Market Like a Genius with Ahmet Dogan
Listen to Ahmet Dogan on Spotify: Hyper-Technical ChatGPT Ads Product Discovery Playbook: Conversational Commerce, Intent Resolution & AI-Powered Recommendations - By Ahmet Dogan, CEO of Citatix ChatGPT Ads Agency
Transcript of the Episode
Welcome back to The ChatGPT Ads Playbook. I am Ahmet Dogan, CEO of Citatix ChatGPT Ads Agency and an official OpenAI Select Partner. Today we are going hyper technical. We are talking about product discovery inside the chat: what happens when shoppers stop navigating traditional ecommerce websites and start asking AI systems what they should buy. Traditional ecommerce is built around navigation. The consumer enters a category, submits a keyword query, applies filters, opens product pages, compares specifications, reads reviews, and eventually makes a purchase decision. Conversational commerce changes that architecture because the interface itself becomes the discovery layer. Instead of navigating a catalog, the shopper expresses intent in natural language. The system then has to interpret that intent, extract constraints, retrieve relevant products, evaluate candidates, rank them, and generate a recommendation. From an AI engineering perspective, product discovery is becoming an intent-resolution, retrieval, ranking, and recommendation problem rather than simply a keyword-search problem. The system can process semantic intent, contextual signals, budget constraints, compatibility requirements, product attributes, preferences, and conversational context. That means brands need to think beyond traditional SEO and ecommerce merchandising. Your product has to be understandable to machines. Product titles, descriptions, structured attributes, specifications, variants, categories, pricing, availability, compatibility information, use cases, and measurable product characteristics can all become signals within the product-discovery pipeline. Think about the process as candidate generation followed by contextual evaluation. A shopper expresses an intent. The system constructs a semantic representation of that intent. Relevant products are retrieved from available product knowledge and commercial data. Candidates are then evaluated against constraints and contextual relevance. The system can produce a recommendation and explain why specific products fit the user's requirements. This is where product data quality becomes strategically important. A vague ecommerce description may be acceptable for a human browsing a website, but conversational systems benefit from precise, structured, machine-interpretable information. If your product is described as powerful, what does powerful actually mean? What processor does it use? How much memory does it have? What workloads is it designed for? What environments does it support? What devices is it compatible with? What are its measurable performance characteristics? The more precisely a product is represented, the more semantic signals an AI system has available to understand, differentiate, retrieve, and compare it. We also need to distinguish search ranking from recommendation relevance. Search ranking asks which products should be returned for a query. Recommendation relevance asks which products best satisfy this particular user's intent, constraints, context, and preferences. Those are different optimization problems. A product can have strong visibility for a broad commercial keyword while being a poor recommendation for a highly specific use case. This changes how ecommerce teams should approach optimization. Instead of optimizing only for keywords, brands need to optimize the product knowledge layer: entity resolution, attribute coverage, taxonomy, semantic relationships, structured product information, product differentiation, and clearly defined use cases. The objective is to make the product easier for an AI system to understand, retrieve, compare, and recommend. Conversational shopping also compresses the traditional marketing funnel. Awareness, consideration, education, comparison, objection handling, and purchase consideration can happen within a single interaction. A shopper can ask what to buy, ask why, compare two products, introduce a budget constraint, specify a compatibility requirement, challenge the recommendation, and continue refining the decision without leaving the conversational environment. That creates a different commercial battleground. The question is no longer simply, how do I get my product in front of consumers? The question becomes, how do I make my product a relevant candidate when consumers ask AI what they should buy? And this is where ChatGPT Ads becomes particularly interesting. Shopping is moving from navigation toward conversation. For brands now.