Artificial intelligence is no longer an experimental channel for the insurance sector but rather a new decision space. This is demonstrated by the first AI visibility ranking in the insurance sector in Spain carried out by Product Hackers at a time of profound transformation of the sector that sees how it is no longer enough to exist on the internet but needs to be a reference for the LLM models that increasingly mediate the user’s decision.
In fact, according to SEMrush data cited in the report, traffic referred by AI tripled in the first half of 2025, going from 2% to 6% of the total, and forecasts suggest that in 2027 traffic from AI could surpass that of traditional organic search.
“This study of visibility in AI arises from the need to understand what criteria determine the relevance for AI today, in a context of incipient but accelerated change,” explains Luis Díaz del Dedo, CEO of Product Hackers. “We are seeing how traffic from AI stops being marginal and becomes a strategic channel, and brands that do not understand its rules run the risk of becoming invisible.”
Hyperconcentration: only 5 brands occupy the visibility zone in AI
Mapfre, AXA, Adeslas, DVK and Allianz. These are the insurers that dominate AI visibility for artificial intelligence models and smart search engines, according to the analysis, highlighting the extreme concentration that the sector is experiencing.
Specifically, Mapfre leads the AI visibility ranking with 8.87% visibility, followed by AXA (6.28%), Adeslas (4.47%), DKV (4.46%) and Allianz (4.02%). Starting from the top 5, visibility drops sharply and more than 20 brands remain below 1.5%, a level that in AI environments is practically equivalent to not appearing in the majority of responses.
If we analyze the second level (positions 6 to 15), we see relevant brands such as Sanitas, Asisa, Santalucía, Caser and Zurich, with shares that are below 3%. Here begins a more pronounced decline, with a more punctual presence and focused on specific verticals (health, life, etc.). From position 20 onwards, we enter the zone of marginal visibility. More than 20 insurance brands have a share of voice of less than 1.5%, which in LLM environments means being out of the question in most scenarios.
This happens because artificial intelligence models do not work like a traditional search engine with 10 results. They offer a short answer, with very few brands, and tend to always choose those that already have repeated authority in many contexts. Therefore, being part of the top 5 is essential to have a presence in them. Much more important even than it was for SEO positioning in search engines like Google.
“LLMs do not distribute authority like Google,” says Ricardo González, Head of Growth at Product Hackers. “They systematically favor the 3–5 best positioned, leaving the rest out. This is what we call the ‘AI visibility zone’: being outside of it implies being excluded from the decision process through this channel.”
Comparators and new players gain weight compared to traditional insurers
After analyzing the sector in nine key categories (health, life, car, motorcycle, accidents, civil liability, travel, death and home) one of the most relevant findings of the Product Hackers study is the weight of aggregators and comparators. 40% of the sources most cited by LLMs belong to this type of platforms, compared to the 25% represented by insurance companies’ own websites.
Domains such as Polizamedica (4.2%), Rastreator (3.88%) and Adity (3.7%) lead the references in the majority of categories analyzed. In some sectors, even YouTube videos help reinforce positioning.
Furthermore, and although the concentration pattern is repeated, each branch responds to different logics. In the general category, the large insurance companies stand out, Mapfre, AXA, Adeslas, DKV and Allianz, but in specific categories other players emerge.
“In the home, for example, Tuio, a relatively new brand, occupies second place thanks to a clear, simple content strategy adapted to the user’s language,” explains Díaz del Dedo. “And in niches like motorcycles, specialized insurers like AMV have gained visibility by having content very focused on the specific needs of motorists.”
This shows that size is not everything. “The best positioned are not always the largest,” adds the CEO of Product Hackers. “There are categories where specialization and clarity surpass dimension.” For example, in travel, where behavior is differential: IATI (18.14%), Intermundial (16.85%) and Heymondo (15.92%) clearly dominate, in an environment where AI prioritizes specialized blogs and experiential content over general comparators.
Content is still king but with different rules
The report concludes that AI does not actively “penalize,” but it does disfavor certain patterns. “Models tend to ignore information that is poorly structured or difficult to interpret, especially if it is not organized in tables, FAQs or clear comparisons,” explains Ricardo González.
“In addition, the lack of consensus is critical: if different sites say something different about a brand, the model does not construct a clear and probable ‘opinion’ about it.” Thus, far from losing relevance, content continues to be decisive, although under new rules. “Content is still king, but with a different strategy,” says González. “The important thing is how we structure it so that the AI consumes it, how we respond to specific questions that the user asks in a chat and, above all, how we distribute it. The reference domains are decisive.”
