Where Data should be placed within in the organization? What a good data governance should look like ? How to secure the sponsorship from the business?
To get answers to these questions faced by many CDOs, we interviewed Isabel Gomez Garcia de Soria , Group Chief Data& Analytics Officer at L’Oréal.
Hi Isabel, can you please tell us how you became Chief Data Officer at L'Oréal?
I joined L’Oréal in 2019 with the ambition to structure data initiatives within the group. L’Oréal gave me three main missions:
1) Scale : many Data initiatives existed within the group but we needed to industrialize our approach.
2) Optimize: We wanted to improve our productivity and we needed to rely on trusted data to take strategic decisions.
3) Monetize: We need to leverage our data to generate new revenues. We launched a new offer “ Beauty Tech” based on a highly customized approach. To optimize client satisfaction & acquisition we have to be excellent in data & analytics.
As CDO, you report to the CFO, can you tell us why L’Oréal made this decision ?
First let me explain what we didn’t want. We didn’t want Data & Analytics to be seen only as a Tech subject. That’s why we didn’t want Data to be attached to the IT department, because in L’Oréal’s culture, the IT department does not drive business initiatives.
Then, we started from our need, L’Oréal needed to put its data initiatives in order, we needed structure and discipline.
That’s why we decided to attach Data to the finance department because our finance department is powerful. Finance has complete on what is going on in every subsidiary of the group. Therefore, it made sense to integrate us into the Finance Department, and it has paid off so far. However, there is no guarantee that this will be the case tomorrow, we are allowing ourselves the right to change.
How do you create a data governance in such a large group with so many subsidiaries?
You have to start with the global model you want to adopt. At L’Oréal, what made sense to us was to have a federated governance. The Data Owners are not in my team, they have to be in the business lines directly, the business lines are the ones who are most likely to use the data.
Then we articulated data governance around two main missions:
1) Data Ready for Business: making data accessible and reliable for the business
2) Business Ready for Business: educate the business units on how to use and leverage data.
How did you run the “Data Ready for Business” and “Business Ready for Data” initiatives?
When I arrived, I spent a lot of time on Data Ready for Business.
It’s a fairly sensitive subject, because we have to explain to the business that in order to start leveraging data and making revenues from it, we must first make a lot of investments, even if there is no immediate ROI. You have to invest in data quality, in a platform, in a data catalog… all of this costs time and money but it is a necessary. There is a lot of education to be done while clearly indicating that the longer we wait to do it, the more expensive it will be for the company.
For the past 12 months I have been working hard on the second initiative, Business Ready for Data. The goal is to get the business units involved in this transformation. If the business doesn’t play the data game, it’s a waste of time. You really have to integrate them into the project.
Many CDOs share the difficulty of getting the business units on board.
Many CDOs share the difficulty of getting the business units on board. How do you go about achieving this objective?
Governance played a key role. I worked directly with the CFO and the CEO of L’Oréal to find 18 Data Owners within the group able to drive our strategy. The CEO appointed these 18 people. He used different criteria to select the “chosen ones” such as their importance in their business unit, their leadership, their creativity. I then personally wrote a letter to these 18 people explaining that they were officially Data Owner and I had the letter signed by my CEO himself.
This formalism was necessary to make them understand that this new role was important, it was actually a reward. We then formed a real data club within L’Oréal. Being part of this club was seen as a privilege by the members.
How did these 18 Data Owners drive the Data & Analytics within L'Oréal?
First of all, we trained them. We explained to them the expectations, the vision, the involvements and we introduced them to Data technologies.
Then, to support them, we recruited 120 people (L’Oréal employs 86,000 people) to work with them. We appointed data stewards in priority.
What is your relationship with the IT department?
It’s a key partner, and our CIO has appointed a person within his team to be in charge of Data. He is a sort of CTO for Data and he is very involved in our strategic issues. For example, he is in charge of the Data Platform and the technological stack.
If you could go back in the past, what would you do differently?
t’s hard to say, but with a little hindsight, I think I would try to be a little firmer on certain subjects with the business. It’s good to be open to dialogue and to allow a certain amount of flexibility, but on the subject of Data, you also have to be firm on certain positions. You have to define a framework and make sure that the different players don’t get out of it.
What do you think of Hubadviser?
I think your concept is very good because it allows you to exchange with peers and experts who have operational feedback. This is very valuable for me because the Data & Analytics discipline is so new, that we need to rely on concrete and proven feedback to justify our decisions.