Project Everest

Experiment Results

[Experiment Results]: Solutions - Currency Testing - FarmEd Fiji July 2019



Lean Phase: Solution (sales / early adopters)



Key Outcomes:  This experiment surveyed a total of 60 people from 10 villages, to determine whether smallholder and subsistence farmers demand a solution (the app) for the diagnosis and treatment of pests and diseases and are willing to pay $1 for a one-month subscription of the MVP.


Timeframe: 2 weeks


Summary of results: 

Number of farmers that bought the service: 43

Percentage of farmers that bought the service: 71.7%

Experiment result: Green light (success point - 60%)



The raw data can be found here:

A summary of our results can be found here:


The results below are collated from farmers who only purchased the app. For the full results and data of each farmers interview see the raw results in the spreadsheet linked previously. The results have also been summarised in the charts below.

  • Average ability of farmers to control pests and diseases (Likert scale): 2.692/5 
  • Average income from the farm: $600 per month - Median $200 per month
  • Average farmers’ estimated yield loss due to P/D: Approx. 25% [Insert pie chart] 
  • Land size: Home - 20.5%, Community - 56.4%, Commercial - 23.1%
  • Success level of the farm (All farmers grade their farm as being level ‘3’ and above on the Likert scale- this means that their farm is enough for the farmers to survive and live comfortably). Mean: 3.42

Over the two week period, the ‘Farm Ed Fiji’ Facebook page was monitored as a secondary channel for selling the application which is more scalable. Although it was not a primary focus of the currency testing experiment it is a channel that should be explored more at a future point. The KPI’s and data on the facebook page are here:


Validated learning: 

The experiment achieved a green light outcome with 71.7% of farmers interviewed purchasing the app. The experiment demonstrated that the majority of smallholder and subsistence farmers demanded a solution to the issue of the lack of information regarding pest and disease control and were prepared to pay $1 for the service. From our primary data, on a Likert scale of 1 to 5, with 1 being completely unable to deal with pests and diseases and 5 completely capable of dealing with pests and diseases, 17.9% of our purchasers scored a 1, 17.9% scored a 2, 41.1% scored a 3, 23.1% scored a 4 and none scored a 5.

Almost all farmers (98.3%) experienced issues with pests and diseases, and the majority (88.1%) were interested in a solution. In regards to missed sales, 65% of farmers that did not purchase the app were also interested in a solution, indicating that technology access was a block to purchasing the app. Farmers with no or limited access to the internet were less likely to purchase the application. A number of blocks were identified during the first three stages of the customer journey and are summarised as follows:

  • The greatest block we encountered during this stage of the customer journey was that farmers already had an effective mitigation strategy in place that was working already (10 farmers of the 17 that didn’t purchase the application).
  • The next block encountered was the lack of access to a smartphone (9 farmers of the 17 that didn’t purchase the application).
  • Some farmers weren't interested in a solution to pests and diseases for reasons they couldn't explain or were not willing to explain (6 farmers of the 17 that didn’t purchase the application).
  • The access to mobile internet was a major block in many villages we visited, however, many do have wifi access. 5 farmers of the 17 that didn’t purchase the application were due to having no access to the internet.
  • The final block encountered in purchasing the application was the lack of trust towards PEV, with farmers unwilling to pay for the service electronically through M-PAiSA (2 farmers of the 17 that didn’t purchase the application).


Next move:

Utility and solutions testing will determine the usability of the application and the desirability of the pests and diseases function. 

As smartphone use is increasing exponentially in Fiji, it is probable that in the future more farmers will want to buy the app as they will be able to access the application and have internet access to send questions and receive answers. In the future, looking into channels to better access tech-savvy early adopters would increase the adoption rate of the app significantly. 

Another block we encountered is that the time frame of this experiment is too short. If we had more time and a broader reach it is possible our Facebook conversion rate may change in future testing. Liaising through Facebook with potential customers back and forth could be time-consuming as there are lag times in responses. This could be due to potential customers not being having proficient computer literacy or from our interviews with village visits, some people did not use their smartphones often. If the experiment time frame was longer, it might provide more time for potential customers to actually purchase the app or Facebook service. We had 12 new messages with a zero conversion rate. 

Moreover, towards the end of the experiment time frame, there were still customers who expressed interest in purchasing the app. However, we did not have time to revisit villages whose meetings got canceled and pitch the app, thus, we might have lost some potential sales as well.



edited on 23rd July 2019, 22:07 by Rhys Taylor

Andrew Vild 6 months ago

This is fantastic, well done!

Reply 1

Eugenia Muñoz 6 months ago

Status label added: Experiment Results

Reply 0