From Search Rankings to AI Recommendations: The New Brand Playbook

In the latest edition of Media Mohalla – MM Talk, we feature Aditya Kathotia, Director & Founder of Nico Digital, an SEO and AI-driven growth agency.

With 15+ years of experience, Aditya is a digital business leader known for bridging technology, technical SEO and business strategy. He began his entrepreneurial journey at 17 and has since bootstrapped and scaled four ventures.

Under his leadership, Nico Digital has grown into an 80+ member agency with offices in Kolkata and Mumbai, serving brands including Škoda, Birla Group and NDTV Foods.

In this edition, Aditya shares his insights on AI, SEO, digital transformation and the future of brand growth.

1. The Discovery Moment Has Already Moved

Nearly seven in ten Google searches now end without a single click, up sharply from roughly six in ten just two years ago. Where a results page once appeared, an AI-generated answer increasingly sits instead, and when it does, click-through to the pages beneath it falls by more than half. ChatGPT alone crossed 900 million weekly users in February 2026, doubling in a year, and India now contributes over 100 million of those weekly users, second only to the United States. The consumer journey marketers have spent twenty years mapping around keywords, rankings and landing pages is being replaced by something that behaves less like a search and more like a conversation.

2. From Keywords to Context

Traditional SEO optimised for a query: a fixed string of words matched against an index. Conversational discovery optimises for intent expressed in context, refined across several turns. A shopper no longer types “best waterproof hiking boots under five thousand rupees.” They describe a trek, a budget and a foot problem, and expect a recommendation, not a results page. Brands built only around discrete keyword clusters will find themselves invisible here, because the AI system is not matching words, it is synthesising an answer from whichever sources it trusts enough to cite.

3. Measurement Needs a New Vocabulary

Impressions and clicks assumed a page the user could see and act on. Inside an AI conversation, there is often no page at all, only a recommendation and, sometimes, a named source. The nearest equivalent to an impression becomes a citation: how often, and how prominently, a brand is named inside an AI-generated answer. The nearest equivalent to a click becomes the action a user takes right after that mention, whether that is asking a follow-up question or navigating directly to the brand. Advertisers should expect targeting to shift away from bidding on queries and toward earning consistent presence across the sources these models draw from, since there is no auction inside the answer itself, at least not yet.

4. Authority Is Built Outside Your Own Website

This is the part most brands underestimate. A large majority of brand mentions inside AI search responses originate not from a company’s own website but from third-party pages: review platforms, trade publications, analyst commentary, independent comparisons. Earned media of exactly this kind is understood to drive the overwhelming share of AI citations, and brands with a presence across several independent, credible sources are markedly more likely to be surfaced than brands relying on owned content alone. In practice, digital PR and traditional SEO are converging. Getting credible third parties to write about a brand accurately is no longer just a reputation exercise; it is now a direct input into whether an AI system recommends that brand at all.

5. What Actually Gets Cited

Format matters more than most content calendars assume. Content structured as direct question and answer performs consistently better at earning citations than narrative prose. Lists, comparison tables and step-by-step formats are cited at meaningfully higher rates than unstructured paragraphs, and content carrying original data or a clearly sourced statistic is cited more often than opinion alone. Freshness compounds this: pages left unrevised for long stretches are considerably more likely to fall out of an AI system’s citation pool within a year. None of this replaces good writing. It simply means good writing now needs to be extractable, current and verifiably sourced, not just persuasive.

6. The Brands That Win This Shift

The brands that adapt first will not be the ones chasing every new AI platform update. They will be the ones that treat authority as something built deliberately across the open web, not just on their own domain, and that structure their content so an AI system can lift a clean, accurate answer out of it. Search is not disappearing. It is becoming a layer the customer no longer sees, which makes being the trusted source behind that layer the entire game.