05 Jan 2017

Creating an adaptive culture and speeding up the time it takes to adapt

How human-centered design helped CARE's social enterprise become more adaptive.

Adaptive_management_icon.png

This blog is part of our series on adaptive management.

Krishi Utsho, or KU, is a CARE social enterprise focused on generating systemic change to improve availability and accessibility of agricultural inputs and services in rural Bangladesh. KU does this by supporting the development of a network of microfranchise input supply shops which serve last-mile farmers. One challenge in the input supply system is that supply companies and their end customers work off different information, as information is retold and becomes distorted so that input supply companies don’t understand the needs of the farmers. One of KU's key goals for the market system is to improve communication between input supply companies and small-scale farmers so companies are more responsive to farmers' needs. 

In early 2016, the KU Team was selected to participate in CARE’s Scale X Design Impact Accelerator, which aims to reduce the time it takes for an innovation to scale from an idea to widespread impact. During a lab facilitated by experts at GRID Impact, the team learnt about Human Centered Design (HCD) a creative approach to problem solving. This starts with developing empathy for the people you are designing for, and then rapidly designing and testing prototypes to quickly learn about what works and what doesn’t in real world situations. The KU team immediately saw the potential that HCD’s approach to rapid prototyping could have in helping the team to become more adaptive. Applying the lessons of HCD, the KU management team spent a day in an input supply shop to get hands-on experience, observe customer behavior, and ask about customer preferences. At first the shop owner and customers were taken aback to see the KU team taking such a hands-on approach in managing the shop for a day, but they quickly accepted their presence.

What did we learn about how farmers want to receive information?

The KU team learnt how farmers prefer to hear information about product promotions. Most farmers indicated a preference for "miking" (promotions using a microphone and loudspeakers) as both farmers and other household members pay attention to this type of advertisement. Farmers said that TV and radio advertisements were easy to miss, and illiterate farmers can't use leaflets. Farmers said they tend to ignore text and voice message promotions, considering them spam. This last point was especially important because KU had been exploring the potential to support development and testing of a voice messaging system targeting farmers. 

care milking.jpg"Miking" - promotions using a microphone and loudspeakers

If miking works, then what’s the problem?

Unfortunately, although miking works effectively to share messages with farmers, it did not solve the underlying problem. Miking has no feedback loop for KU and input suppliers to learn from the farmers about their demands, or their satisfaction with the products and services they receive. And the absence of a feedback loop between farmers, input shops and suppliers is a key market failure KU needs to address to get more appropriate inputs to smallholder farmers. 

So now what?

We focused on prototyping solutions to test at field level. This ranged from ICT-based messaging systems to simple comment boxes. The most popular solution with farmers was a suggestion box where customers fill out a feedback form after their purchases. The form uses pictures to ask self-explanatory questions about current and previous purchases. To motivate farmers to fill out the form, KU will randomly select a farmer through a lottery-style prize drawing every month. KU is now testing this prototype in a single store. The response from customers so far has been positive. 

How is Human Centered Design helping us be more adaptive?

Applying HCD helped the KU team be more adaptive in two important ways. First, by spending time in the field working in an input supply shop and interacting with customers, we were able to better understand the problems that we were trying to solve. Compared to more traditional research methods, this experience provided us with a different, and important, perspective on the challenges as well as the needs and interests of the stakeholders. It also helped us become more “farmer-focused” in terms of the types of solutions we design. The insights continue to guide our decision-making, we are more willing to question our initial assumptions and have developed a more adaptive culture within the entire KU team. We plan to continue to do this type of hands-on action research periodically going forward. 

Second, the use of rapid prototyping of rough designs and approaches, as opposed to more traditional "piloting" of fully fleshed out ideas has helped us generate learning about what seems to be working more quickly and more cheaply. This has encouraged us to be more flexible in our approach and test more "out of the box" ideas. Rapid prototyping allows us to iterate quickly and help the team to be more nimble and adapt strategies and approaches in response to what we learn through the prototyping process. For instance, through our new approach to prototyping we are now learning a lot about geographic variations in farmer demand for various inputs, and we can adapting our strategies based on this geographic differentiation.

KU's long-term strategy is to eventually transition to an ICT-based system to support two-way communication between input suppliers, input shops, and farmers at scale. We will build on our experience with the suggestion box by continuing to prototype and explore a number of ICT options such as direct "comment lines" to facilitate this two-way communication. Our approaches, strategies, and activities will continue to adapt based on what we learn through applying HCD to the challenges we face.

This blog is part of our series on adaptive management.

Add your comment

Sign up or log in to comment and contribute.

Sign up

Contribute

Write a blog post

Share