signal.
Collection: secure payment processor records + chatbot-vs-direct attribution tag, Ramadan 2027 cohort comparison
4
Impact estimate qualification rate
% of completed donations where the 4-condition validity gate passed (estimate shown) vs. fallback message shown, with per-condition failure breakdown. Measures responsible AI execution in practice.
Collection: structured events log — impact_gate_result event with {pass, failed_condition} per completed donation

Measurement infrastructure

Structured Events Log
Conversion funnel: step entered → cause selected → partner matched → checkout opened → donation completed
AI Quality Tests
Automated test suite run before each deploy — verifies intent parsing accuracy and that AI answers stay grounded in the knowledge base
Web Analytics
Public-facing funnel analytics for Every Pregnancy and Gates Foundation reporting

Year 1 measurement report — findings reported to Gates Foundation

1
Donor persona report — conversation patterns by behavioural segment (entry path · cause choice · Zakat vs. non-Zakat path · returning vs. first-time donor · typed-input vs. button-only navigation) and by country-level IP geolocation (disclosed in the privacy notice; country only, no precise location, no PII linkage)
2
Intent → completion funnel breakdown by chatbot path (A/B/C) and drop-off step
3
Zakat/Ramadan giving pattern insights — seasonal behaviour, Zakat-path conversion compared to non-Zakat paths, and Ramadan timing effects (English-only in Year 1; multilingual register analysis is complementary-funding work)
4
Impact translation accountability report — 4-condition validity gate results with per-condition failure breakdown, impact estimate qualification rates, and responsible AI audit findings

Complete your donation

🔒 Secure checkout
$
edit amount
Processed via a secure payment provider. Every Pregnancy never stores your card details.
Your donation is tax-deductible where applicable.