Thought Leadership July 10th 2019

Why and how did McDonald's change how it attributed its sales?

Data science and big data have enabled McDonald's France to move away from a product-centric strategy to a more customer-centric strategy. Here’s how on the occasion of the Big Data Show on 11 March, 2019.

With a five-billion-Euro turnover and 1,450 points of sale in France, McDonald's is still the fast food leader in France. To maintain this position, the company decided to change how it measured its sales to identify new development and growth opportunities. “With our slogan - Come as you are - we have always tried to talk to our customers directly,” explains Romain Girard, Head of Business Insights at McDonald's France. “However, in the past we looked at sales by point of sale, product, time, and type of protein but never in an in-depth way i.e. by type of customer.”

Realising that the same customer could order the same burger in very different contexts - alone, with the family or in a group - the company decided to change how it analysed its sales. The complexity? “The change in approach required using data belonging to the general management team, and the marketing, market research, IT, digital and insights teams - not forgetting the HQ in the USA”, added Romain Girard. Highly dispersed data that needed to be identified.

For McDonald's France, the aim of using this data was to maintain its lead in a context of market disruption with the arrival (or return in some cases) of e.g. O'Tacos, Five Guys, Burger King, KFC and Marie Blachère but also Uber Eats and other retailers jumping on the fast food wagon (e.g. Carrefour and Franprix in Paris). And not forgetting the changes brought about by home delivery services for small outlets, changing eating habits and digitalisation driven by the likes of La Fourchette.

Analysing transactional data

McDonald's contacted Ekimetrics, a data science consultancy company. Its first mission was to conduct an audit to map the company’s available data to better understand their customers. “Two million French people visit McDonald's restaurants every day,” explains the Head of Business Insights at McDonald's France. “However, the transaction data owned by the brand was underutilised as we were not trying to connect it to business opportunities. Cash register receipts represent an incredible source of information but also millions of pieces of data requiring organisation.” Transactional data i.e. representing two terabytes of data, was enhanced with data from loyalty cards, home deliveries and points of sale. Then types of customers were formed based on key criteria (defined in workshops with the teams) e.g. consumption behaviour (alone, as a family, in a group, etc.) and the time of consumption (morning, afternoon or evening in particular).


Deployed in just three months with “convincing” results according to the brand, the selected cloud architecture made it possible to create simple dashboards for the business teams to explore the data. “We adopted a ‘test and learn’ approach to validate the potential using an iterative approach in a start-up format,” added Romain Girard. In this way, McDonald's France was able to calculate the volume of transactions for single customers or the products most purchased by families for example. These customer segmentation insights will be integrated into its next OOH marketing campaigns in mid-March 2019, and on TV in the following months.



Article published by, also available following this link. For more information:

Thought Leadership July 10th 2019

McDonald's France : business strategy, behavioural analysis & big data

Whether I’m in class or talking to friends, it is often difficult to explain to them what the meta-data in their daily lives can do. It is easy to imagine how big data is used in scientific calculations of meteorological models or for managing transport, but it is harder to see how it can be used in our daily lives.
Thought Leadership June 19th 2019

How to combine business approaches, advanced statistics, and technology?

It is sometimes difficult to create a link between highly sophisticated statistical approaches and business reality. The transformation of data into value is an art that needs to combine three different pillars: business understanding, advanced statistics and technology. This triangular approach needs to be adopted by all machine learning projects.