How to develop a data driven culture?

What is a data-driven company?

A data-driven company is a data-driven structure. This means that it uses data as a lever for making informed decisions to drive its business, with a view to improving performance and creating value.

Aware of this, companies must start a data transformation process in order to become data-driven structures to continue to exist in an increasingly complex environment. A study conducted by Deloitte in 2019, shows that of the panel of companies wishing to become data-driven, only 32% succeed.

This transformation of companies is not simple. Before being a technological issue, it implies first and foremost a challenge in terms of culture, processes and above all people. To create a data-driven culture, companies must take action.

Have the COMEX address the data issue

“The first thing was really this managerial will to place data at the center of Microsoft’s transformation” Lionel Gourvitch, Director of Data&IA Strategy – Microsoft.

The issue of data is above all a strategic and transversal lever of action carried by the COMEX. Without a strong commitment from top management tointegrate data culture into its strategic plan, this transformation will never take place.

What’s more, this approach ensures that all corporate bodies, right down to the lowest level, pay particular attention to data.

With a clear vision and measurable objectives at all levels, management is the driving force behind the project and encourages all employees to get involved in the process, enabling the company to succeed in this transformation.

The COMEX sponsorship of the data transformation also ensures a budgetary envelope, enabling investment in the infrastructures necessary for this transformation. Indeed, becoming a data-driven company implies heavy investments to collect, clean, model, analyze, visualize, secure and store a large amount of data.

Implementing data governance

“Data Governance Is About People and Process, Not Technology” Terence Siganakis, CEO of Growing Data

This mass of data produced by companies requires ad hoc data governance. Indeed, the challenge lies in being able to move from an anarchic, visionless data management mode to a reliable organization that guarantees access to quality and compliant data. Data governance must impact the organization of the company in order to make it able to manage its data efficiently. To achieve this, companies need to reconsider the way they process data.

Data governance consists of putting in place a data management framework based on a set of processes, rules, standards and roles that guarantee the use of company-generated data to develop business and innovation.

Good data governance is built around 2 pillars:

• Define a frame of reference for data use and processing. Data governance must make it possible to define rules and standards to guarantee the quality, compliance (GDPR), security and availability of data (decompartmentalizing data).
• Define the players, roles and bodies involved. To ensure smooth collaboration within the company and create a sustainable ecosystem, it is essential to identify the players involved in the data governance process. Roles, responsibilities and decision-making bodies are used to decide on strategic directions.

Demonstrate the usefulness of data with simple, high value-added use cases.

“Start small and demonstrate measurable business outcomes” Randy Bean, author of “Fail Fast, Learn Faster: Lessons in Data-Driven Leadership in an Age of Disruption, Big Data, and AI”

In the imagination of many employees, data-related issues are primarily the concern of technical teams. It’s a common misconception in many companies, and one that holds back organizations engaged in data transformation. An EPSILON study on “data maturity” underlines that 81% of the employees surveyed consider that data is not just a trendy topic.

To make a success of this transformation, it’s imperative forcompanies toinvolve their business in data projects. Because data only comes to answer a problem, or create value for the customer. Without business, data is worthless. A successful data project must be based on a “Pain Point” clearly identified by the business.

However, the choice of the use case that inaugurates the data work in the company is strategic. It must be clearly identified, understandable to all and have a strong impact. It emerges through collaboration between data experts and business representatives.

By starting with a critical use case, decision-makers can quickly demonstrate the value of data and establish the legitimacy of investments through “Quick Wins. The objective is to create a positive dynamic around a significant success story for the whole company, highlighting clear KPIs, and to communicate on the return on investment generated by the project. This usage-based approach helps convert even the most skeptical.

Building on this high-impact success, the organization will iteratively engage in a continuous learning cycle that will lead to process refinements, larger projects, and a data culture. The more data projects there are, the more the business will be involved and understand the issues, and the more the data culture will spread throughout the company. Around this use case, the challenge is to show that the success of a data project is the fruit of cooperation between data experts and non-experts.

In this learning process, one point of attention must be followed, that of the stigmatization of failure. Indeed, the data mining phase does not always lead to conclusive results for the business. The managerial line must be made aware of the management of failure in order not to slow down the dynamics of the teams.

Data transformation, a matter for specialists?

“Data transformation is essentially a matter for non-specialists ” Julien Levy – Affiliate Professor at HEC Paris

With the growth of data-related activities, several new professions have emerged. If you want to succeed in your transformation, you need to surround yourself with data specialists.

• Data manager: oversees the collection and organization of company data, with a view to optimizing its use.
• Data engineer: ensures that data is available within the company
• Data analyst: exploits and interprets data to extract observations and trends useful to the business.
• Data scientist: develops algorithms to make the masses of data available speak to business needs.
• Data protection officer: ensures compliance with data protection legislation (GDPR)

But beyond the recruitment of data profiles, it is essential to set up an annual training program for all employees on data issues. This training plan must be long term and evolving to become and remain a “data driven” company. The speed at which the world of data is evolving means that companies must continue to invest in training to give their teams the ability to adapt to these changes.

The goal here is to ensure that all employees follow a common training path. This enables them to immerse themselves in the fundamentals of data culture and understand their role in the data value chain. Each employee must be able to perceive his or her own interest in the data production/processing process.

The effort to acculturate non-specialists to data literacy is as important as the effort to recruit data profiles. Ensuring that each employee knows how to read, manipulate and use data is a fundamental issue. Many experts agree that data literacy will be the most in-demand skill by 2030.

Meaning at the heart of the transformation

“The main engine of our action is meaning” – Michel Lejoyeux, Management Expert – University of Paris-Diderot.

As explained above, a data transformation project requires the involvement of the men and women in the organization.

Aware of the aversion to change in human beings, it is essential to provide a change management system adapted to this type of project. To do so, this work must be done at the beginning of the project. The earlier the change process is accompanied, the greater the chances of success.

To encourage change, it is essential to engage employees in understanding why to change. Help them visualize a clear picture of what their company will look like tomorrow, thanks to this transformation.

The challenge is to bring meaning to the project and to define a vision, a clear ambition in order to gain the support of the greatest number. This work will allow for deep reflection on change. They imply the implementation of an organized and targeted communication to try to dissipate the brakes generated by the change and limit the risk of failure.

Data transformation is a long-term project that requires patience and commitment. The key success factor in transformation projects is change management. Without buy-in and acceptance of the new organization, the transformation will not take place.

In the years to come, the stakes around data skills will be higher and higher. The tension on these profiles is increasing on the labor market. Companies will need to be innovative in attracting and retaining these skills to ensure constant growth in the organization’s data maturity.

At SQORUS, we support our customers in a wide range of data transformation issues: strategic Council, data consulting, change management, data analysis and processing, analytics, dashboards…