Case Study / Analysis and management in AI engines

Case Study: Brand Presence Analysis in Artificial Intelligence Search Engines of Iberdrola

berdrola, a company in the energy sector with a strong international presence, identified the need to analyze its presence in artificial intelligence search engines in order to understand how its brand was being represented in generative response systems. The evolution of conversational search environments has introduced new dynamics of digital visibility, where information is no longer only ranked but also synthesized and delivered directly as an answer. In this new context, it became essential to systematically evaluate how the brand appeared across different language models and what types of context and sources were being used to construct the responses.

Contact with us

Solution:

The project focused on analyzing the behavior of AI engines in response to different types of queries. The main objective was to build an initial snapshot of brand presence, informational treatment, and citation patterns in automatically generated responses.

To achieve this, an evaluation framework was designed based on different types of prompts that replicate the main ways in which users and stakeholders interact with artificial intelligence engines.

The analysis included:

brand-oriented queries
sector-related questions
transactional intent scenarios
informational searches

This segmentation made it possible to observe how brand representation varies depending on the intent of the query and the context of the question posed to the system.

Based on a broad set of previously defined prompts, the responses generated by different AI engines were reviewed, paying attention not only to the textual content provided but also to the narrative structure of the response, the positioning of the brand within the sector, the type of sources associated, and the sentiment conveyed in the mentions. The analysis examined the coherence of the information, the consistency across engines, and the recurrence of certain external references.

The applied methodology focused on comparative and qualitative observation of results, aiming to identify common patterns of appearance, forms of contextualization, and citation behaviors. This approach made it possible to detect differences between models, variations depending on the type of prompt, and opportunities for improvement in terms of the quality and alignment of the sources used to construct generative responses.

Result

As a result of the study, a structured understanding was obtained of how artificial intelligence engines interpret and present information associated with the brand across different query scenarios. The analysis made it possible to identify areas for strengthening presence through both owned and third-party sources, as well as areas where representation could improve in terms of precision and contextual coherence.

Based on these findings, an initial framework for managing positioning in AI environments was also defined. This framework focuses on prioritizing narrative development through owned assets, reinforcing strategic publishing in media sources used by AI engines as references, and establishing intervention guidelines for managing negative mentions or references. At a sector level, the diagnosis also revealed opportunities to anticipate and occupy conversation spaces ahead of competitors, contributing to building a stronger and more consistent narrative in the responses generated by different systems.

Conclusion

This case reflects how digital presence management is evolving towards automated response environments and how analyses based on structured prompts make it possible to assess brand representation in AI systems at an early stage. Having this type of diagnostic enables more informed strategic decision-making and helps adapt digital visibility to the new ecosystem of generative engines, maintaining a prudent, general, and continuous improvement-oriented approach.