Mobile Market Linking Service
Clients: CGIAR, CIMMYT and Viamo
Location: Nepal
Duration
-
Design: 3 months, from May to July 2018
-
Impact: Ongoing, with expansion to Indonesia, Niger, and Rwanda.
My Role
-
UX Research & Design
-
Innovation Team Facilitator
-
Horticulture Value Chain Research and Analysis
-
Prototyping
-
Usability Testing
​
Press: CGIAR Impact Stories
​​
Brief
In Nepal, farmers struggle to access food system information and market linkages due to an inability to make connections with required stakeholders. Research and analysis showed that disconnections in value chain relationships could be resolved through digitizing communication processes, thereby improving accessibility and cutting out the wait times associated with farm gate sales. Our innovation team successfully designed an interactive voice service with an algorithm, which matches farmers with buyers, labourers, transportation, and other various stakeholders and uses voice-based messaging to create matches via SMS.
DISCOVER
User Research
Nepalese culture is built on community; building relationships and establishing trust between stakeholders would be key for any process – digital or otherwise – to work.
Our user research focused on key participants along the value chain. We started by conducting one-on-one interviews with actors in the local horticultural market, including farmers and consumers. Then, we facilitated two ideation workshops: one with small-scale farmers and the other with wholesale and retail vendors.
From this research, there were a few key takeaways:
-
Successful market linking in Nepal could not revolve around the transaction process, but instead needed to focus on fostering connections.
-
There was certain key information that a buyer needs to know, as well as a specific amount of information most farmers would be willing to share.
-
Precise language and phrasing choices would be vital for the AI to share the right information while still building relationships. For example, “I’m a tomato farmer expecting a 5-10 kg yield per week this season” was more appropriate than “I have 10kg of tomatoes I need to sell tomorrow” when creating a mobile market link between seller and buyer.
Market Research
While examining users, it was important to research the horticulture value chain. Working with an agronomist from the University of Pennsylvania, we conducted both desk and field research to answer the following questions:
-
What are the local and international economic forces that influence the horticulture marketplace in Nepal?
-
Who are the key actors in the Nepalese horticulture value chain?
Then, we then reviewed comparable marketplace systems, facilitating informal discussions with their creators in order to best understand the challenges involved.
​
Our market research provided a few key takeaways:
-
Nepalese food systems are informal and dis-aggregated. These issues impeded the flow of information and resources, which limits the ability to improve income or livelihoods.
-
An increase in the dissemination of market price information may support more equitable negotiations but would still leave farmers with minimal bargaining power if there is a reliance on farm gate sales.
-
Challenges exist in making connections between stakeholders, whether it be for farmers seeking services such as labor and transport, or for buyers seeking sellers of product.
Define, Ideate, Test
It became apparent mid-way through the project that our team had hit a wall in the brainstorming process. The outcomes and takeaways from the discovery phase did not smoothly transition the team into the ideation phase, and innovation on my end was required to jumpstart this task. I facilitated a five-day, collaborative design sprint, creating an environment for us to brainstorm in new ways as a team.
At the end of the sprint, we had designed two different prototypes of the market-access system. The first was built around interactive voice response (IVR) and the other on social media.
Rapid Prototyping & User Testing
The prototypes were tested through a series of focus groups and interviews to assess the usability and feasibility of the product. Participants included numerous individual farmers, four farmer cooperatives, two agri-wholesalers, three agri-retail outlets, and two commercial buyers.
Interactive Voice Response Protoype
The IVR Prototype involved farmers calling a number and answering questions about their crop, the quality, quantity and price. We made observations and asked follow-up questions post-call. Findings included that integer questions which required farmers to type in amounts using the keypad proved difficult, with many users holding down buttons that would lead the system to register their responses multiple times. Farmers also didn’t differentiate between quality levels in their product, and found questions such as “How many kg of high quality tomatoes do you have” and “How many kg of average quality tomatoes do you have” repetitive.
The Social Media Prototype
The Social Media Prototype included a Facebook page with advertisements of produce for sale. We showed the Facebook page to two head chefs from hotels in Surkhet, a small retail shop, and numerous farmers. Findings included that smartphone access was low amidst farmers, and those with smartphones often had low digital literacy. Additionally, the two head chefs said they would prefer dealing with a middleman who could aggregate what they needed versus scrolling through a page and contacting various people.
Prototype Key Takeaways​
-
Forming lasting business relationships is the most important factor for farmers when they are using a service to access new aggregators and/or buyers.
-
Farmers need to be connected to buyers as far up the value chain as possible to derive a higher price for their goods.
-
Farmers and buyers are comfortable being linked with one another by only being asked a few simple questions about their situation.
-
Phone calls are farmers and buyers preferred method of communication for new and existing business communication.
Our conclusion from these insights was that the IVR system had the most potential, given the usability across education levels, literacy, and access to technology.
​
After making this decision, we were able to simplify the questions in the IVR pilot in order to maintain the minimum parameters needed to make a useful match while maximizing functionality. We settled on multiple choice questions that provided details about location, ranges for average yield per week, prices for wholesale or retail, and availability of transportation.
Deliver
We continued iterating the system during the pilot test of the IVR system, collecting data and analyzing engagement in order to improve the platform. We conducted call back surveys for those who registered in order to learn what challenges users were still facing.
​
Since its formal launch, there have been over 45,000 unique callers within four months. The service is active in Nepal, connecting farmers and buyers across the country. Additionally, since the initiative proved successful, it is now being implemented in Niger and Indonesia to alleviate the burden of farmers finding buyers for agriculture products during a country-wide lockdown due to COVID-19.
Reflection
This three-month long endeavour brought together a multicultural team comprised of people at different points in their careers, working on a complex project. As the team facilitator, it was necessary that I created an environment that encouraged creativity and problem solving, getting members excited about working together. From an internal perspective, I learned the importance of innovative team facilitation, encouraging the team to learn and iterate on our past failures, and taking the time to create an environment that encourages experimentation in problem-solving.
From an external perspective, there were many things learned from this process. Primarily, it was an interesting challenge to design a completely new digital system for individuals who do not heavily rely on mobile technologies to do their work. To find the best possible option, we needed to try out as many design methods as we did potential solutions.
