Experiment adopted

[JAN 19] [TIMOR] Price Testing [2/2]: Briquettes (Solution)

This range of Price Testing experiments are intrinsically connected with the Cost Analysis experiment (linked below). They should be read and understood as two parts of the same overarching feasibility experiment for each solution.

Lean Phase: Solution

Assumption: That our customers are willing to pay a price for briquettes that is above the failure points set out in the cost analysis experiment.

Time Period:
2 Weeks in country, repeated with new prototypes as necessary.

Success Metric:
There are two key metrics being measured.
 1. What each customer’s perceived value is for briquettes (what do they think is a fair price).
 2. Asking if customers think that $X is a fair price, and if they would pay that much for this solution. This should be a yes/no clear cut answer.

See the experiment build for more detail on this process and the price ranges.

A minimum of 30 data points should be collected before conclusions can be drawn from the data. Ideally a box and whisker plot of prices given by customers can be made, and then a graph of prices vs percentage of positive responses for each solution can be made too.

Green Light - Proceed (What is the next move given successful validation?)
Move forward with medium scale manufacturing of briquettes and begin usability and engagement testing. Continue improving and iterating on the recipes/ingredients and manufacturing processes.

Success point: There is sufficiently high conversion at a price that is above 40% gross margin. See Cost Analysis for more details on success metrics.
30% of potential customers identify a price point within ±$5 of each other.

Orange Light Optimise - Understanding why such a variation in price occurs. Do we need to target more specific customer segments with different value propositions and different levels of affluence. What is holding our customers back from paying more for this solution. Do we need to explain the solution better, give demonstrations

Failure Point - Tied in with cost analysis to determine feasibility of briquettes in Timor. Failure should be declared once alternative pitches, presentation and designs/types of briquettes have been investigated. When we can clearly state why people aren’t willing to pay our minimum viable price for briquettes based on margins then briquettes are considered failed.

Red Light - Assess other solutions for feasibility. See cost analysis experiment for more details.


Experiment build:

Standard empathy approach to start talking to people about fuel/cooking and any issues surrounded with that. This is like a basic offer test and ideally there are positive responses to these questions. Flowing from the offer test intro, presenting one or multiple of the solutions to the potential customer. Explaining what they do, how they work etc. Once the customer understands the product well and has had a chance to ask any questions they have, ask “how much would you pay for this product?”

To avoid the scenario of “how much does it cost?”,
if the customer does not really have much interest or understanding of briquettes, refer to the price testing calendar and test at the designated price without first asking for the customer to come up with a figure, as this will often result in low-ball prices if they do not have a good understanding of benchmarks for briquettes.

What prices to test?
We had many discussions in December around how to price test briquettes, as if you pay per kg then how much is 1kg, how long does it last etc. Conversely we don’t as of yet know how briquettes stack up compared to firewood. Therefore testing of this is paramount during briquette price testing, to determine what kind of performance briquettes on average give compared to firewood. Once this is established at least roughly, then we can confidently price test.

The best way we have thought of so far, and this can be improved upon, is to test relative to current fuel prices. This eliminates the discrepancies in household consumption, family size etc. There are obviously still issues with this, if you pay 25% more than you normally do for wood will the briquettes last the same time, or if you give enough for one week how much will that be, is that feasible etc. These are all questions that need answering with briquettes.

Once the above questions start to be answered the price testing is rather simple:
“If you normally spend $4/week on wood, would you spend $6/week on briquettes?” [50% increase]
“Do you think $3/week is a fair price for these briquettes [2kg bag for example]?”

The proposed calendar linked below is subject to change, depending on how many tests are able to be conducted on one day out etc etc. The calendar is just a template to ensure a spread of prices tested.

Understanding the performance of briquettes is critical and will make the price testing of them far simpler. If anyone reading this has novel ideas for how we can test in other ways before doing performance experiments I would love to hear them.

edited on 15th January 2019, 07:01 by Harry Telford

Thomas de Heus 3 months ago

It's great to see the fuel project looking at three unique solutions to the problem! Having gone down the briquette route with the Cambodia fuel project, I can say that they are quite difficult to accurately price test. You're right in that you need to compare it to their existing options (wood in this case), but it is difficult to get an accurate idea of how much someone would actually pay. We had 100% of people (out of 25 people) say they would buy it if it was cheaper than what they were using, but this wasn't a reliable test and it didn't guarentee they would actually purchase it over a bag of charcoal if it came down to it.

We were competing with wood charcoal in Cambodia, which meant briquettes were a familiar concept. It also meant our briquettes (which we had on hand to allow people to test) didn't offer a clear cut advantage over what was already being used for cooking. Since wood is the fuel of choice for the Timorese people, a product that burns for longer with less smoke can certainly be sold for a higher price if they believe it offers additional value.

It's mentioned at the end, but understanding how briquettes will stack up against wood is thus crucial. This information is likely avaliable online to provide a rough idea which would be sufficient for this initial testing (e.g. briquettes deliver up to 50% more heat per unit mass than wood logs ( )). They need to be pitched appropriately, accomodating for the fact that people are unlikely to have a technical understanding of why they're better than wood. The key points there are obviously burn longer, require less mass for the same cooking effect, and produce less smoke. It is also difficult because briquettes are so much more dense, so a 1kg bag compared to a 1kg bag of wood visually appears less, but actually lasts so much longer.

In summary, from my experience it is a difficult experiment to get good results from. There are so many variables in the performance of a briquette and they are inherently difficult to pitch. In Cambodia we brought ours with us (we only made around 25) which was good for generating interest and explaining the solution, which is definitely an option here. The briquettes we made (coconut husks) worked very effectively for an initial prototype, and allowing those cooking to place one in their stove while we had a chat to them and then letting them observe it after burning for some time was a good way to allow the person to make a more informed decision about how much they would be willing to pay for such a solution. In Cambodia we did attempt some performance measurement against charcoal for coconut briquettes, but were unable to get reliable results due to limited time and the briquettes requiring several days to dry properly. I think for an initial price test, having a rough idea of performance from secondary sources and having briquette prototypes on hand to test and observe is the most effective way to conduct this experimet. After the first day of testing it will be important to regroup and discuss what worked and what didn't, because collecting unreliable answers from 30 people is a lot of time wasted.

Keen to see how this turns out, because it's definitely an interesting area for me!

Reply 2

David Gailey 3 months ago

Hi Thomas,

Thanks for your comments and reflections from your experiences in Cambodia. We will definitely keep them in mind when we do our own price testing.

David Gailey

Reply 0

Harry Telford 3 months ago

Status label added: Experiment adopted

Reply 0