Grupo Bimbo is the world’s largest baking company headquartered in Mexico City, Mexico. They operate more than 100 brands, among which are some of your favorites: Bimbo, Wonder, and Sara Lee. This year, we partnered with them to develop a decision-making tool for their highest level executives (and to see if we could get some free pastries out of the deal). The application shows their executives key performance indicator data, in real time for each of the markets in which they operate - helping them make better, more informed decisions faster that will help them manage and grow their business. The application was built to leverage MongoDB’s big-data capabilities, is available on IOS, Android, and desktop, and even offers offline availability.
Grupo Bimbo is the world’s largest baking company headquartered in Mexico City, Mexico. They operate more than 100 brands, among which are some of your favorites: Bimbo, Wonder, and Sara Lee.
With 65 manufacturing plants and 2.5 million sales centers located in 22 countries, across 3 continents, the Grupo Bimbo team does a lot of traveling. As executive members meet with their direct reports around the world, they need quick access to the sales information and other KPI’s for each country, channel, and brand. In an effort to enable faster, more efficient communication, Grupo reached out to us to help them create a mobile sales dashboard.
The project began with a Design Sprint, a 3-week phase to bring alignment on product vision and design. Going into this phase, Grupo Bimbo had done a lot of groundwork to determine the scope of the project and its necessary features. For example, the team already knew they needed the ability to analyze sales data by organization, region, channel, and brand, and to compare the data across a specified month or year. In addition, they needed the product to analyze returns data across similar metrics, as well as analyze non-monetary data, such as POS, clients, and routes.
From insights teased out in collaborative work sessions, we produced a prototype for the product’s look and feel. Once approved by the executive team, we dove into code! (Watch out: made-up-sounding technical names and acronyms ahead.) We used two frameworks, MeteorJS and React Native, to develop an application that works across all the top platforms (Web, iOS, and Android). Then, in collaboration with the Grupo Bimbo BI team, we connected data stored in a variety of sources to MongoDB. Using MongoDB’s massive parallel processing (MPP) capabilities in harmony with AWS Lambda’s (Serverless) we were able to tackle the big-data problem we were facing.
Our greatest hurdle was digging out all of Grupo Bimbo’s data. As Bimbo is spread across multiple parts of the world, several of their regions have their own method and data structure for handling and saving sales data. We helped them resolve this
About 70% of their sales data is stored in a centralized data lake, while the other 30% is handled manually (though Excel). While you might expect that the manually processed data would make it into the data lake before being manipulated by other tools (like the one we built), it doesn’t (at least not yet).
Identifying this snag was a battle; solving it was simple. We created a bucket to catch CSV files produced by the BI team at the end of their process. Now those files are read and saved into MongoDB for processing.
The app is available on IOS, Android, and desktop (as well as available offline), giving Grupo Bimbo’s executives quick and easy access to critical, up-to-date information regarding sales, returns, and equipment wherever they are, and on any device. The project was a success and we are actively planning new projects to work on together with Grupo Bimbo in the coming months.