One-line solution summary:
Supporting small farmer livelihoods and building food system resilience through digital connections
Pitch your solution.
COVID-19 has disrupted agricultural supply chains in India, contributing to widespread hunger. Public and private sector buyers are struggling to find reliable, aggregated supplies. Difficulties farmers face selling produce through traditional channels highlight the need for a digital marketplace that enables lower transaction costs and greater value capture for farmers.
The digital solutions we are developing address these problems for both smallholder farmers and buyers. A voice-enabled WhatsApp chatbot will provide seamless market connections, enabling smallholder farmers to improve their incomes amid economic disruptions caused by COVID-19. A searchable digital marketplace will allow buyers to discover local produce, assess quality via photos, and purchase directly from farmers.
By scaling the solution across six states in India, we will improve livelihoods of 1,000,000 smallholder farmers within 3 years. Replicating the solution in Bangladesh, Kenya, and Ethiopia within the next 5 years will benefit an additional 1,000,000 farmers.
Film your elevator pitch.
What specific problem are you solving?
Globally, 2 billion people rely on 500 million small farms to supply their food. In India, with many regional markets closed or operating at limited capacity to contain COVID-19, food wastage is growing as farmers are unable to reach markets. Market closures affect farmers’ incomes as well as their own food supplies -- local markets provide about 80% of food consumed by the region’s rural residents. A recent survey in 12 Indian states found that more than 50% of respondents have reduced the number of times they are eating each day and 68% have reduced items in meals. Economic shocks compound existing challenges in rural India, where about 415 million people already live below the poverty line. A World Bank analysis predicts that 12 million people in India will fall into extreme poverty (living on less than $1.90/day) in 2020 as a result of COVID-19.
Even absent pandemic conditions, Indian farmers struggle to capture much of the value of their produce, which typically travels through numerous middlemen. Farmers can receive as little as 25% of the final sale price of their crop, throttling productive investment and slowing the transition from subsistence to market based farming.
What is your solution?
Loop allows farmers to share the type, quantity, and price of crops they wish to sell using a chatbot accessed via WhatsApp, a digital messaging platform. Enabled by local language support and speech recognition, farmers enter this information along with photos. Agricultural buyers, including small purchasers from the local community looking for nutritious foods or large industrial and retail buyers, use the same chatbot interface to discover available produce, using farmer uploaded photos to assess quality. The buyers directly contact farmers via WhatsApp to complete the transaction. With 88% of India able to access 4G networks and more than 500 million internet users across the country, rural India is poised to benefit from digital solutions that would have been unimaginable only a few years ago. In cases where users struggle, our government extension partners, who can be reached by both farmers and buyers through Loop, are available to backstop the service and provide user support. This innovative solution enables buyers to bypass regional markets and middlemen to purchase directly from farmers.
Who does your solution serve, and in what ways will the solution impact their lives?
We work with smallholder farmers in India (among vast majority with holdings under 2 hectares) who lack access to quality market information. Working with government extension partners in Odisha, Bihar, and Jharkhand, our local teams - which include members from agricultural backgrounds - engage in human centered design assessments to understand community needs and co-develop solutions with farming communities and government systems that serve them. We have identified several challenges:
- Farmers struggle to sell all of their produce at a reasonable price;
- Households are unable to access nutritious foods, due to cost and availability;
- The 30% of rural Indians who lack basic literacy cannot access existing technology-based solutions;
- COVID-19 shutdowns have brought younger, technology savvy urban migrant workers back to rural villages, boosting on-farm mobile literacy while leaving households struggling to feed additional mouths.
To address these challenges, we’re using WhatsApp (installed on over 95% of Indian smartphones) and local language voice input and audio and video responses to enable more farmers to participate. Linking farmers and buyers via an easy-to-use WhatsApp chatbot will improve system-wide resilience to shocks, lower food buyer prices through improved sourcing and reduced wastage, and allow farmers to capture more value from their sales.
Which dimension of the Challenge does your solution most closely address?
Improve supply chain practices to reduce food loss, scale new business models for producer-market connections, and create low-carbon cold chains
Explain how the problem, your solution, and your solution’s target population relate to the Challenge and your selected dimension.
Loop addresses the challenge by creating a digital network of farmers and food buyers to bypass inefficient markets. When functioning, these markets often provide extremely low returns to smallholder farmers. When subject to outside shocks, such as the current COVID-19 pandemic or future climate-linked disturbances, these markets often cease functioning entirely, causing widespread food loss. By providing farmers and wholesale food buyers with an alternative digital marketplace, Loop will increase availability of food and lower costs for consumers by eliminating supply throttles, thus reducing food loss and boosting incomes of smallholder farmers.
In what city, town, or region is your solution team headquartered?Delhi, India
What is your solution’s stage of development?
Growth: An organization with an established product, service, or business model rolled out in one or, ideally, several communities, which is poised for further growth
Who is the primary delegate for your solution?
Rikin Gandhi, Executive Director
Which of the following categories best describes your solution?
A new application of an existing technology
Describe what makes your solution innovative.
Loop represents two significant innovations. The first is grounded in its unique local language voice recognition enabled chatbot. Loop’s local language voice recognition technology makes the service accessible to farmers who struggle with literacy or those who are new smartphone users. Furthermore, through our partnerships with government extension agencies, farmer queries can be backstopped by support from trained extension agents.
While chatbots have been deployed in many interventions, the narrowly bounded set of variables required by Loop - crop, quantity, and price - allow us to deploy a product that can effectively interact with users’ voice-based inputs. Previous efforts at farmer targeted chatbots, like FarmChat, which delivers potato advisories via chatbot, or the Indian Farmers’ Fertilizer Cooperative’s Oracle chatbot for delivering financial and inventory data to farming cooperatives, have struggled to effectively respond to broader sets of potential queries, limiting their usefulness to farmers.
Loop’s second major innovation comes from its deployment as part of an integrated suite of Digital Green solutions. Loop was developed when our video-enabled approach to agricultural extension cost-effectively boosted production, but farmers struggled to translate that boost into significant income gains. By layering Loop on top of our video extension outreach to over 1 million farmers, we were able to deliver more robust impact by supporting farmers from seed to sale. Transforming Loop into a self-service digital application to connect farmers to markets enables us to cost effectively deploy that impact at even wider scale.
Describe the core technology that powers your solution.
Loop’s core technology is an AI chatbot with local language voice recognition functionality. Farmers and agricultural buyers will interact with this chatbot using WhatsApp, a messaging platform with near ubiquitous adoption among the rapidly growing community of Indian smartphone users. Farmers will speak to the chatbot, which will support Hindi and Odiya languages, to report their cropping plans and produce available for sale using natural language voice or text based queries. This chatbot will be supported by the local agricultural extension staff of our government partners to respond to queries that require human intervention. Agricultural buyers, which may be local purchasers or large food processors or retailers, will rely on the same WhatsApp chatbot interface to discover available produce, accompanied by photos to allow them to assess quality, prices, and quantity available.
Provide evidence that this technology works.
KisanDiary, a digital ledger to help farmers track on-farm transactions and cropping plans, and earlier iterations of Loop as a farm-to-market logistics services demonstrated early successes in improving rural market linkages. Through Loop, Digital Green deployed a farmer-facing app and human staffed call center for booking produce transport to market. Loop served 19,713 farmers and was responsible for the sale of 15,200 metric tons of produce that resulted in $2.9 million in farmer earnings. We also launched a digital supplier directory this year, which provided similar functionality to the solution proposed in this application. While results were promising, the directory relied on staffed call centers to populate its produce database. Our current iteration of Loop provides a similar impact to these efforts while ensuring cost-effective scalability through deployment of the AI enabled chatbot that addresses the high labor cost required to operate an effective logistics management service and call center. While efforts to deploy chatbots in rural settings are not new, low literacy rates can impede their effectiveness. Our deployment of local languages, voice support, and human backstopping will allow us to sidestep these constraints. Likewise, chatbots that attempt to do too much often fail to respond accurately to queries. Because of the narrow set of functions processed by this chatbot - tracking the volume and prices of commodities - we’re able to sidestep many of the technology’s challenges when negotiating a wider set of variables.
Please select the technologies currently used in your solution:
What is your theory of change?
Through Loop’s WhatsApp-based, local language voice chatbot, farmers list the quantity, price, and type of commodity they’re offering for sale. Buyers search available produce by type or seller.
Buyers purchase produce directly from farmers. Loop has already seen robust demand for the service, with sustained demand even as we shifted to a fee-for-service model. Farmers reported strong demand for assistance with market access and 80% of farmers were repeat users of the service.
Participating farmers see a net increase in income of at least 25% (after commission) for produce sold through Loop and time savings of up to four days every month as they avoid market trips. As validated by third-party studies, farmers have realized a 15% price realization gain and saved 45% on transportation costs. Agricultural buyers see cost savings as middlemen are eliminated from their supply chain. Local availability of nutritious fruits and vegetables increases in project areas. Market disruptions due to COVID-19 are reduced, leading to reduced food loss, in turn exerting downward pressure on food prices as a whole.
Increased farmer income will lead to increased investment in market oriented agriculture, aiding smallholder farmers as India transitions from subsistence to commercial farming. This will lead to further income gains, as farmers learn how to produce, grade, and sort their crops in line with market demands. As farmers and buyers engage in less travel to often distant markets, they’ll save time and reduce the climate impacts of marketing their produce. Food buyers will benefit from increased availability of nutritious food and lower food prices. Loop will also build the resilience of systems to function during system-wide shocks, such as the current pandemic or future health or climate linked crises, contributing to ongoing food availability, lower food prices, and improved nutrition.
Select the key characteristics of your target population.
Which of the UN Sustainable Development Goals does your solution address?
In which countries do you currently operate?
In which countries will you be operating within the next year?
How many people does your solution currently serve? How many will it serve in one year? In five years?
Through our suite of related solutions - Loop, KisanDiary, and our Supplier Directory - we have served about 30,000 farmers. In one year, we will have enrolled 200,000 farmers in Loop. In five years, we will have scaled the solution to 2 million farmers.
What are your goals within the next year and within the next five years?
Over the course of the first year of deploying Loop, we foresee process and systems goals and reach and impact goals.
Process and system goals include:
- 30 days: Roll out digital marketplace and initial chatbot.
- 90 days: Prototype and test enhanced chatbot with voice recognition, validate with farmer focus groups.
- 120 days: Chatbot with Hindi voice recognition goes live, distributed through government partners.
- 6 months: Chatbot with Odiya voice recognition goes live; digital marketplace enrollment accelerated.
- 12 months: Additional language support, increased linkages to distant markets, and linkages to food producers who may accept lower quality produce.
Reach and impact goals include:
- 6 months: 50,000 farmers participating in Loop.
- 9 months: 150,000 farmers participating;
- 12 months: 200,000 farmers participating;
- Farmers realize 25% increase in incomes through reduced marketing costs, decreased food wastage, increased volume sold, and greater price realization; and
- Consumers benefit from increased availability of fruit and vegetables in local markets and reduced fruit and vegetable prices through simplified supply chains that reduce intermediate sellers.
● 3 Years: Expand Loop to three additional Indian states, Bangladesh, Ethiopia, and Kenya. One million farmers sell produce through Loop.
● 5 Years: 2 million farmers sell produce through Loop.
What barriers currently exist for you to accomplish your goals in the next year and in the next five years?
Technical barriers: Building a voice-based chatbot usable by farmers with limited literacy who are new to smartphone technology requires innovative approaches.
Market barriers: Farmers and buyers will require support to conduct transactions using digital payments.
Financial barriers: Farmers will require surety that their transactions are safe in the event of disputes over sale quantity or quality.
Capacity barriers: A shortage of data scientists and AI researchers in the agricultural development field and the social impact sector overall.
How do you plan to overcome these barriers?
Technical barrier mitigation: By focusing on a narrow domain - crop, location, volume, and price - we’re able to sidestep the challenges of responding to broader sets of natural language queries.
Market barrier mitigation: Both mobile connectivity and digital payment gateways are rapidly expanding in India, even under ongoing pandemic conditions, as is farmers’ technological literacy. We may initially deploy the service using cash transactions as we introduce farmers to digital payment processes. As WhatsApp deploys its WhatsApp Pay feature in India, we anticipate an even more streamlined end-to-end transaction flow.
Financial barrier mitigation: We will introduce buyer reputation metrics and reviews to mitigate disputes.
Capacity barrier mitigation: Work with the Solve network to identify highly skilled data scientists and supply-chain optimization researchers for collaboration.
What type of organization is your solution team?
How many people work on your solution team?
Twenty members work on the solution team; engineers, designers, operations managers, and administrative and finance staff.
How many years have you worked on your solution?
We have worked with many of the communities reached by Loop for twelve years, when Digital Green emerged out of a project within Microsoft Research’s Technology for Emerging Markets group. Our work on Loop, as a transport-to-market logistics service started over two years ago. The work on the self-service, WhatsApp-based chatbot began in April of this year to address challenges associated with the COVID-19 pandemic.
Why are you and your team well-positioned to deliver this solution?
Digital Green bridges information gaps by building the capacity of government extension systems to deliver localized advisory information, using low-cost, locally produced digital videos that leverage peer-to-peer learning and existing social networks. In India, more than 17,000 government extension agents have reached 1.8 million smallholder farmers (90% women) since 2008 using our video-enabled approach. Peer reviewed randomized control trials (RCTs) demonstrate that our video-based approach is up to 50% more effective at influencing farmer adoption and up to 10 times more cost-effective than traditional state-delivered extension. An RCT in India shows that farmers who adopt promoted practices realize yield increases of 46%.
While videos are the core delivery mechanism, Digital Green integrates other communication channels - SMS and interactive voice response (IVR) - to reinforce messages and improve service delivery for smallholder farmers. Demand for information and farmers’ ability to access it is demonstrated by our YouTube channel, with more than 58.5 million views and 213,000 subscribers.
Our successes in boosting farmer productivity taught us that to ensure farmers benefit fully from their engagement with Digital Green, they must be better integrated with agricultural markets. This led us to deploy Loop, KisanDiary, and a Supplier Directory, detailed above, equipping farmers with effective tools to realize more value from their produce. This experience, paired with deep and longstanding partnerships with farmers, agricultural buyers, and public and civil society extension partners, equips us to deploy the latest iteration of Loop effectively, drawing on consultative farmer-centric experience and structural food-systems expertise.
What organizations do you currently partner with, if any? How are you working with them?
To deploy Loop, we will build on our existing long-standing, large scale extension partnerships with state-government rural livelihood missions: JEEViKA, the Jharkhand State Livelihood Promotion Society, and the Odisha Livelihood Mission in the Indian states of Bihar, Jharkhand, and Odisha, respectively. We are also partnering with Reliance Industries AI chatbot firm, Haptik, which has a close association with Facebook and WhatsApp. We are also building on our existing network of local traders who have taken part in Loop and our Supplier Directory, as well as larger food buyers such as Ninjacart and Zomato.
What is your business model?
Loop’s primary beneficiaries are smallholder farmers in India. Typically, these farmers are already engaged by Digital Green’s government extension partners, who use our video-based approach to boost on-farm productivity. Digital Green supports these partners, with whom our interventions are co-developed, and who finance the intervention’s operational cost through capital and dedicated personnel.
We publish intervention data and impact results and open-source our technology so that outside organizations may replicate the approach and contribute to it. Digital Green’s India offices in Bhubaneswar, Bangalore, Delhi, Ranchi, and Patna provide video-based training and backstopping support, enabling partners to use our products and services.
Our government partners work with us because we share their objectives of improving farmer productivity and livelihoods and inclusively connecting them to markets with a focus on women and youth. Independent randomized control trials have proved our impact and cost-effectiveness and our partners recognize our ability to make their work more affordable, impactful, and scalable. Farmers, who want to sell more produce at better prices, take part in the intervention because it enables them to meet those goals.
The enthusiastic adoption of earlier iterations of Loop confirmed that farmers are willing to pay a commission for an effective route to selling their produce without traveling to traditional markets. Farmers will pay a 10% commission for all sales conducted through Loop. This commission will, once Loop has reached sufficient scale, support costs not covered by government co-investment.
Do you primarily provide products or services directly to individuals, or to other organizations?
Individual consumers or stakeholders (B2C)
Why are you applying to Solve?
As we seek to address technical barriers to chatbot deployment, we’ll look to Solve to help us collaborate with data scientists and supply-chain optimization researchers who can help with AI development, including Karen Zheng and Tata Center for Technology Design doctoral candidate Somya Singhvi. To build our internal capacity, we’ll look to the Solve network as we recruit additional staff and board members. We also hope to draw on the network for assistance with farmer- and consumer-level impact assessments to ensure that Loop’s deployment is translating into meaningful results.
In which of the following areas do you most need partners or support?
Please explain in more detail here.
We hope to continue our collaborations with MIT faculty and PhD candidates (noted below), who are developing chatbot technology and conducting local language speech recognition research.
We look forward to partnering with researchers at MIT’s Sloan School of Management to evaluate Loop’s impacts at both farmer and local consumer levels.
We plan to build on our existing partnerships with state government extension partners in Bihar, Jharkhand, and Odisha, with Haptik, and with our network of agricultural buyers as we introduce Loop and expand to additional Indian states, Bangladesh, and Ethiopia.
What organizations would you like to partner with, and how would you like to partner with them?
In the past, we’ve worked with Sloan professor Yanchong Karen Zheng and Tata Center for Technology Design doctoral candidate Somya Singhvi, who have studied Loop and supported our work through the development of price prediction algorithms. We'd like to continue to work with them as we automate farmer and buyer market linkages, and would also be interested in partnering with MIT Media Lab's Empathy Learning, Socially-Aware Agents (ELSA) project as we continue to iterate the Loop chatbot.
Do you qualify for and would you like to be considered for The Andan Prize for Innovation in Refugee Inclusion?
Do you qualify for and would you like to be considered for the Innovation for Women Prize?
Do you qualify for and would you like to be considered for the AI for Humanity Prize?
Explain how you are qualified for this prize. How will your team use the AI for Humanity Prize to advance your solution?
Through our work boosting the productivity of over 1 million farmers using targeted local video extension, Digital Green maintains data on the location, contact information, cropping patterns, and agronomic practices followed by each of these farmers. This existing network has allowed us to deploy advisories to farmers through diverse ICT channels, including video, IVR, and SMS. To better serve the communities with which we work, and to create an advisory approach that is resilient even in the face of crises such as the current COVID-19 pandemic, our WhatsApp chatbot that can deliver targeted advisory services directly to farmers. Drawing on our robust farmer dataset, we will target messaging tailored to farmers’ location, crops, and season. Loop will be deployed on top of this network to channel farmers’ production directly to the buyers who value it.
We've worked with MIT’s Yanchong Karen Zheng Karen and Somya Singhvi to develop price prediction and load optimization algorithms for our initial Loop deployment, and hope to continue our collaboration as we deploy deep learning to improve the chatbot and its speech recognition capabilities. We also hope to apply these tools to the crop image dataset provided by farmers to conduct automatic quality assessments of produce and mitigate potential sales disputes or inaccurate listings.
Do you qualify for and would you like to be considered for the Future Planet Capital Prize?