Angie Fu
Mo'Coffee Uganda
Project Background partnered with Kyagalanyi Coffee Ltd. (KCL), a coffee buyer in Uganda, to establish a solution aimed at improving the uptake and use of mobile money with their farmers. Our small team consisted of a business designer and design researcher as co-project leads, a communications designer, and myself as an interaction designer, in partnership with the UN Capital Development Fund, and the Bill & Melinda Gates Foundation starting in September 2016. The project is currently closing up it’s live prototyping phase, and will continue stewardship through our partners until the end of the coffee harvest season in early 2018.
Interaction design
UX design
UI design
Interactive prototyping
Communication design
Coffee farmers sorting their harvest
Theory of Change
Coffee farmers in rural Uganda are among the poorest in the country, lacking in adequate exposure and familiarity to financial health tools and services like bank accounts and loan products. Often, these types of tools are out of reach for the poor, who can’t afford to pay the high fees or spend time travelling to distant, unwelcoming banking institutions. However, over the last decade mobile money has begun to shift the paradigm surrounding digital finances, as mobile network operators (think Verizon, AT&T) now have mobile wallets attached to every phone number—so as long as you have an active SIM card and a phone, you can load, send, and receive mobile money.
This has a lot of implications; the most practical of which assures that the majority of Ugandans—regardless of socioeconomic status—now have access to this more open form of digital finance, and with more use and exposure, can develop the confidence to use other, more advanced services that contribute towards poverty alleviation.
Design challenge
Prior to the 2016 coffee harvest season, KCL worked with a local developer to create a web-based payment platform that would empower staff members on the ground to start paying farmers in mobile money. However, staff members were hardly using it, and few farmers were asking for and accepting mobile payments. Our design challenge was simple:
How might we increase the uptake and use of mobile money of rural Ugandan coffee farmers?
Boots on the ground
Did I say simple? I meant whatever the opposite of simple is. In our research trip to the field, we quickly observed the complex ecosystem of influences and actors far beyond what we had originally anticipated. Led by our design researcher, we created research guides and activities to learn and understand the dynamic between coffee farmers and KCL, as well as perceptions of and experiences with mobile money.
During this initial research trip, we conducted interviews with 8 KCL staff members, 3 mobile money agents (foot soldiers of the network operator who help convert cash to mobile money and vice versa), 4 merchants, and dozens of coffee farmers.
Critical painpoints
Bottleneck —
I led interviews with KCL staff members to better understand why the existing web-based payment platform wasn’t being used. Asking for a demonstration would typically reveal an avalanche of insights—the laptop was hidden away under boxes and paperwork, took minutes to startup, relied on a USB WIFI adapter to connect, and often required running a generator for power. Logging in required a lengthy email and password that aggravated forgetful staff members, and the cumbersome UX meant that entering farmer data could take up to 5 minutes. Even worse, depending on the network strength, the harvest payment could take between 1-6 hours to arrive in the farmer’s mobile wallet. For staff members, the platform was an extreme undertaking in every capacity—from performing basic functions to coaxing anxious farmers that their mobile money would arrive in time.
Lacking both in value and understanding —
While mobile money access in Uganda is widespread, we saw in interviews that farmers don’t know how or why to use it. Those that do have active accounts exclusively use them to send and receive remittances from distant family members, meaning that every interaction with a mobile money agent is coupled with steep cash out fees. Outside of remittances no other mobile money services had made it out to these rural communities, so farmers simply weren’t seeing the value in keeping their money in the system.
Trying to load the web-based platform
Establishing a push—pull ecosystem
During synthesis, the team realized that just
mobile money more efficiently to farmers would only address part of the problem. Without a
a desirable, tangible mobile money use case—there would be little value in accepting it in the first place. To solidify and elevate this connection between a push—pull ecosystem, we saw an urgent need to provide a layer of
for farmers to start internalizing the short and long-term value of mobile money adoption.
Push —
Improved payments
To more seamlessly push mobile money to farmers, we began to imagine a more optimal device that would cater to KCL staff member workflow. They needed an easily transportable tool that wouldn’t rely so heavily on a generator or WIFI adapter, so we explored the pros and cons of utilizing a smartphone. However, after considering the staff’s low tech literacy and need for increased readability and target sizes, we moved forward with a tablet as our device. To illustrate the value of this more useable payment app, I started with rough sketches of key user flows and migrated into Sketch and to bring the tool to life.
Inspire user confidence
It was necessary to prioritize and establish safeguards for the most common issues staff members experienced, like forgetting their username and password (takes 3 days to reset), running out of mobile money (up to a week to top-up), and workflow being halted when there’s no network connection. To address the issue of network connection, I added a ‘Payment Queue’ feature that could be accessed from each screen, where staff could store payments until the connection was strong enough. On top of this, the interface was designed with intentionally high-contrast to maintain readability when the sun is bright, large text fields and otherwise fairly minimal layouts in order to reduce visual clutter.
Payment transparency
The most persistent issue faced by both staff and farmers was the uncertain, lengthy processing time of mobile payments—in part due to the fact that staff members had no visibility into what was going on behind the scenes and therefore weren’t able to give farmer a time estimation. We designed a ‘Pending Payments’ feature that gives staff members visibility into the precise progress of payments and clear error messages to help them better communicate issues with farmers. An additional part of my role was working with our Ugandan development partner to prioritize backlog issues that would move us closer to our goal of making mobile payments faster than cash. Over the course of several sprints, we’ve been able to move the needle from 1-6 hours to 5-15 minutes, and are still working to achieve our reach goal of a ~2 minute average.
UX walkthrough for our development partner, communicating design intent
Pull —
Generating demand
Given the steep mobile money cash out fees, we needed a use case that would keep a farmer’s mobile money in the digital ecosystem, thereby avoiding cash out fees. After comprehensive research to further understand mobile money use cases that would be of value to coffee farmers (i.e. paying for school fees, bill payments, rent, agricultural inputs, etc.), and taking into account executional feasibility, we designed
—a shop situated at KCL stations that would exclusively accept mobile money for payment of goods.
The location was critical, as it’s proximity to the station removes the time and hassle for farmers to travel to more distant markets or village merchants. This not only reduces the cost of roundtrip travel, but also valuable time—time that could be better spent harvesting more coffee or otherwise strengthening farming businesses.
Immediate trust and value
We recruited local farmers to own and operate the Mo’Agro prototype, leveraging existing trust in farmer communities. We equipped these managers with starting inventory, receipt books to document mobile money sales, as well as bold, cohesive brand assets to support and legitimize the shop. Ultimately, we aimed to establish an immediate, trusted, and safe environment for farmers to learn how to use mobile money to pay for goods and services.
Godfrey, our Mo'Agro prototype manager, helping a farmer make a purchase
Educate —
Empowering farmers
To maintain and amplify the momentum created by a strong push and pull, we needed to ensure that farmers were internalizing the value propositions of mobile money. We worked with a certified financial coach from our NY office to co-design a financial education class aimed helping farmers make an emotional connection between budgeting, savings, and achieving family goals.
The more we narrowed in on the curriculum, the more targeted our ongoing research was—ensuring that we were providing the most compelling topics and family goals, and striking the right class format and setting. We have since done 3 iterations of live class prototypes, trained three teachers, and rolled out our latest curriculum to over 800 coffee farmers.
Class considerations
To supplement the class curriculum, we designed several dozen iterations of an activity worksheet that would serve as a long term reminder of the goals farmer’s had set in class. Due to low literacy levels, we focused on making each activity highly, if not entirely visual—which was challenging as it was coupled with the fact that these worksheets eventually needed to be sourced and printed as simply as possible. When viability was less of a focus, we kicked off the session with a visioning exercise; asking farmers to cut out and collage images of goals they had an emotional connection to—e.g. a cow to generate supplementary income to pay my daughter’s school fees. In rolling out our longer term prototype, we adapted the activity such that farmers could identify their goal by coloring in line illustrations. This small change significantly reduced the cost of production, while still maintaining the interactive, highly visual nature of the worksheet.
Early success stories
In January 2017, we tested our first comprehensive class prototype in three classes, making up about 40 KCL farmers total. We’ve since been doing qualitative follow-ups to inform updates and changes—i.e. to see what was working, where more guidance was needed, and what behavioral changes were happening if any. Out of the eight in-depth follow ups we’ve made since then, three have successfully saved up for the goal they set at the beginning of the year with mobile money! These three were all women, adding proof to an early insight we had—that Ugandan women are typically more consistent and goal-oriented than men.
Mo'Coffee —
Where we are now
In Uganda, the coffee harvest season typically runs from August to January, so our three live prototypes will continue to run with our remote stewardship and guidance until January 2018. We’ve rolled out the
prototype (tablet app build) to 5 of 6 KCL stations, and the
prototypes to 2 of those 5. While it’s early in the current season (as I’m writing this), based on the initial correlation analysis collected from the stations with all three prototypes, versus the stations with just the tablet app, there is an optimistic uptake of mobile money acceptance with farmers. We’re using these quantitative insights to inform our ongoing qualitative remote research. Talk to me later in the year and hopefully I’ll have more to share about this beast of a project!
Previous Project
Airtel Tanzania
Back to
Top of page
Back to
All projects
Next Project
CGAP + Amret MFI