Today, the most successful organizations use data to gain a competitive advantage, supporting the same critical business decision-making processes. In this way, all analytics applications developed to provide essential information have a set of four components, all of them equally important: data, process, organizational model, technology, people, and culture.
Any informed decision-making process begins with data, which is meaningless in itself. Determining the context is critical for proper analysis and use by organizations, as well as for achieving a broader knowledge of data.
The process and organizational model
Data literacy can be defined as the ability to read, write, and communicate data in its context, including an understanding of sources, analytical methods, applied techniques, and above all, the ability to describe a use case, application, and resultant value. Organizations will need to think deeply about their degree of analytical maturity and the data strategy that must be implemented to gain a competitive advantage in their sector of activity. However, the equation enters four pillars: technology, governance model, organizational structure, and business value.
With regard to technology and the governance model, it is necessary to develop skills in data management, information governance and analytics, as well as define a governance structure with proper definition of responsibilities and decision-making processes. One of the most interesting and least unanimous points among companies is the organizational structure. It is possible to find success cases where the data and analytics activities management model is central and many other cases where there is a decentralized or hybrid model.
High-performing analytics organizations feature deep functional knowledge, strategic partnerships, and a clear center of gravity for organizing analytic talent. To implement an effective program, they must locate the analytical unit within the organization, which is most effective when linked to a business intelligence unit with a broad corporate perspective, choose the most appropriate organizational model (centralized, hybrid or decentralized), which depends on the type of company and its analytical maturity, and define the role of sourcing External, which must be minimized, because analytics is the “brain” of the organization.
Once the organizational model has been defined and the data strategy defined, it is essential to understand, based on business needs, which are the best analytical techniques and methodologies to use. Analytical developments always have data as a starting point and go through different levels of complexity: descriptive (seeking to answer questions about what happened), diagnostic (with greater focus on why it happened), predictive (supporting research on what will happen) and didactic (aiming to Determine the best course of action to take). The first two methodologies belong to the business intelligence domain, and are implemented through the development of custom dashboards, reports and analytics. The predictive and educational components belong to advanced analytics, which seeks to respond to business needs by developing predictive and optimizing models.
people and culture
The culture is largely dependent on the direction of leaders, especially the CEO, who, although he must take on a visible role, will have to choose an operating partner, such as a data area manager (Chief Data Officer) – a role that grows in importance, clarity and scope .
Francesco Costigliola, Professor at AESE Business School