Data Visualization of Mosaiq
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
Photo by Nancy Lee
A glimpse at the flourishing industry of wine and spirits in Quebec, renowned gourmet destination!
When I embarked on the journey of learning how to become a data scientist at the NYC Data Science Academy, I knew from the start that it would be a great challenge! For my first project, I needed to build an interactive Shiny application that could create visual representations of large sets of data. That is when I partnered with a company that could use such a tool, Mosaiq Inc.
Mosaiq is an independently owned company based in Montreal that imports a great selection of beers, wines and spirits from all over the world and sells them to a government owned corporation named The Sociรฉtรฉ des alcools du Quรฉbec (SAQ).
The SAQ's mandate is to sell alcoholic products to consumers. It provides a major income stream to the Quebec and Canadian governments in the form of taxes and duties. It also makes dividend payment to the Quebec government which is its sole shareholder. The SAQ has over 400 retail outlets in the province of Quebec!
Building the app
One of the problems Mosaiq had was having to work with huge raw csv sales files that were difficult to analyze. The SAQ sells these files which contain a long list of every sales transaction in Quebec in a week. Mosaiq currently uses the files to calculate total province-wide sales but they couldn't evaluate how their products were doing by city or region.
For this project we have used sales data for the month of December 2017 which is the most important month of the year for the industry!
Mosaiq wanted a tool that would allow them to see the percent market share of their products broken down by city or region. To see the sales by different SAQ banners was important as well. For instance SAQ Express outlets have the best selling products and SAQ Signature outlets have high end, exclusive and rare products. Knowing this information would help focus their sales, promotion and marketing teams on certain regions.
The application I have developped solves Mosaiq's problem! I have named the tool HallStat after the charming town Hallstatt I fell in love with when I visited Austria.
Data analysis
Some challenges with the raw csv files Mosaiq purchased were the need to:
- ignore certain transactions such as the sales from warehouses.
- merge multiple reference files as the raw sales files only had codes.
- determine factors to convert various bottle sizes to nine liter equivalent cases which
is the industry standard to compare volumes.
- group all non-Mosaiq products into one group called "Other"
I am proud to now present the app's functionality. I wanted to keep it as simple and easy to navigate as possible so Mosaiq can use it weekly, monthly or even every time they meet with their customers (which are the wine and spirits makers).

The tab "Explore the market" displays interactive graphs and text that updates every time a new criteria is selected.

The user has the option to select all the banners as well as the divisions, without having to manually select them all.