FrenchTechFriday: The Dear Muesli Algorithm

Jun 22, 2018
Vote on Hacker News

 
French startup Dear Muesli was not, upon launching, a French Tech startup. But a little foodtech never hurt anybody as Dear Muesli is finding out in its latest product: DM+.
 

The Abercrombie & Fitch testing ground

 
As often, the idea for a new startup springs from a noticeable lack. Dikom Bakang and his brother Bakang Bankantonje were both working at A&F as students. They would bring their lunch which often included muesli. However, the industrial brands had two main drawbacks: too much sugar, too little choice.
The brothers didn’t care though: they would get their muesli ingredients and make their own. This was noticed by their A&F coworkers who began to ask them for this à la carte muesli and pay for it.
Dear Muesli was born.
 

Dear Muesli

 
Launched in March 2015, Dear Muesli is an online store offering all different sorts of muesli to accommodate the pickiest. Basing its communication on social networks (mostly Instagram), Dear Muesli doesn’t have a brick-and-mortar store. Their products are featured at Colette and they own a mobile Muesli Bar funded with Kickstarter.
 

Getting some tech in the process

 
With Dear Muesli, consumers can choose from a wide variety of grains, nuts, seeds but they don’t always pick the ones that go well together. Not all consumers are alike. dietary needs vary according to age, sex, pathologies even location. Yet, how can you tell which mix is most suited? And how do you avoid having a disappointed client that mixed the wrong ingredients.
 
The Dear Muesli team is getting Data Scientists in the game and will provide, starting next September, a brand new experience for consumers.
 

The DM+

 
Rather than resort to trial and error to find the best combination of muesli, the Dear Muesli user will fill in a questionnaire. Pretty much like a trip to the dietician, the DM+ will use an extensive algorithm to predict the best Muesli for the consumer. Location, age, current health status (a pregnancy changes the dietary needs), activity, goals and phenotype, all are taken into consideration. Last, but not least, after the user has picked all the ingredients, the algorithm will also make one vital final check: taste. A huge data bank is being set up to verify the given combination is tasty.
 

25% of the team is for data management

 
The startup launched with only 3 people: the founders (the two brothers and a friend: the Muesli Boys). A fundraising just took place, drawing in over 500,000€. This allowed the team to expand all the way to 12 people including three data scientists. All efforts are on conjuring the most complete algorithm for DM+.
 

“…Somebody else will”

 
“Go for it otherwise another less talented person will” is Dikom’s motto. Dear Muesli is looking to develop as B2B in Europe although it is already sending its muesli overseas to hungry web users. The launch of DM+ should give Dear Muesli a greater reach especially among sport and healthy lifestyle enthusiasts. Another good example of a happy marriage between data science and French food.