How a Heritage Jewellery Brand Took Control of Its AI Search Narrative: A GEO Case Study

What Happens When AI Becomes the New Search Engine

Search has changed. For years, ranking on Google Page 1 was the finish line. Today, the finish line has moved. When someone searches a brand name or a product comparison, Google increasingly shows an AI Overview before any traditional link. When someone opens ChatGPT, Gemini, or Perplexity and asks "is this brand worth it," they get a generated answer, not a list of websites to click through.

This shift is the entire premise behind GEO, or generative engine optimization. GEO is not a replacement for SEO. It is the next layer built on top of it. If SEO answers the question "how do I rank on Google," GEO answers the question "how do I get cited inside the answer an AI gives, whether that AI is Google's own AI Overview, ChatGPT, Gemini, or Perplexity."

This case study walks through how Heramb applied a GEO and AI SEO strategy for a large, well known Indian heritage jewellery brand (name withheld for confidentiality) that was facing a specific and measurable problem: its brand was being discussed constantly online, but the conversation was happening without the brand's side of the story, and AI platforms were already citing that conversation directly in their answers.

What is GEO and Why Does It Matter More Than Traditional SEO Right Now

Before getting into the case study, it helps to define the terms clearly, because this is one of the most searched and most confused topics in digital marketing today.

What is GEO? GEO stands for generative engine optimization. It is the practice of structuring and placing content so that generative AI systems, including Google's AI Overview, ChatGPT, Gemini, and Perplexity, are more likely to read it, trust it, and include it when generating a response to a user's question.

What is AEO? AEO stands for answer engine optimization. It is closely related to GEO and sometimes used interchangeably, with a slightly tighter focus on optimizing content to directly answer specific questions in a way that voice assistants and answer engines can extract cleanly.

GEO vs SEO: what is the actual difference? Traditional SEO optimizes a page to rank in a list of ten blue links. GEO optimizes content to be selected, summarized, and cited inside a single generated answer. SEO cares about keywords, backlinks, and page structure on your own website. GEO cares about whether your brand's information exists in the places an AI is already pulling from, which are very often Reddit threads, Quora answers, YouTube comments, and independent review platforms, not brand owned websites at all.

This is the part most businesses miss. You cannot GEO-optimize your own homepage and expect ChatGPT to start recommending you. AI systems are trained to distrust brand-owned content because it is inherently biased. They are trained to trust community generated, peer reviewed, independent content. That is precisely why this case study centers on Reddit and Quora rather than the brand's own website.

The Brand's Starting Situation

The client is one of India's most recognizable heritage jewellery retailers, operating hundreds of stores nationally with a multi-decade brand history. Despite this scale and recognition, an audit of how the brand appeared across AI platforms and Google's AI Overview revealed a critical gap.

A simple search like "is [brand] making charges high" returned a Google AI Overview response that cited a Reddit thread and a Quora source directly in its answer panel. The AI Overview stated a making charge range without the comparative context that would have favored the brand, framing the cost as high without explaining the value or transparency behind it. This was not a hypothetical risk. This was the live, current AI generated answer being shown to anyone searching that exact phrase, sourced directly from existing third party community content the brand had no presence in.

This single screenshot became the clearest possible proof point: AI Overview is already citing Reddit and Quora for this brand's most sensitive queries, and the brand was not part of that conversation at all.

The Strategic Approach: Building a GEO and AI SEO Presence From the Ground Up

The strategy was built around one core principle. Instead of trying to optimize the brand's own website for AI visibility, which AI systems are inherently skeptical of, the focus shifted to becoming part of the actual conversation happening on the platforms AI already trusts and cites.

Step One: Mapping Where the AI Is Already Looking

Before creating a single piece of content, the existing conversation was audited in detail. This meant identifying exactly which Reddit threads, Quora answers, and review platforms were currently ranking on Google Page 1 and currently being cited inside AI Overview, ChatGPT, and Perplexity responses for the brand's most important queries.

This is a critical and often skipped step in any serious AI SEO services engagement. Without knowing exactly which existing sources an AI is pulling from, new content is published blindly. With this audit complete, every subsequent piece of content had a specific, confirmed target.

Step Two: Becoming Part of the Conversation on Reddit and Quora

Reddit and Quora were treated as the two highest priority platforms, for a specific and provable reason: both were already directly cited inside the brand's AI Overview results. New, original, story-format content was created for Reddit, written as genuine first-person experiences rather than promotional posts. Long-form, structured Quora answers were written to directly and honestly address the exact questions buyers were already asking, the same questions AI systems were already trying to answer.

The content was never designed to deny or argue against existing criticism. It was designed to add a missing, credible, detailed perspective into the same conversation, so that when an AI system summarized "what people say about this brand," it had more than one side of the story to draw from.

Step Three: Expanding to YouTube Comments and Long-Form Articles

Beyond Reddit and Quora, two additional channels were brought into the strategy specifically because of how Google's AI Overview and other large language models are now sourcing information.

YouTube is the second largest search engine in the world, and Google's AI Overview has begun pulling directly from YouTube video pages and their comment sections for review and experience-based queries. Genuine, specific comments placed on high-view jewellery review and shopping guide videos created an additional layer of indexed, AI-visible content addressing the brand's core narrative points.

Long-form articles on high domain authority publishing platforms, including Medium and LinkedIn Articles, were used to directly target informational and "how to" style queries. These platforms carry domain authority in the 90s, frequently higher than most news websites, and AI systems weight this kind of high-authority, long-form, well-structured content heavily when constructing explanatory answers. An article explaining how a specific billing practice works, published on a trusted third-party platform, became exactly the kind of source an AI system would choose to cite when explaining that same concept to a user.

Step Four: Structuring Every Piece of Content for AI Extraction, Not Just Human Reading

A defining part of this AI SEO and GEO methodology is that every piece of content was written with two readers in mind at once. The first is the actual human inside that Reddit thread or reading that Quora answer. The second is the AI crawler that will eventually read that same content and decide whether to extract it into a generated answer.

This meant leading with clear, direct answers to the exact phrasing of common queries, structuring long-form answers with scannable subheadings, and ensuring that factual, specific details, not vague brand praise, were present in every piece. This is the single biggest difference between traditional content marketing and true GEO content. Traditional content is written to be read. GEO content is written to be read and extracted.

Measuring Success: The KPIs That Actually Matter for GEO

A core part of this engagement was establishing the right way to measure progress, since traditional SEO metrics alone do not capture what GEO is actually trying to achieve. The KPIs tracked included Google Page 1 presence for the brand's priority keyword list, checked on a recurring basis, manual AI mention audits across ChatGPT, Gemini, and Perplexity using a fixed list of priority prompts, with before and after screenshots as evidence, sentiment ratio tracking across all newly identified mentions, and direct monitoring of whether Google's AI Overview began incorporating the brand's positioned narrative for its most sensitive queries.

This is the part of GEO that most ai seo tools cannot automate. Software can track keyword rankings. It cannot manually prompt five AI platforms with the same question every month and document exactly how the generated answer is changing. That verification work has to be done by a person, repeatedly, on a fixed schedule.

Why This Matters for Every Brand Considering AI Marketing in 2026

This case study reflects a pattern playing out across nearly every consumer facing brand in India right now. Whether the category is jewellery, banking, FMCG, or healthcare, the underlying mechanism is identical. Generative AI platforms are answering category-level questions using a relatively small set of trusted, third party sources, and most brands have no presence inside that source set at all.

This is the gap that ai marketing and traditional digital marketing agencies are not built to close. A normal performance marketing agency optimizes ads and your own website. A GEO and AI SEO focused agency works inside the actual conversation an AI is reading from, on Reddit, on Quora, on YouTube, and across high-authority publishing platforms, which is where the real influence over generative AI answers currently lives.

As more consumers move from typing into Google to asking ChatGPT, and as Google's own AI Overview becomes the default result for an increasing share of searches, this is no longer an experimental marketing channel. It is becoming the primary battleground for brand perception, and the brands that establish a credible, well-structured presence early will be the ones generative AI systems trust and cite for years to come.

Key Takeaways