What is the name of your organization?
Sakhi
What is the name of your solution?
Sakhi
Provide a one-line summary or tagline for your solution.
Sakhi empowers mothers in underserved communities with AI-driven, culturally-rooted maternal health support, right on WhatsApp.
In what city, town, or region is your solution team headquartered?
India
In what country is your solution team headquartered?
India
What type of organization is your solution team?
For-profit, including B-Corp or similar models
Film your elevator pitch.
What specific problem are you solving?
India faces a maternal mortality rate of 103 deaths per 100,000 births, affecting 30 million women annually, with a 2030 SDG target of 70 deaths per 100,000 births. Three delays drive this crisis: seeking care information, reaching facilities, and receiving care. We focus on the first delay. In rural India, 9/10 individuals lack health literacy, with 64% relying on public healthcare supported by 78,000 NGOs and frontline workers (ASHAs, Aanganwadi Workers, ANM Nurses). We spent 3 months doing field research with over 80 ASHA workers, District Health Officials, Pregnant Women, and NGO officials. We found three interconnected failures in the current system—systemic disempowerment of women in healthcare decisions owing to a lack of education and information. There is 1 ASHA worker per 1000 people; hence, they are overworked and underpaid. Because of this, healthcare providers and recipients have a broken feedback loop. Healthcare providers get mothers' information with 3-7 day delays, standardized knowledge is lacking, and 70% lose INR 10,800 ($130) in government benefits due to information gaps. Existing government programs have moderately leveraged technology, with one-way voice-based messaging services like Kilkari only having 23% adoption.
What is your solution?
At Sakhi, we bridge the feedback loop gap through a WhatsApp-based AI platform leveraging high adoption rates (83%) to connect previously isolated mothers, ASHA workers, and NGOs. We have three key components:
A human in the loop, a Retrieval Augmented Generation LLM-backed chatbot with an expert-curated knowledge base providing personalized healthcare information, antenatal care reminders, and government medical schemes in their local dialects. We've integrated voice features for users with limited literacy.
Second, a verification platform where healthcare and frontline workers can review LLM-generated answers to questions not in our dataset. Verified answers are added to our knowledge base, ensuring culturally sensitive and easy-to-understand information. Using behavioral science principles, we've designed our communication to motivate active behavioral change beyond passive information consumption.
Third, a real-time analytics dashboard for NGOs with key metrics for program improvements, M&E data, high-risk pregnancy tracking, and migration movement monitoring for immediate intervention. Our solution is designed to work within the existing care delivery system and optimize human effort, not replace it. This is a systems-level problem, and our technology is very human-in-the-loop and complementary.
https://drive.google.com/file/d/1xZCikB4m9CXJh2nPnVNNsDj-4AYcheTE/view?usp=sharing
Who does your solution serve, and in what ways will the solution impact their lives?
Our solution is designed to serve approximately 19.2 million women in rural and urban poor parts of India (64% of 30 million women) who rely on public health to give birth. These are women aged between 19 and 30, predominantly first-time mothers, with education levels below 10th grade, in households earning approximately $2,400 annually. 65% of these women have smartphone access (either personal or shared). We also serve the 78,000 NGOs that work in India in RMCH programs. As described in the problem statement, these women lack timely and unbiased health literacy, owing to the low healthcare provider to pregant women ratio.
Sakhi builds confidence in healthcare decision-making through reliable, culturally aligned information, reducing information seeking time from 3-4 days to under 2 hours, facilitating access to government benefits and frontline workers, and maintaining continuity of care during migration to maternal homes in late pregnancy, ultimately to increase antenatal care compliance. At the same time, it provides the 78,000 NGOs with relevant program metrics, high risk pregnancy detection, to reduce time taken to action and also save M&E costs.
