Isn't data a subject for geeks? It all depends who is using it and how…
When we mention DATA, we quickly think of either rooms full of servers, a slightly (too) airy-fairy vision of the cloud, or what's left on our mobile phone monthly bundle. But you will have guessed the concept of data is much broader than that, and it can even be helpful for working on your environmental impact …
Data is the digital representation of a concept. Any type of concept. A logo is data. The date Quechua was established is data.A price is a data. In essence, data is a piece of information
Otherwise, Wikipedia tells us it is also character in Star Trek, a saga we won't be going into here.
At DECATHLON, there are three types of data: data linked to finance, data associated with logistics (the supply chain), and sustainable development data.
Each piece of data has to be mastered, which means somebody has to thoroughly understand what each piece of data encapsulates.What is the context, what are the limits? How to calculate it? How to report it? For certain types of data, it is a piece of cake, while for others, it is trickier (cotton is a good example: it is a raw material among 1,600 listed at Decathlon).
Once the context is well defined, the concept is clear, then remains a vital question: is it correctly deployed (and used) by our teams?
This is where data governance comes in: in other words, providing a set of rules, a context for each person in charge of a piece of data (the data owner) to adhere to. The goal: ensure each piece of data is of good quality.
Once the context is defined, the data collected (through “sources tools”, where they are directly input by the engineers, for example), it is then about getting the data to "talk" to each other. And that happens via a data lake (a unique and centralised data repository). It is then about ensuring the data are consistent with each other within the company, but also “enriching the data”: cross-referencing one with the other, ideally getting insights (this is the data analysis part).
We can, for example, subsequently cross-reference the types of products sold and eco-design %. Analysis of results can help to identify if a scope is lagging behind on this issue, and why this might be the case.
And then we are finished, right?
Well, nearly. All that remains is to give it a visual: that is the role of business intelligence. Data science will then be in charge of observing past performance to prepare for the future.
Sustainable development is a data-intensive area. To report information about the environment, you need concrete data: What materials? What manufacturing processes? What factory? What is the manufacturing percentage of this or that factory? The sustainable development domain is major user of company and cross-functional data: logistics, design, supply chain, you have to know everything!
And in the end, the environmental assessment will be a means of converting all these "real" data into impacts. It is essential to get this conversion right.
Because without data or reliable data, how can decisions be made? How do you take decisions or actions without being in control of product design? How do you know what has the biggest impact on climate change throughout a product's life cycle?
Especially since there are other issues at stake, given environmental impacts are very systemic. And it is impossible to test “in situ” to see what is happening. You need actual data.
Given Decathlon has been designing its products for… quite a long time (roughly 40 years), and is subsequently best placed to carry this work on a very wide range of products: clothing, footwear, bikes, “heavier” equipment such as tents or treadmill and even nutrition. Composition, quantity of material, production processes… we know our products even better than if we had made them.
2012: It's the start of data being used for environmental assessment purposes
1,600* different raw materials, for which we have environmental data.
400* pieces of manufacturing process data
1,300* sites that we request monitor their energy consumption
200* pieces of data to convert energy consumption into environmental impact
100* pieces of data to calculate transport
63,4% of DECATHLON products display an environmental score.
*approximately
The other (major) benefit, for the company, in using data is the ability to rely on it when carrying out its carbon footprint assessment. Many companies base themselves on financial data, which they convert into CO2. A method that averages everything out and does not let you identify potential levers to reduce emissions. .
DECATHLON has decided to calculate its carbon footprint based on actual data (raw materials, manufacturing countries…), life cycle analysis (carried out for all our products) and rely on the PEF method. It subsequently lets us define strategies, identify the means that could help to speed up regarding the issue, calculate the best past energy / “carbon benefit” ratio obtained… And, therefore, take significant decisions regarding the company's sustainable development.
Environmental labelling, CSRD… just some examples where the use of these data are required. It is, therefore, just as much of strategic importance for the company as a legal requirement. What is more generally now at stake: is the harmonisation of rules of calculation for all companies to ensure a reliable result, making it easy for each citizen to compare and understand.
