Gates Foundation Grand Challenge

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

Personalized AI matching that connects everyday givers to the right maternal health partner — based on their values, geography, and giving preferences — across a coalition of 50+ frontline organizations.

$91M+
Raised, Ramadan 2026
50+
Partner organizations
192K+
Donors across 3 campaigns
136
Countries represented

The gap between donor intent and informed action

Maternal health is complex — spanning prenatal, delivery, and postpartum care across dozens of countries. Donors arrive motivated but face barriers that suppress giving: the cause feels abstract, the choices are overwhelming, and the path from intent to impact is unclear.

50+

Organizations, one abstract cause

Maternal health spans before, during, and after birth. With 50+ partners across 15 countries, donors can't see where their money goes — the cause feels like an "abstract term" rather than a concrete action.

79%

Never complete their gift

Of donors who reach a donation page, nearly 4 in 5 leave without giving. The intent is there — the path is unclear.

60%

Intend to return, but don't

Of those who abandon, most say they'll come back. The vast majority never do. Every moment of friction costs lives.

Sources: Fundraise Up, Donation Page Friction Study (2024); M+R Benchmarks; Snowball Fundraising

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

Helps donors navigate a complex, multi-layered cause. AI identifies donor preferences — geography, intervention type, faith-based giving — 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

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 — Report

Unified Impact Data System

Partners report diverse data — healthcare workers trained, delivery kits distributed, C-sections performed — in different formats via Google Drive. This pillar builds a dashboard-based reporting system with AI that translates diverse partner data into a uniform, compelling donor story.

From intent to impact — the causal pathway

Donor arrives with intent

Motivated to support maternal health — but unsure where to give among 50+ organizations

AI reduces discovery friction

Conversational matching recommends 1-3 partners based on donor values, geography, and preferences

AI profiles donor preferences

Builds understanding of the donor — geography, intervention interest, giving capacity, faith-based motivation — to create personalized matches and segment into actionable personas

Evidence contextualizes the need

WHO/UNICEF data grounds each partner's work in real maternal health gaps — not abstract appeals

Donor gives with confidence

Higher conviction, better match, to the organization whose work resonates most

Feedback loop sustains giving

Donor sees their $10 become $30 through the Pregnancy Boost Fund, funding 10 months of antenatal care. Personalized impact stories — not receipts — drive repeat giving.

More funding reaches frontline organizations

Where it's most needed and most aligned with donor values

Smaller partners get discovered

AI prioritizes under-funded organizations when donors are undecided — leveling the playing field within the coalition and directing funds where they're needed most

Sector-wide learnings published

Open data standards, matching patterns, and findings shared for replication

Two friction points that suppress giving

When donors arrive at Every Pregnancy, they're motivated — but two distinct barriers stand between intent and donation. Each requires a different AI intervention.

Problem 1

Decision paralysis

"I want to give, but there are 50+ partners — which one should I fund?" The donor is motivated but overwhelmed. They don't know enough to choose, and the cost of choosing wrong feels high. Many leave without giving. Even those who pick one partner wonder if they're missing a better fit elsewhere.

What's needed: A way to narrow 50+ options to 1–3 matches — or skip choosing entirely by allocating across multiple partners at once, weighted by evidence.
Problem 2

Last-mile hesitation

"I'm on a partner page, but I'm not sure this is the right one." The donor has already found a partner — via a social media link, a friend's recommendation, or browsing. They have intent, but not enough conviction to click donate.

What's needed: Contextual evidence and impact clarity right on the partner page — reducing hesitation at the moment of highest intent, not adding more to read.

These are different donors at different stages. Problem 1 is an entry point problem — "help me decide." Problem 2 is a last mile problem — "help me commit." Each gets its own AI tool below.

Solving decision paralysis — "Help me choose"

The Navigator starts by asking the most important question: do you already know which partner you want to support? Donors who arrive with a specific partner in mind go straight to that partner — no alternatives shown, no split suggested, no diversion of funds. Donors who are undecided get the guided flow: 2–3 questions on motivation, region, and budget, then the AI narrows 50+ partners down to the best matches. Only after a donation is complete do we invite the donor to make an additional gift to a similar partner — protecting each partner's relationship with their loyal donor base.

EP

EP Impact Navigator

AI-powered matching
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 the Navigator 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).
Paradox of choice has a measurable threshold
More than 3 options at any single decision point reduces both participation rate and average gift size. The 3-option ceiling is the published optimum for charitable-giving decisions. Iyengar/Lepper 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:

  • Ternary entry gate (Partner / Cause / Help me find) — exactly 3 options, the published optimum.
  • G1–G6 partner-intent guardrails — once a donor signals partner intent, no portfolio or split UI ever appears, eliminating Corazzini-style intent override.
  • Concrete data framing throughout — every partner card shows WHO MMR with country source citation, every Zakat claim is verified, every dollar amount maps to a specific outcome.
  • Up-sell only after the gift is locked in, framed as additive (a separate new gift), never as a re-allocation.

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

How the conversation flows

Three paths, one principle: the donor's original intent is protected

The Navigator branches at the very first question — three options, one per intent level. Each branch has guardrails baked in so we never cross partner interests. Persistent ↶ Back to step and ↻ Start over controls live in the chat header so a donor can revisit any prior decision without losing the conversation.

STEP 0 · ENTRY (TERNARY GATE) "Which best describes you today?" PARTNER IN MIND CAUSE OR REGION HELP ME FIND PATH A · SINGLE PARTNER 1Searchable pickerAll EP partners 2Walkthrough onlyWHO data · Zakat 3Checkout 100%No split, no diversion 🛡 G1 GUARDRAILNo portfolio button on this path PATH C · CAUSE / REGION 1Single-tap filter3 causes / 4 regions 2Top 3 + See allG7 — never dead-ends 3Pick OR portfolio→ checkout 🛡 G7 GUARDRAILEmpty filter offers broaden / picker PATH B · GUIDED MATCHING 13 questions (adaptive)Region skipped if "anywhere" 2Top 3 + See allInline portfolio CTA 3Pick OR portfolio+ Increase / Add / Both 🛡 G2 GUARDRAILPortfolio only on matches screen COMMIT GATE · STRIPE Donation processed — intent locked in Resume flow if donor closes modal mid-payment + OPTIONAL UP-SELL · ADDITIVE · SINGLE-PARTNER ONLY Celebrate-first bubble: "Mothers in [Country] will feel that…" Suggest 2–3 similar partners (region + cause match) as separate new gifts 🛡 G4 NEVER A SPLIT
Path A · highest intent
Path C · cause-led
Path B · open intent
Commit gate
Guardrail
0
Step 0 — Entry: "Which best describes you today?"
A 3-option gate (the published optimum, per Iyengar/Lepper). The donor's answer routes to one of three paths designed for a specific intent level.
Path A · highest intent
"I have a partner in mind"
Focused single-partner path
1Searchable picker — full partner list, vertical 2-column grid
2Walkthrough — that partner only · WHO context · Zakat status
3Checkout — 100% to this partner · single Stripe transaction
🛡 G1 · No portfolio button anywhere on this path
Path C · cause-led intent
"I have a cause or region in mind"
Single-tap filter to results
1Pick one — Causes section (childbirth · workers · Zakat) OR Countries section (4 regions)
2Top 3 of N matches · "See all" expansion shows the full filter
3Pick a partner OR optional Impact Portfolio → checkout
🛡 G7 · Empty filter never dead-ends — bot offers broaden / open picker
Path B · open intent
"Help me find the right match"
Adaptive guided matching
1Motivation · Region · Budget — region skipped if motivation is "Maximum impact (anywhere)"
2Top 3 of N matches · inline "Try Impact Portfolio" CTA · "See all" expansion
3Pick partner OR open portfolio (with Increase budget · Add partners · Both selectors) → checkout
🛡 G2 · Portfolio only on the matches screen, never after a partner is picked
Commit gate — donation processed via Stripe
Whatever the donor chose — single partner or portfolio — the payment lands first. The donor's original intent is now locked in. Closing the modal mid-payment shows a Resume/Change amount/Start over recovery in the chat.
+
Optional up-sell — single celebrate-first bubble (single-partner gifts only)
"Mothers in [Country] will feel that. If you'd like to extend the impact, here are partners working on similar programs — each would be a separate additional gift, processed on its own." Suggests up to 3 similar partners (region + cause match) as new transactions. Skipped entirely after portfolio donations — those donors already chose breadth.
🛡 G4 · Always additive · never a re-allocation of the locked-in gift
Header controls — always visible in every state
↶ Back to step opens a dropdown of every prior decision the donor reached (Step 0 entry · Path B questions · Top matches · Walkthrough for each partner viewed). Selecting one re-enters that step additively so the prior conversation stays visible above. ↻ Start over wipes everything and restarts from Step 0.
Core principle
The Navigator never moves a donor's intended gift away from the partner they came to support. Every branch, every button, every up-sell, and every back-step option is designed so that partner trust is protected and donor intent is honored.
Research on proven conversion rate

The published evidence supports a +15–35% checkout lift — conditional on the design choices we already made

We did not build the Navigator to test a hypothesis — we built it to match what the literature already shows works. Below are the verified numbers that justify the projected lift, with verification tier per claim. Anything weaker than Tier B is excluded from this view.

+164%
Conversion lift
Fundraise Up AI-powered forms vs. M+R baseline (29% vs. 11%)
$250
Avg gift via chatbot
vs. $121 industry baseline — chatbot-sourced donors give larger
+28%
Embedded-giving lift
Revenue per visitor when checkout is on-page (no redirect)
Tier A — peer-reviewed academic study
Wang et al. 2025 — Journal of Theoretical and Applied Electronic Commerce Research, MDPI
4 between-subjects experiments · ~1,000 participants · UK + China · 2×2 factorial design
AI chatbots delivering concrete, data-driven copy outperform the same AI delivering abstract appeals (p < 0.001 across all 4 studies). When the AI carries an anthropomorphic avatar (warm visual + conversational tone), the framing penalty disappears and trust matches a human agent. Effect replicates cross-culturally.
→ Maps directly onto the Navigator: warm EP avatar, MMR numerics with WHO citations, $X-funds-Y outcomes, mono-font for data.
Tier B — primary industry benchmarks & peer-reviewed economics
M+R Benchmarks 2025 — the baseline
Static donation page conversion: 11–12% desktop, 8–11% mobile. Average one-time gift: $121.
Fundraise Up — the upper bound
AI-optimized donation forms convert at 29% vs. 11% industry baseline. Average gift $169 vs. $126 (+34%).
2025 Brand Discovery in the Age of AI
Donors arriving via chatbot give an average gift of $250 (vs. $121 industry baseline). Slower to convert on first visit, larger when they do.
Corazzini et al. — J. Public Economics
When donors face multiple competing recipients, both total donations and number of successfully-funded projects decrease as option count rises. 3 options is the optimum.
When chatbot integration reduces conversion — the 4 documented failure modes
1. Intent override. Suggesting alternatives to a donor with chosen recipient. → G1–G6 prevent this.
2. Paradox of choice. >3 options at any decision point. → Step 0 ternary gate; QR rows capped.
3. Funnel bloat. Steps that add no value. → Adaptive Path B skips region for "anywhere".
4. Intent misunderstanding. Bad NLU killing flow. → Pill-driven decisions; free text is fallback only.
Bottom-line projection

A chatbot-integrated donation flow increases conversion by roughly +15–35% on checkout completion with a meaningfully higher average gift — conditional on avoiding the four failure modes above. The Navigator v2 design satisfies every condition the research says matters.

The single scenario that would reverse this lift: surfacing the Impact Portfolio split to a donor who arrived with partner intent (Corazzini et al., Tier B peer-reviewed). Guardrails G1–G6 already prevent it — which is why the v2 refinement matters.

Full source list, primary URLs, and weaker-tier claims (excluded from this view) in docs/chatbot-flow/research-conversion-evidence.md.

The gap

Demo 1 solves the "where should I give?" problem well. But not every donor arrives undecided. Many land directly on a partner page — from a social media post, a friend's link, a Google search, or a Ramadan campaign. They already know who. They need to know why this one and why now.

A chatbot doesn't serve this donor — it adds a new interaction on top of the page they're already reading. What's needed is contextual AI that works within the partner page experience: surfacing evidence, answering hesitations, and reducing the distance between "I'm interested" and "I'm donating."

Solving last-mile hesitation — "Convince me here"

This is for donors who are already on a partner page with intent. The AI widget doesn't summarize what's already visible — it adds what's missing: WHO health data for the partner's country, how donations are used, Zakat eligibility, and impact at specific giving levels. It meets the donor at the point of highest intent and removes the last reasons not to give.

everypregnancy.org/partners/srd
SRD Partner Page on everypregnancy.org
AI
EP Impact Assistant
Powered by AI

Evidence-based context, not abstract appeals

WHO/UNICEF data grounds each partner's work in real maternal health gaps. AI generates the connecting narrative — never fabricates claims.

🇧🇩 Bangladesh — Maternal Health Context

Data for partner and country pages
Maternal mortality ratio123 per 100K
Skilled birth attendance59%
Antenatal care (4+ visits)47%
Births per year~3.0 million
EP partners operating4 organizations
Total raised via EP$1.2M
Source: WHO/UNICEF/UNFPA/World Bank, Trends in Maternal Mortality 2000-2020 (2023 report) · CC BY 4.0. Note: 2023 estimates now available — to be updated for final submission.
✨ AI-Generated Context
"In Bangladesh, where only 59% of births are attended by skilled health workers, Mercy Without Limits provides prenatal checkups, safe delivery kits, and emergency obstetric referrals in rural Sylhet — one of the country's most underserved regions."

Health data: WHO/UNICEF (attributed, dated). Partner description: partner-approved. AI generates the connecting narrative only.

Understanding givers to serve them better

Before AI can match donors to partners, it needs to understand who's giving and why. We're building a donor intelligence layer from partner interviews, behavioral data, and persona-driven A/B testing.

🎤
Research

Partner Interviews

Interview partners to understand their existing donor base — age, gender, geography, giving patterns. These insights seed the initial persona models and reveal what motivates real donors.

👤
Segmentation

5-6 Donor Personas

Build actionable donor personas that guide segmentation, communication, and AI matching. Each persona represents a distinct motivation and giving pattern — from first-time Zakat givers to recurring monthly donors.

🧪
Optimization

A/B Testing & Lead Scoring

Test persona-driven experiences against controls. Iterate based on what segmentation works. Build toward automated lead scoring that improves matching accuracy over time.

From foundation to Ramadan 2027

Months 1–3
Foundation
  • Partner Impact Profiles (8-10)
  • WHO/UNICEF data pipeline
  • Matching algorithm prototype
  • A/B testing framework
Months 3–6
Build
  • Navigator chatbot UI
  • Impact context blocks
  • Post-donation experience
  • Internal alpha testing
Months 6–9
Test
  • Beta launch with real donors
  • A/B test vs. control group
  • Donor behavior tracking
  • Iterate based on data
Months 9–12
Scale + Publish
  • Ramadan 2027 full deployment
  • Measure vs. 2026 baseline
  • Extend to all 50+ partners
  • Publish sector findings

$300K total investment — half from us

The Grand Challenge awards up to $150K per project for 1 year. Co-funding is welcomed. We're matching the grant with $150K in EP resources — demonstrating commitment and extending reach.

Gates Foundation Grant
$150K
AI/ML development & integration$70K
Product design & frontend$30K
Data infrastructure$20K
Testing, measurement & publication$20K
Partner outreach & coordination$10K
EP Co-Funding
$150K
Engineering team allocation (existing staff)In-kind
Platform infrastructure (servers, APIs, Stripe)In-kind
Partner relationship managementIn-kind
Ramadan campaign operationsIn-kind
4+ years of donor data & baselinesIn-kind

Grant allocation is EP's proposed budget. Co-funding per team discussion (Apr 10, 2026). Grand Challenge: co-funding welcomed but not required (FAQs).

How we measure impact

>60%
Match acceptance rate
Percentage of Navigator users who click through to a recommended partner and proceed to donate.
+25%
Conversion rate lift
Navigator users completing a donation vs. organic visitors who browse without AI guidance.
+15%
Donor return rate
Donors who give again within 6 months — sustained engagement beyond a single transaction.
-30%
Time to donation
Reduction in time from first visit to completed gift. Faster decisions, less friction, more lives reached.

EP operates in the gap

No existing AI tool solves the coalition navigation problem — routing donors across a curated network of organizations in a shared cause area.

PlatformApproachEP's Differentiation
Charity NavigatorAI ratings across 1.5M+ charitiesEP goes deep within one cause across a curated coalition
DaffyConversational AI for DAF holdersEP serves mass donors, especially Ramadan/Zakat givers
Fundraise UpAI-optimized single-org donation formsEP innovates on which org — the routing layer
DonorSearchAI tools for fundraising teams (B2B)EP's Navigator is donor-facing — donors are the users
Muslim Charity UKAI-personalized Ramadan emailsEP applies AI across 50+ orgs with Zakat routing

Nine reasons this is grant-ready

1

Unsolved sector problem

No AI tool tackles coalition-level donor routing — a structural gap in philanthropic technology.

Source: Competitive analysis — Charity Navigator, Daffy, Fundraise Up, DonorSearch serve different niches (see Competitive Landscape above)

2

Born from the Gates Foundation

EP originated from a Gates Foundation planning initiative on maternal health and the Muslim community in 2023. This proposal extends a relationship the Foundation already invested in.

Source: everypregnancy.org; Devex profile for Marleen Vellekoop; Gates Foundation grant records

3

Execution-ready

Platform, partners, donors, and data already exist. The grant funds the AI layer — not building from scratch.

Source: EP production platform (everypregnancy.org), 50+ active partner integrations, Stripe payment infrastructure

4

Natural experiment built in

Ramadan provides a 30-day controlled window with 3 years of campaign baseline data ($13M in 2024, $21M in 2025, $91M in 2026). Plus $150K in co-funding — $300K total commitment.

Source: EP Ramadan campaign reports (2024-2026); co-funding per team discussion Apr 10, 2026; Grand Challenge FAQs confirm co-funding is welcomed

5

All three challenge areas

Matching, engagement, and data infrastructure as interdependent layers — not stapled together.

Source: Grand Challenge criteria — connect, convert, infrastructure (gcgh.grandchallenges.org)

6

Culturally specific, globally transferable

Ramadan/Zakat AI is under-researched. Muslim Charity UK achieved 4.1x increase in donations per email using AI-personalized Ramadan outreach — for a single org. EP applies this at coalition scale.

Source: Muslim Charity UK, Insight in Fundraising Conference 2024 award; JCASC AI-Enabled Islamic Philanthropy Framework (2024)

7

Committed to sector learning

Open findings, open schema, replicable patterns. $150K produces knowledge, not just features.

Source: Grand Challenge values "approaches that generate insights" and learning (FAQs); SSIR framework on AI for nonprofits

8

Levels the playing field

AI matching can prioritize smaller, under-funded partner organizations when donors are undecided — ensuring the coalition's smallest members benefit from the platform, not just the largest.

Source: Team ideation meeting (Apr 10, 2026) — addressing equity within the coalition

9

Scale of proven demand

$91M raised in Ramadan 2026 from 192K+ donors. The intent exists at massive scale — the gap is between intent and informed action.

Source: EP Ramadan 2026 campaign results (everypregnancy.org)

Complete your donation

🔒 Secure checkout
$
edit amount