AI is changing how brands are discovered, evaluated, and ranked. Traditional SEO signals like backlinks and keywords are giving way to new AI-native signals such as structured data, community credibility, and verified factual content. This shift means marketers must now build both conversational relevance and informational authority to stay visible in AI-driven discovery.
Marketers are watching with anxiety as the familiar world of organic traffic and clicks erodes under the rise of AI search. The old rules are changing, but what are the new ones? Guesswork will not work.
The Semrush “AI Visibility Index Study” provides a data-backed look into this new reality. By analysing 2,500 prompts across major platforms like ChatGPT and Google AI Mode, the study decodes how AI models discover, filter, and rank brands. Its most critical finding reveals a Two-Stage AI Decision Process:
- Stage 1: Discovery. The AI asks, “Who is being talked about?” It scans listicles, reviews, and Reddit threads to build a consideration set based on user sentiment and discussion volume.
- Stage 2: Authority. The AI then asks, “Who can I trust?” It seeks clear, factual, structured content (like pricing and specifications) to validate the brands discovered in Stage 1.
This post distils the most surprising takeaways from the research. These are the counterintuitive truths you must understand to win at both stages in the new phase of AI-mediated discovery.
1. Your Google Rank Means Almost Nothing to AI
The study’s first finding dismantles a core assumption of digital marketing: the link between high traditional Google rankings and AI visibility is surprisingly weak. This marks a fundamental shift that renders old assumptions redundant.
The reason is that AI models operate on a “two-stage logic.” They prioritise a different set of signals than traditional search engines. The report found that factors like structured data, accessible pricing information, and third-party validation from industry publications now matter more than many legacy SEO metrics. AI is looking for clear, factual, and verifiable information it can easily parse and trust to move from discovery to authority.
Brands can no longer rely on existing SEO dominance to carry them. Winning in AI requires adopting a new mindset focused on AI discovery optimisation, building visibility through AI-native signals rather than conventional search engine algorithms.
2. AI Cares More About Reddit and Reviews Than Your Website
To win Stage 1: Discovery, you must dominate the platforms where real people talk. AI models learn about the world by reading what they say, relying heavily on user-generated content (UGC) from reviews, forums, and social media. The study calls this trove of public opinion the “new currency” for visibility - a crucial element in AI discovery optimisation, where community sentiment drives brand discovery.
The data is striking:
- Business & Professional Services, ChatGPT’s top sources are not corporate websites but Wikipedia (151.93%) and community discussions on Reddit (141.20%).
- This trend culminates in the “YMYL Paradox,” where for high-stakes “Your Money or Your Life” topics like Finance, Reddit is ChatGPT’s #1 source, cited in a staggering 176.89% of prompts.
This is a game-changer. AI learns about your brand from the unfiltered voice of the customer. This makes community engagement and review management essential marketing levers. Your reputation on platforms like Reddit, Quora, and industry forums is now a direct driver of your visibility in AI search.
3. Being “Mentioned” Isn’t the Same as Being “Cited”
The study uncovers a critical concept it calls the “Mention–Source Divide,” which is the direct result of the two-stage decision process.
In the world of AI, Popularity ≠ Authority.
- Mentions are the currency of Stage 1: Discovery. They reflect brand popularity and sentiment in user conversations.
- Citations are the currency of Stage 2: Authority. They reflect trust in your brand as a source of factual, verifiable information.
The “Zapier Paradox” from the Digital Technology & Software sector illustrates this perfectly. Zapier is the #1 cited source for Google AI Mode in its category, thanks to its vast library of factual integration guides. However, it ranks only #44 in brand mentions. Users talk about other tools, but AI learns from Zapier’s content.
The strategic implication is clear: brands need a plan to get mentioned to build sentiment and a separate plan to get cited as an authority. Excelling at one does not guarantee success in the other.
4. There Is No One-Size-Fits-All Strategy for AI
A key finding from the report is that AI search treats every industry differently. As the report states, “AI doesn’t treat every market the same, neither should you.” A winning strategy in one sector can fail completely in another, because the data ecosystems AI relies on are unique to each vertical.
Two industries highlight this divergence perfectly:
- Consumer Electronics: Google AI Mode operates with a clear “ecommerce focus,” citing retailers like Amazon (66.67%) and Best Buy (29.29%). In sharp contrast, ChatGPT takes an “editorial approach,” prioritising review sites like Tom’s Guide (58.63%) and TechRadar (49.20%).
- Fashion & Apparel: Visibility can be achieved through different routes. Patagonia (21.96% Share of Voice) dominates by owning the “ethical fashion” conversation in niche sources, which is a good example of AI search optimisation. Meanwhile, a high-fashion powerhouse like Gucci (14.20%) wins through decades of features in popular publications.
The lesson is that a successful AI strategy must be tailored, integrating AI SEO strategies that reflect each industry’s unique data and discovery patterns. Brands must first understand the unique information ecosystem of their specific vertical before they can hope to compete effectively.
5. Niche Dominance Beats General Fame
While iconic brands perform well in AI search, the study shows that being the undisputed leader in a specific niche is an exceptionally powerful driver of visibility. It’s more effective to be the definitive AI answer for a focused need than to be vaguely known for many.
Garmin, for example, achieves an impressive 31.15% share of voice in Consumer Electronics, demonstrating how a specialised category leader can compete directly with mass-market giants like Apple and Samsung.
Patagonia offers a masterclass in this strategy. Its success stems from executing a clear approach to own the “ethical fashion” subcategory:
- Category Definition: Rather than competing on general “fashion,” Patagonia defined and now owns the “ethical fashion” conversation, making it the default AI recommendation for conscious consumers.
- Consistent Editorial Placement: The brand appears in virtually every listicle and review for “sustainable brands” that AI models reference.
- Cross-Platform Authority: Patagonia has a strong presence across diverse sources, from Vogue and Forbes to Reddit discussions on corporate responsibility.
6. Hiding Your Price Is a Recipe for Negative AI Sentiment
For B2B and enterprise brands, this may be the most urgent takeaway: hiding your pricing is no longer a viable strategy. It is an active liability.
When AI models cannot find official pricing information behind a “Contact Sales” wall, they turn to user speculation on platforms like Reddit and LinkedIn to find an answer. These discussions often carry negative sentiment, complaining about high or opaque costs. This negativity then becomes directly associated with your brand in the AI’s response.
The solution is to provide LLM-crawlable pricing pages, supported by a strong LLM content structure. This allows you to win Stage 2: Authority by providing factual, verifiable information. You must control your narrative at both stages, or AI will find someone else’s.
Conclusion: Building for Authority, Not Just Attention
This brand new world of AI search demands a new playbook, where Generative Engine Optimisation principles integrate into content and platform strategies. Visibility is no longer a by-product of traditional SEO tactics but a new discipline entirely. Success requires a dual focus: achieving conversational relevance to earn mentions in Stage 1 while simultaneously building deep informational authority to be trusted as a source in Stage 2.
As the Semrush report puts it:
“In a world where AI mediates discovery, visibility is no longer earned by shouting the loudest, it’s won by being the most useful, the most cited, and the most trusted.”
Partner with Vajra Global for AI-Ready Marketing
At Vajra Global, we help businesses prepare for this new AI-driven marketing environment by aligning their content and data strategies with Generative Engine Optimisation best practices. From optimising your content for AI discovery to designing data-rich authority frameworks, our MarTech and digital specialists ensure your brand remains visible, verifiable, and valued across AI ecosystems. Partner with us to future-proof your marketing for the age of intelligent discovery.
