Project Everest

Project Summary
Work Update

[Project Summary]: FarmEd Fiji Project Summary July 2018

Felix Zerbib
Felix Zerbib | Jul 27, 2018 | in FarmEd - Fiji


In the past, FarmEd has explored supply chain management, face to face consulting, drone consulting and agricultural expertise workshops. While there is value in the above products, moving forward, FarmEd aims to focus on the development of the app itself as the app has proven to be the most scalable and accessible way of providing agricultural expertise. Over July 2018, the 3 FarmEd teams focused on determining the viability of the FarmEd app:

  • FarmEd I tested the user experience of a potential app, particularly the ability of farmers to give and receive information;
  • FarmEd II tested the ability and willingness to pay for agricultural expertise through offer and currency tests; and
  • FarmEd III tested the validity of  the market for the future FarmEd app by collecting data on smartphone and internet use/access, as well as assessing language capabilities.


Overall, our findings concluded that in terms of inputting accurate information, farmers provided the most satisfactory information through the text method, with the success rate being 22% more accurate than information provided through a phone call. Close-ended text questions also received a 58% response rate, much higher than the 44% for calls and 37% for open-ended text questions. Furthermore, in terms of outputting advice and following instructions, farmers interacted best with the videos, followed closely by tables and then voice calls. This is important, as the videos and voice calls were still interacted with satisfactorily, regardless of the farmers English language ability. To summarise, the recommended method of input is close-ended questions with multiple-choice responses, and the recommended method of information output is voice calls, with videos and tables being highly viable options.

Of the 37 people offered the MVP currency test, 34 showed they were willing to pay - that is 92% successful currency tests. This measurement was designed to test whether there is a market for the app and show that farmers were willing to invest money in agricultural advice.

Of those who were willing to pay for the MVP demonstration, 31 expressed interest in a continuing subscription to the product by signing an MoU. The currency tests showed that a large percentage of people we spoke to, from a diverse range of circumstances, were willing to pay for a technology based farming management recommendation system. A breakdown of content that farmers would want to see in such an application was also established, with the 5 most common features desired being:

  1.  Soil measurement and analysis
  2.  Cropping recommendations
  3.  Pest management advice
  4.  Farming technique recommendation
  5.  Seasonal recommendation

Regarding the existence of channels and platforms for rolling out the FarmEd prototype, data was collected and analysed from various sources regarding:

  • Demographic Information
  • Phone and Internet Access
  • English Comprehension
  • Chatbot Ability

Overall, it was established that approximately (52.6 ± 18.0)% of Fijian farmers have access to a smartphone, and have on average a moderately high ability to read and write as well as interact with a chatbot and photo portal (ability scattered between 3 and 4 on a scale of 1-5).


        Moving Forwards


Now that we have a thorough understanding of the agricultural landscape in Fiji and an idea for a final product, the logical next step is to develop and implement a basic prototype of the final product.

The focus over summer 2018/19 should be to distribute a prototype of a platform that provides farmers with agricultural expertise and then gain constant feedback on the way farmers interact with the prototype. The prototype should be distributed to a number of farmers that teams build strong relationships with and continuously visit for feedback over the three month period.

The current FarmEd team envisions that customers' interaction with the prototype will be iteratively analysed, and this will be done with three larger goals in mind:

Front-End Development : distribute the prototype and conduct 20 close case studies with farmers using the prototype. Ultimately the core aim being to gather feedback on usability. The top priority for Team I is to determine how customers interact with the prototype, whether it be the ease of use itself or the content that it provides.   This team should work closely with farmers, and try to understand the progression of their use of the prototype.

 Back-End Development : gaining 120 data points every month. For the app to successfully provide agricultural recommendations, we need a database of ‘best practice’ in Fijian agriculture. This database should contain information about how to best grow certain crops and the price crops can be sold at. A combination of agricultural and market data will enable the app to give farmers recommendations on how to increase the yield and profitability of their farm. In the past, research into best agricultural practice has been attempted by having our own experimental farm. While there is value in this project, it is more efficient to build the database by looking at what existing successful farmers do right.

Scalability : explore the systems that would enable FarmEd to scale and be fully independent whilst there are no teams in country. Currently one of FarmEd’s biggest blocks is its reliance on us to empathise, run trials, and provide feedback. If this can all be streamlined and scaled then it would allow for enormous data collection and feedback gathering both while teams are in and not in country.

For more details, see the full project summary attached.

edited on 6th September 2018, 01:09 by Justin Hakeem

Isaac Crawford Jul 27, 2018

Hey Will, I would love for you to have a look at the overarching future actions/ experiments for FarmEd and gage your initial thoughts on effective ways to collect data on efficiency of customer acquisitions.

There will are a couple more posts in the works explaining the future actions/ experiments in more detail.

Users tagged:

Reply 0

Andrew Vild Aug 6, 2018

Status labels added: Project Summary, Work Update

Reply 0