Project Everest

Adopted Experiment

[Proposed Experiment]: Channels- Sales Agents- Fuel Fiji December 2019


The following experiment is based on the proposed sales agent pipeline experiment from July 2019 and tests the use of sales agents as a viable distribution channel for selling the Buka Stove 4.0 with the changes and developments listed here. The results for the July experiment can be found here:


Qualified leads: For the purpose of this experiment, qualified leads will be defined as the number of deposits collected to secure the purchase of the Buka Stove purchase via sales agents

Lean Phase: Channels

Timebox: 3 weeks (December) 

Assumption: An autonomous sales system that involves sales agents collecting and submitting quality leads to an in-country distributor will result in sales of the Buka 4.0. 

Success metric: Number of qualified leads obtained via trained sales agents 


Green light: This distribution channel can be scaled 

Identify the sales agents which produced the greatest number of sales. Contact them and find out:

  • the villages/areas they have visited;
  • an estimate of the total number of people they have approached within the timebox;
  • the qualifying questions they asked;
  • the types of people they approached; and
  • any other reasons they believe their work has resulted in many sales

Use this information to retrain existing sales agents as well as new sales agents

Success point: 10% conversion of total EOIs to qualified leads

Orange light: 

Optimise the system in place by ensuring that sales agents send EOIs to PEV through Rose Gooding (phone number provided on Sales Agent Handout) as frequently as possible, and that sales agents set up meetings between Natasha and the customer as soon as possible.

Contact EOIs that did not convert to qualified leads and/or sales and enquire about what prevented them from buying the product. 

Failure point: less than 5% of total EOIs results in qualified leads

Red light: 

  • Alter sales agent training to be more comprehensive of the sales pipeline
  • Assess whether leads generated are within a specific, validated customer segment. If they are not, clarify the target market with sales agents. 
  • Consider optimising remuneration to sales agents to encourage more intensive efforts 
  • Have a stricter procedure for qualifying future sales agents based on desirable qualities, such as being intrinsically-motivated 

Experiment Build

The following experiment is in response to the results obtained from the autonomous sales agent’s pipeline experiment conducted in July 2019 where local sales agents involved with PEV were found to be a feasible distribution system. The details on this can be found at: 

The experiment will be carried out during the month of December 2019 and will fundamentally test the autonomous sales system by re-contacting the sales agents employed during July 2019 since the July experiment failed to test the effectiveness of sales agents as a distribution channel due to time constraints. The results on the sale agents experiment conducted in July can be found at:

To strengthen and incentivise the sales agents further, the team created a change in the financial plan with the sales agents receiving a $10 commission with a $5 immediate deposit on confirmation of purchase, defined as the ‘qualified lead’. On successful purchase and collection of the Buka Stove following the meet-up organized by the sales agent, sales agents receive the remaining $5 commission along with Natasha receiving her $5 commission. 


A sales training will be conducted on 191203 to solidify the distribution channel and the overall aim of restoring sales of the Buka 4.0 using in-country distributors. This will involve preparing existing and new sales agents for collecting more EOIs. Sales agents and/or village representatives from July 2019 were re-contacted to investigate about their progress on gaining EOI’s and their lack of motivation in obtaining EOI’s. Sales agents that positively responded to our calls were encouraged to attend our improved Sales Training. The Sales Training will discuss a new payment breakdown ($5 natasha $10 sales agent) and more interactive sessions such as scenario based sessions and FAQ’s to engage our potential sales agents. This new incentive must be validated by cost and revenue analysis (on 191202).

Promoting Sales Training 

In addition to the new sales training, changes have been made to the poster to further promote the training with an attempt to get approval for poster placement in market/on council property. A radio ad was also introduced to target educated, well connected young people (18-25) in Sigatoka to encourage attendance to the sales training. Newspaper ads were also explored but were rejected due to budget constraints. 

Collection of EOIs

After 1.5 weeks and after 3 weeks, sales agents will upload a copy of their EOI sheet to PEV through best means of communication to be discussed during the scheduled Sales Agent Training Workshop and also by means initially created by the July experiment (Project Buka Facebook page, mobile phone number). 

Closing the Sale

Sales agents will organise a time suitable for the customer to collect the stove from a market stall in Sigatoka Markets run by Natasha Singh (our cook). These details will be passed to Natasha by the sales agent, through appropriate means of communication agreed by both Natasha and the sales agents. Natasha will contact the customer to confirm the Buka stove pickup and payment. Details of the role of Natasha in the sales pipeline are outlined in the following MOU:

The customers directed to Natasha will be cross referenced with the deposit payment receipt given to PEV via the sales agents.


Sales Agents will be paid $10 FJD per close that can be attributed to their leads and sales. Natasha will paid $5 FJD per close for her role in the sales pipeline. 

Data Comparison

Call leads and sales agents and find out why certain sales were a success. Use data points mentioned in the green light section above to determine what to focus on when recruiting and training new sales agents.



Tagged users
edited on 18th December 2019, 03:12 by Charlotte Jones

Mira Gaponova 2 months ago

Hi Fuel DEC team,
It is good to see you are running another sales experiment! I had a few questions about what type of people were being used as 'sales agents' and if the quality of the agent impacts the sales - such as using uni marketing students from FNU vs women in villages. Additionally, if collecting EOIs was the best option vs going into a village and selling the stoves straight away. These are some points to consider. Have an amazing month!

Reply 1

Vibhu Soni 2 months ago

Hi Mira!

Thank you for your question! The sales agents we are working with this month are all from local villages. We discussed this as a team and found that the following are ideal qualities in the sales agents:

- Intrinsic Motivation
- Ties to the Community
- Commitment / Committed
- Empathy / Empathetic
- Resiliency / Resilient
- Communication Skills
- Positive Thinking
- Charisma / Charismatic
- Patience
- Accountability

Currently, we have not had the opportunity to reach out to any FNU students as majority of them are not located in Sigatoka. The rationale is that we have gone through the stage where stoves are directly sold by PEV teams (some 96 Buka stoves have been sold in the past two years!). We have found that the product is effective; and we want to scale the product to make a bigger impact in the local communities! Thus, the main idea is to automate the whole business in the local area and reduce the dependence on PEV teams. This will push the business to become more self-sustaining and increase the impacts at the same time. However, this might be a suitable goal for future teams.

For the EOIs, the idea is that some customers will need some time before they can decide that they will buy a stove. The sales agents may not be able to sell the stoves straight away, but with the EOIs, they are able to keep in touch with the customers and push for more sales. In addition, the EOIs is part of the KPI system for sales agents. Of course, if there are customers who would like to buy the stove straight away, after collecting EOIs, the sales agents sell the stove straight away.

Hope this gives you some information you need. If you have further questions, we will be happy to answer them as well!

Stay tuned!

Shupeng & Vibhu

Reply 1

Andrew Nguyen 2 months ago

Hi Fuel DEC team,

It is great to see that sales agents are scalable as a channel for distribution. A concern I have is with the current financial plan and its viability. If effective, the financial plan will be further implemented. Albeit if appropriate it may be reiterated to provide sales agents with stronger incentive. I hope the experiments will yield successful results. Have a great month.

Regards, Andrew

Reply 1

Hao Guo 2 months ago

Hi Andrew!

We have conducted a thorough financial analysis of the buka stoves and the costs involved, including commissions. We are currently testing the sales agent channel and have changed the commission of $10 to the sales agent for a stronger incentive, however this is still part of the experiment. Stay tuned for the results and the conversion rates of our sales agents!

Reply 3

Sarah Song 2 months ago

Hi Fuel DEC Team

I really like the fact that the team is trying to identify/contact the sales agent which produced the largest sales and using that data to retrain existing/new sales agents. However, I feel like it will be beneficial for the team if they also took into considerations of the sales agents who have not made as much sales. As then we will be able to see what didn't work and make sure we retrain the exisiting/new sales agents about what not to do. I hope that the experiments produce some promising results. Have an AWESOME month.


Reply 1

Vibhu Soni 2 months ago

Hi Sarah,

Thank you for your question! What actually happened was that we had contacted all the past sales agents from July. But since the on-going support system and rapport-building between PEV and the sales agents were almost entirely missing from July to December 2019, most of the sales agents said they were no longer interested. And to clarify, we did recruit every past sales agent we could. From the validated learning section of the July experiment results, we feel that time is a crucial factor in what went wrong: the July team recruited the sales agents but did not have the time to properly train or support them, which is why our primary focus is to fully flesh out the whole process of collaborating with sales agents.

Hope this gives you some information you need. If you have further questions, we will be happy to answer them as well!

Reply 1

Nikhita Sahay 2 months ago

Hi Fuel DEC team,

This looks like very well thought out experiment!
I'm wondering what how you have incentivised the radio ad for sales training, and promoted prospects and skills gained for beyond this project.
I'm also curious about what has informed your failure and success points. Are they artibrary, from sales theory or has it previously been a good indication?
Also, in assessing the number of closes of the sales people, is there are further analysis of thier indiviual closing rate and the size of thier pipeline? Are those metrics being assessed? Is improving that part of thier intitial or ongoing training?
Very impresive work! I hope you all obtain fruitful data!

Warm Regards,

Reply 1

Vibhu Soni 2 months ago

Hi Nikhita,

Thank you for your question! I hope I didn’t misunderstand your question, but if I do just reply to the post:)

For the radio ad, one incentive for the public to participate was definitely the fact that it was free. And yes, we did mention to the participants that these sales skills can be applied to context beyond this project.

Regarding the success and failure points, the exact number is arbitrary. But I believe we do have good reasons.

The first reason for this is that we didn’t really have a point of reference: there were no sales were made nor any EOIs collected by sales agents previously.

The second reason is we do understand Fiji is a very friendly nation. People here go the extra mile to be nice to each other. A lot of people would say yes they are interested, but would not actually buy it. In this context, we figured it would be more practical if we set the conversion rate a bit lower than what we would expect in other contexts.

The third one is that we also need to take those sales agents who might not be doing well into account to see the entire picture. The idea is that if we have 10 sales agents and they collected 100 EOIs in total in the next month, only 10 of them need to be converted into sales to show that the sales agents are a valid channel considering we have only sold around 100 stoves in the past two years.

As for the metrics, yes we are also developing the metrics as part of the on-going support system for sales agents, which will allow us to better assess the sales agents’ performance. I’m impressed you thought of this step! Stay tuned for more experiment post!

Hope this gives you some information you need. If you have further questions, we will be happy to answer them as well!

Best regards,

Reply 1

Nikhita Sahay 2 months ago

Thanks Shupeng! That's exactly what I was after. Very good reasoning!
Sales training is obviously beneficial coming from a Western background, I'm worried it doesn't translate through the cultural barrier as well as we think it would. What is the group consensus over there about that?
Regarding the second point on failure, is there some ideas about qualifying buyers? I know it has been mentioned but I haven't seen a clear plan. No-risk deposits is the most traditional approach however I'm sure a more creative solution is out there!
Keep it cracking ;)

Reply 0

View all replies (2)

Harrison Mccowage 2 months ago

Hi Fuel DEC team,
great work guys I am curious about the effectiveness of the sales agents? I can see long term it is a great solution to help achieve autonomy but I wonder how high the sale conversion rate is, including a comparison with PEV staff, and if this could reflect the knowledge they have of the product, and commitment which I would expect PEV staff do better in this field, as well as the advantages they get by being locals and therefore seen as more trustworthy.
Again nice work team keep it going,

Reply 1

Georgia Kappos 2 months ago

Hi Harry,

Thanks for your message! Conversion rate is the exact measurement that we are testing to validate sales agents as a viable sales channel. We have just trained out sales agents and our results are currently inconclusive, so stay tuned for when we post them next week and I hope your questions are answered then. As for long term, we wont be able to test the difference between sales agents and PEV staff until the sales agents are fully trained and selling autonomously.


Reply 2

Tom Woodhouse 2 months ago

Hi fuel DEC team, I am excited to see the outcome of this experiment! Just a few quick questions regarding the manufacturer of the stoves when considering automated sales. With the increase of sales which are hoping to be achieved, as well as being automated therefore year-round, will the current supplier be able to meet such demand? Is there a backlog of inventory which could be utilised if needed? Will there be enough material readily available/will alternate materials have to be sourced? How will these factors alter the manufacture price/will offset this get pushed onto the customer. Keep up the great work. Kind regards, Tom.

Reply 1

Vibhu Soni 2 months ago

Hi Tom,

Thank you for your question! Yes I believe they will, although this does require further assessment. The reason I believe they will is that part of the reason the manufacturing price is very high at the moment is that the number of stoves contracted is small. The stove is made out of recycled gas tank; personally I’m not too worried the manufacturer would run out of material. However, the implication of altering the manufacturing process on both us and the customers is hard to estimate at the moment. Changing the price of the stove will probably require a new financial analysis, which is not in our current frame of work. But yes it is definitely an area worth exploring after completing the current experiment with sales agents.

Best regards,

Reply 1

Hao-Nam 2 months ago

Hi Fuel Dec Team,

It is great to see the team giving comission for each closing sale to the sales agents as well as Natasha. Just a question; considering that Natasha is employed to fulfill other roles at PEV, why is Natasha paid less than the sales agents. Since for each closing sale of the Buka Stove, the sales agent is getting paid $10 FJD whilst Natasha is getting paid only $5 FJD.

Keep up the great work and I am looking forward to the results for this month's experiment.



Reply 0

Tom Woodhouse 2 months ago

Hi Hao-Nam, I may be wrong, but i’d assume it is because Natasha will receive $5 from every sale, whereas the sales agent only receives $10 for sales they personally orchestrate, so collectively Natasha will earn more. Regards, Tom.

Reply 1

Vibhu Soni 2 months ago

Hi Hao-Nam,

Thank you for your question!

Tom's answer is correct; however the main reason to award Natasha a $5 commission instead of the $10 is that our focus for the December Project is to establish and stabilise a base of Sale Agents that can develop their skills over the coming months and continue stove distribution for periods when we are out of country. Therefore, this change is temporary for the moment, to incentivise the sales agents to join the team.

In order to test the long term feasibility for this change, the future teams (January and February) are encouraged to conduct a financial analysis that would either support or deny this change.

I hope this helps, I look forward to hearing from you!

All the best,

Reply 0

View all replies (2)

Vitoria Dahara Campos Deveza e Silva 2 months ago

Hi Fuel Dec Team!

First of all, good work! It's clear you guys have come up with insightful, valuable observations about the project and its effectiveness in terms of sales which no doubt will be of great help when the Jan team takes over. So thank you for your hard work! (:

A couple of questions for you - so as to further incentivise people to attend the training session, what are your thoughts on offering some kind of certificate so they have something tangible to show for it when they eventually apply for other jobs? Would that be viable?

Also, in terms of promoting the training, have you guys considered using facebook ads as well? Or is there an indication of radios having a higher degree of marketing penetration?

Looking forward to hearing from you!

Warm regards,


Reply 0

Vibhu Soni 2 months ago

Hi Vitoria,

Thank you for getting in touch with us!

You raise a great point, we have discussed this with our group leader and providing a certification document/ recommendation letter is a concept that we look to implement in the near future. We would encourage you to bring this point up with your group leader so you will be able to communicate this directly to current/ new sales agents.

Facebook plays an integral role in the overall communication between the Fuel team and the sales agents; however we have found that part of the population does not have access to a smartphone. For this reason, facebook advertising would not receive as much attention, compared to the radio advertisement along with any posters/ flyers.

Feel free to reach out to us if you have any further questions or concerns!

Best Regards,

Reply 1

Edward Antony 2 months ago

Hi Fuel DEC team,

I am excited to see the outcome of this experiment! Since one of the main purpose of the stoves is to indirectly reduce in-house pollutions coming from other cooking methods, I was just wondering whether its a good idea to raise the awareness and the experience of the sales agents (in respect to health and the benefits of the stove in tackling health problems) first in order for them to grasp a first-hand experience. Maybe this can be given in the form of a cooking demonstration or a communal kitchen everyone can use (I think this was mentioned by someone, but I can’t seem to find the comment anymore).

In terms of the incentives, maybe introducing a milestone system for the sales agents can also motivate them to sell and reach out to more people. For example, for every 100 sales they managed to get, they will be given a plane ticket to go on a holiday (a bit too extra, but hope you get the point..).


Reply 0

Rose Gooding 2 months ago

Status label added: Adopted Experiment

Reply 0