In the world of data applied to utilities, finding the right balance between a centralized Business Intelligence (BI) strategy and a self-service approach is key to maintaining both agility and control.  

More and more organizations in the sector face the challenge of building an architecture that combines user flexibility with guarantees of quality, traceability, and security.  

The centralized model offers consistency, control, and governance, but it is often slower and less flexible. In contrast, self-service provides agility and autonomy to business areas, although it introduces risks such as duplicated analyses and loss of traceability.  

According to Gigliola Yemini, Project Manager Data & AI at Quanam, “an organization relying only on self-service would lack effective control and tend toward disorder; that’s why most organizations combine both models.”  

Pablo Werner, also Project Manager Data & AI at Quanam, highlights that “self-service is particularly suitable for areas like Marketing, where criteria and needs change so dynamically that traditional BI cannot respond in time.”  

With a hybrid approach, the BI team is no longer the main producer of reports but becomes a facilitator. It provides certified datasets, defines common standards, and ensures proper use of tools.  

Werner summarizes it: “The BI team becomes the provider of raw material, allowing each area to combine, process, and generate value based on its own needs. It also complements this role with support and monitoring tasks to ensure platform health.”  

The success of self-service depends not only on data access but also on user training.  

“It’s not just about teaching tools, but about understanding what data means and how it should be interpreted”, Werner warns.  

From a technology standpoint, many organizations operate with multiple platforms. IBM Cognos, Tableau, Power BI, and open-source tools like Superset are combined depending on user profiles and analytical needs.  

“Just a few years ago, working with two BI tools at the same time was unthinkable. Today, it’s the norm in most organizations” , Yemini notes.  

Not all units are ready to adopt this model. It is critical to assess their data maturity, prior experience, and level of analytical autonomy. When implemented without these conditions, it can lead to errors and duplicated efforts.  

Yemini points out: “Not every user is prepared to perform reliable self-service. Skills, knowledge, and certain controls are needed to ensure analysis quality.” 

Werner and Yemini describe the experience of a telecommunications company that began its analytics strategy with a traditional, centralized approach, supported by tools such as IBM Cognos.  

Over time, increasing data volumes and the need for faster responses drove a shift toward a more flexible architecture, based on Big Data and self-service solutions like Tableau.  

The result was a hybrid model, where multiple analytical platforms coexist, each tailored to different user profiles and needs. This transition accelerated processes in areas like Marketing and enabled the creation of monetizable data products, adding business value through internally developed analytical models.  

Predictive models, recommendation systems, generative assistants, and automated visualizations are expanding self-service capabilities. This increases user autonomy but also requires new forms of validation and governance.  

In this context, artificial intelligence does not replace self-service, it amplifies it. It enables users to move beyond consuming data and become active contributors to organizational knowledge. 

Finding the right balance between centralized BI and self-service is essential for utilities to respond with agility without losing control. With a robust architecture, skilled teams, and trained users, self-service becomes a strategic advantage rather than a risk. 

Gigliola Yemini

Project Manager Data & AI

Pablo Werner

Project Manager Data & AI

Alejandro Acle

Journalist