Estudios, datos y casos reales sobre cómo eligen las IA a quién citar.

Shopping ad impressions are falling while CTR climbs, and the industry is reading it as loss. After fifteen years inside the Google Ads auction, I…
Leer el artículo →
A B2B audit found that when an AI answer engine quotes a person, that person is on LinkedIn 74% of the time. The number is being read as a platform…
Leer el artículo →
AI content is now infinite and nearly free to produce. A wave of detection tools and platform rules is the market repricing human origin. The lesson…
Leer el artículo →
A new Semrush study shows AI Overviews expanding across high-value commercial searches – exactly the keywords where advertisers pay the most. For…
Leer el artículo →
A new study of 23 million AI citations found that 44% appear once and vanish, and the typical citation lasts 11 to 15 days. The number is real and…
Leer el artículo →
LinkedIn's Hiring Assistant ranks people the way AI engines rank brands: not by how often you state something, but by whether your structure lets a…
Leer el artículo →
On June 3, 2026, Anthropic explained how it makes Claude answer correctly using internal data: not better models, but structure. The same framework…
Leer el artículo →
The entire AI visibility market measures the same thing: how often AI mentions your brand. It is the easy number. It is also the symptom. The cause –…
Leer el artículo →
A clickstream study of 846,000 Google sessions shows that when an AI Overview is on the page, even users who searched for your brand by name pause to…
Leer el artículo →
At Google Marketing Live 2026, Google unveiled an entire generation of AI-rewritten ad formats. Gemini synthesizes the creative, builds the…
Leer el artículo →
Google deprecated FAQ rich results on May 7, 2026. It is not a feature update – it is the latest signal of an infrastructural shift. Search is going…
Leer el artículo →
Google has rolled out personalized AI search at scale. Two people, same query, different answers. The market is realizing what mature AI visibility…
Leer el artículo →
Google AI Mode now labels Reddit threads and forum discussions as "Expert Advice" inside its AI responses. For brands measuring AI visibility, this…
Leer el artículo →
TikTok just rolled out Tako, AI Overviews and AI-powered Related Posts. A new AI discovery surface enters the conversation — here is what changes for…
Leer el artículo →
xAI has released Grok 4.3 with a 321-point ELO jump on real-world agentic tasks and 40-60% lower pricing. A more capable, cheaper Grok means more…
Leer el artículo →
OnlyCFO calls 2026 the year of churn — and identifies AI brand strength as a direct churn trigger. If a competitor appears prominently in AI…
Leer el artículo →
A brand radar measures what's in the field of view of the systems that matter. In 2026, those systems are AI models — and most brands are invisible…
Leer el artículo →
Google has enabled direct purchasing inside Gemini, powered by the Shopping Graph. Users can now ask, compare, and buy without leaving the chat. The…
Leer el artículo →
Anthropic has released Claude Opus 4.7 — a significant leap in agentic reasoning, long-horizon task execution, and autonomous decision-making. For…
Leer el artículo →
Crazy Egg surveyed 49 shoppers on in-platform purchasing. 81% knew it existed. Only 27% had ever done it. 80% still prefer brand websites. The brand…
Leer el artículo →
New research shows that 64% of users in AI Mode never click anything — they read the AI's answer and declare their finalists. The funnel is not…
Leer el artículo →
Meta's Superintelligence Labs has released Muse Spark — a proprietary model that will power Meta AI across Facebook, Instagram, WhatsApp and Ray-Ban…
Leer el artículo →
Semrush analyzed over 1 billion lines of clickstream data. ChatGPT referral traffic grew 206% in a year — but 21.6% of it went to Google alone, and…
Leer el artículo →
Most AI tools give you one model. Your AI Room puts three in genuine conversation — Claude, ChatGPT and Gemini reading each other's responses,…
Leer el artículo →
AI agents are becoming buyers. They don't respond to charm pricing, they skip "contact sales", and they evaluate brands based on structured,…
Leer el artículo →
New research maps how licensing deals and ownership structures determine which social platforms each AI model can cite. It's a structural constraint…
Leer el artículo →
OpenAI is testing ads in ChatGPT. Google is building toward it. Anthropic is refusing entirely. While the industry debates the ad model, the real…
Leer el artículo →
Anthropic has launched computer use for Claude — the AI can now autonomously perform tasks on your machine. A step forward in agentic AI, and a…
Leer el artículo →
AI is compressing agency margins from both sides. Clients bring work in-house, budgets shrink, and execution gets commoditized. Here is why Citation…
Leer el artículo →
AI Overview increases impressions but reduces clicks. Why this phenomenon is becoming a structural problem for SEO, brand visibility, and the role of…
Leer el artículo →
AI is becoming a traffic channel. CTR is losing relevance. Here are the three foundational pillars to build your brand's citability — content,…
Leer el artículo →
At AI Tech Expo London, one thing became clear: most tools measure citation as a binary fact. That's structurally broken. Here's why probability…
Leer el artículo →
There is no perfect prompt. But there is a structure that eliminates arbitrary interpretation. A JSON-inspired method to get clearer, more replicable…
Leer el artículo →Datos, estudios y casos reales sobre cómo aparecer en las respuestas de las IA. Sin spam: te das de baja cuando quieras.