EverYield
What is EverYield? EverYield is an B2B, agribusiness SaaS platform designed to deter deforestation by incentivizing major commodity stakeholders to engage in the transition to sustainable agriculture.
Summary of Research: Remote, mixed-methods, bilingual research was conducted across three distinct stakeholder groups to address the highest-risk assumptions for the EverYield MVP. The research aimed to assess farmers' sentiments toward sustainable transition methods and explore how to develop resources, such as insurance and carbon credits, that farmers could utilize through EverYield.
Constraints: Budget constraints required the team to be creative in obtaining meaningful quantitative insights on a smaller scale, while language barriers added challenges to maintaining time efficiency.
Outcomes: The research team successfully conducted 36 hours of bilingual research over a two-week period by employing a show me/tell me approach, which allowed for the quick and cost-effective gathering of quantitative insights with qualitative depth.
These insights were pivotal in shaping the direction of the MVP, which subsequently secured an additional $1M in follow-up funding.
Project Type
B2B
Research Type
User/Market Research
Research Characteristics
Quant/Qual Generative/Evaluative Attitudinal/Behavioral
Methods
Mixed Methods Participatory Design
Tools
Miro, Airtable, Google Meet, Fireflies, Notion, Figma,
Company
NAX
Role
Lead Researcher & Strategist to a Team of 3
Timeline
10 weeks; Oct-Dec 2023
Context
Understanding systems and theory of change to inform the MVP
A multinational insurance client engaged our innovation consulting firm to develop a product aimed at reducing deforestation on a large scale.
Using systems analysis, we identified key drivers and stakeholders contributing to global deforestation. We determined that targeting the agriculture sector would have the most significant impact.
However, the challenge remained: Could stakeholders be motivated to change their behavior? Further research revealed that commodity traders faced mounting pressures from impending legislation and consumer demands to address deforestation in their supply chains.
With these insights, we developed an MVP for EverYield, a SaaS platform that enables commodity traders to detect deforestation in their supply chains and connect farmers with the resources (insurance and/or carbon credits) needed to transition to more sustainable practices—ultimately reducing the likelihood of future deforestation.
The team had made the decision to focus the MVP on coffee in Minas Gerais, Brazil due to the relative sophistication of coffee farming in this region.
I was then tasked with identifying, prioritizing and leading a team of 3 to conduct research aimed at addressing the MVP's highest risk assumptions.
Stakeholder mapping
Causal Loop Diagram: Deforestation Drivers
Causal Loop Diagram: Theory of Change
Timeline
The project was conducted over the course of 10 weeks.
Assumption Identification; Scoping the Project: 1 week
Study Design: 1 week
Identifying Needed Resources and Setting Budget: 1 week
Recruiting Translator and On-Ground Recruiter: 1 week
Recruiting and Scheduling Participants: 1 week
Conducting and Synthesizing: 2 weeks
Analysis: 2 weeks
Share out with Stakeholders: 1 week
Team
I worked with a cross-functional team to set the research scope and a team of (3) to conduct, synthesize, and analyze the research
My Role and Responsibilities:
Determine research priorities with broader business, product and engineering teams according to business and user needs
Setting of timeline and budget
Coordination of timeline for conducting sessions across timezones and language barriers
Study design and execution
Analysis and share out of study insights
My Team:
1 research assistant: collaborated on note taking, synthesis and analysis
1 on-ground coordinator in Brazil: worked with to establish timeline, recruit participants and distribute incentives
1 translator: worked with to compile translated research materials for sessions
Determining Research Scope
I worked with the product, business and engineering teams to identify and prioritize Everyield's highest risk assumptions
Description: I worked with the broader product, business, and engineering teams to scope the research based on both business and product needs. I then categorized the assumptions according to whether they dealt with the desirability, technical feasibility, or business viability of the product.
Based on the assumptions, the scope of the research would be geared around the topics of:
Determining what might prevent the adoption of certain sustainable land management practices and overall what practices would be most feasible for adoption
Determining the desirability of carbon insights and insurance offerings on the platform
Activity: Assumption identification and prioritization workshop
Deliverables: Assumption backlog; Prioritized assumptions to act as research goals
Tools Used: Miro, Airtable
Outcomes: Team alignment on most critical research priorities for immediate research project as well research priorities for the near future
Example Assumptions:
"EverYield can provide automated suggestions regarding sustainable land management practice adoptions for mid-sized coffee farms in Brazil."
"Farmers in regions such as Brazil are unable to pay for insurance premiums therefor they will be unwilling to engage with an EverYield insurance offering".
Assumptions Mapping Workshop
Assumptions Backlog
Developing The Research Plan
With the high risk assumptions as a foundation, I developed the research goals/questions and laddered them up to business objectives in order to obtain leadership buy-in
Description: I translated our highest priority assumptions into research goals that spoke to both the business and product needs of the MVP. I then translated my research goals into research questions that were actionable, specific, and practical.
I paid special attention to detailing how answering the research questions would serve the goals of the business to facilitate easier budget approval and leadership buy-in.
Activity: During the development of the research plan, I held frequent touch points with the business, product, and engineering teams to ensure alignment on the foundational aspects of the research plan.
Deliverables: Defined research goals and questions (6) for the project
Tools Used: Notion
Outcomes: Defined scope of research
Example Research Question Aligned with Business Goal:
Business Goal: The business needs to determine land management practices that farmers are most likely to adopt order to ensure product viability and problem-solution fit.
Research Question: "What are the pain points and influence drivers of farmers who have transitioned to sustainable land management practices?"
Method Selection
I selected a mixed qual/quant approach to best serve project constraints
Description: I realized that for each research question, it was valuable to understand the answer relative to a scale (for example: EverYield would not only suggest one land management practice to a farmer…we needed an understanding of how different variables ranked against each other and why they were ranked that way).
Additionally, I knew the sessions would need to be conducted almost exclusively in Portuguese. The participatory design methods would help to ease the language barrier by providing visual stimuli and less translation time during analysis.
For this reason, I elected to conduct mixed method research with a combination of participatory design methods such as scale/stack ranking and concept testing. This approach would enable to team to gather both quantitative data as well as qualitative data to add color to the quantitative responses.
Methods Used and Rationale:
Participatory Design: included card sorting, rankings, and concept testing as well as followup questions to dig deeper into the "why" for a mix of quantitative and qualitative insights
Note: I realize focus groups could be used as a method to drive efficiency in this scenario but chose not to utilize this method as, in my experience, focus groups are not the most effective method to elicit genuine responses.
Tools Used: Miro
Participatory design exercise for Farmers
Concept testing exercise for Cooperatives
Participatory design exercise for Cooperatives
Screening and Recruitment
The research questions necessitated the recruitment of 3 different participant profiles-each with their own screening criteria
Description: Our research questions were best suited to be answered by four separate audiences: Brazilian coffee farmers that had transitioned to sustainable land management practices, Brazilian coffee farmers that had not transitioned to sustainable land management practices, Brazilian farm cooperatives, and Brazilian agronomists.
Constraints: The team did not have any existing connections with any of these audiences. Additionally, English is not widely spoken in rural Brazil.
To solve for these constraints, I worked through an urban cooperative to enlist the help of Thaissa, a bilingual agronomist willing to make connections with all the audiences we needed to speak with.
Participants Recruited:
(6) Agronomists
(6) Transitioned Farmers
(6) Non-Transitioned Farmers
(6) Farm Cooperatives
Incentives: Each participants was offered monetary compensation. In speaking with our on-ground recruiter, we also learned that household wares such as coffee mugs and blankets were valued by our participants so we expanded our incentives to include these items as well.
Example Screening Questions:
[for agronomists]: "Do you have at least 5 years of experience advising farmers in the adoption of sustainable land management practices?"
[for transitioned farmers]: "Have you fully modified at least one aspect of your coffee production to be more sustainable according to national standards within the past 3 years?"
[for cooperatives]: "Are at least 25% of the farms within your cooperative utilizing sustainable practices?"
Planning the recruitment participants in Brazil with our on-ground coordinator, Thaissa.
Compensation, especially when working with vulnerable populations, is incredibly important. When we learned farmers valued household wares, we were quick to provide merchandise (produced locally in Brazil) in addition to monetary compensation.
Conducting Research
A three person team consisting of our translator, on-ground coordinator and myself conducted 24, digital, 1 on 1 sessions over the course of 2 weeks and 3 international time zones.
Sessions Conducted: (24) 1.5 hour sessions with (1) 15 minute break. Prior to each session, each participant reviewed and signed a written explanation of purpose as well as a participant consent form. Additionally, each session was recorded and transcribed.
Platform Used to Conduct and Record Sessions: Google Meet; Fireflies.ai
Sample Questions and Exercise Prompts:
[for farmers] "Can you tell me about the last time that you felt pressure to mitigate risk within your farms? What risks were these and where did the pressure come from?"
[for cooperatives]"Have you ever had a farm within your supply chain that had contributed to deforestation? How did you approach this situation?"
[for farmers] "Use the sticky notes to the left to write down all risks you are concerned with addressing in your farms. Once listed, please plot the risks on each of the spectrums below (examples included ease with which the risk can currently be addressed, willingness to pay to address, etc."
[for cooperatives] "(shows the participant a low fidelity mockup of a proposed feature) Do you currently have information like this regarding your farms? If no, what would understanding this information enable you to do?"
Conducting a 1 on 1 session with a Brazilian coffee farmer via Google Meet.
Conducting a 1 on 1 session with a Brazilian agronomist via Google Meet.
Synthesis and Analysis
After translation of the material, the team synthesized according to research objectives and then used thematic analysis to draw insights from the data. This was done entirely in Miro.
Description: Given the sessions were conducted in Portuguese, it was necessary for our translator to translate the content before the team was able to synthesize.
Once translation was complete, myself and (1) research assistant synthesized and analyzed the material entirely in Miro. Since the research artifacts already existed in Miro, it was relatively straightforward to synthesize all the insights and compare across all participants directly in Miro.
To synthesize, we utilized a method of tagging insights according to our research objectives, making sure to attribute each piece of evidence to the corresponding participant.
To analyze, we then took the tags we created and worked collaboratively to group and draw out common themes/insights from the material.
Note: we also took special care to re-listen to the recordings of each session for any pieces of information that might not have been captured in Miro during the sessions. We were then able to capture these pieces of information to include them in our analysis.
Team: myself, (1) research assistant, and (1) translator
Tools Used: Miro
Example of a completed exercise from a research session
Snapshot of analysis in progress
Outputs and deliverables
The insights generated were encapsulated in a research report shared with the broader EverYield team.
These insights were also instrumental in the iteration of existing personas as well as the formation of a new journey map detailing farmer behavior.
Research Report: the main deliverable for this project was a robust, 25 page research report. The insights obtained from the farmer, agronomist, and cooperative sessions centered around the following:
The behaviors and attitudes of farmers when it came to certain sustainable land management practices
The desirability associated with offering direct financing for sustainable farming transitions via the platform
The technical limitations of sustainable land management practices and the willingness of buyers to pay a premium for these goods
The sentiment of farmers towards transition vehicles such as carbon credits or insurance options
The sustainable land management practices most feasible for farmers to adopt and the efficacy of suggesting these practices remotely via the EverYield platform
Farmer Journey Map: since many of the insights obtained had to do with farmer mindset and preferences, I believed that it would be a missed opportunity not to encapsulate these insights into a journey detailing the farmers actions and viewpoints prior to engaging with EverYield.
In total, three farmer journey maps were made to shed light on three different farmer segments: the skeptic farmer, the sustainable farmer, and the conventional farmer.
Farmer Personas: additionally, I used the new farmer insights to iterate on the previous farmer persona and provide three new, segmented farmer personas to account for attitudes towards and past experiences with agricultural transition.
Outcomes:The report seamlessly connected assumptions, business goals, and research questions as to provide actionable insights to the EverYield team as well as elicit buy-in from leadership to
Mapping the journey up the first touchpoint with EverYield for a farmer segment
Sample slides from the deck share out with the broader team and leadership
Conducting a 1 on 1 session with a Brazilian agronomist via Google Meet.
Impact
The research insights were instrumental for moving product development forward and securing additional funding.
The research team successfully conducted 36 hours of bilingual research over a two-week period by employing a show me/tell me approach, which allowed for the quick and cost-effective gathering of quantitative insights with qualitative depth.
These insights were pivotal in shaping the direction of the MVP and fleshing out the imperative "solution" aspect of the EverYield platform, which subsequently secured an additional $1M in follow-up funding.