Onlineshop-Geflüster - Der E-Commerce & Shop Podcast · Berend Heins

So wirst du von ChatGPT empfohlen (🎙️ Maurice Marinelli von findling & finseo.ai)

·52 min·4 clips

Maurice Marinelli explains how to get ChatGPT to recommend your online shop instead of just claiming you're the best.

1. Onlineshop-Geflüster hosts Maurice Marinelli, founder of Findling SEO agency (clients include Snogs, LFDY, Jägermeister, Hertha BSC) and co-founder of finseo.ai, to discuss how brands get recommended by ChatGPT and AI search. 2. Maurice is a returning guest who appeared in a previous SEO three-part series, and this episode extends that earlier content into AI-specific optimization. 3. The episode's central thesis is that getting found by AI search tools is not a new discipline separate from SEO, but an evolution of it — with 82% overlap according to Maurice's working rule of thumb. 4. The host describes a personal behavioral shift: he uses ChatGPT more than Google for research, though Google's own AI overviews have recently pulled him back, illustrating the fragmented and evolving search landscape. 5. Maurice explains that the fundamental question is not 'will ChatGPT overtake Google' but how human search behavior is changing — people now supply full context in queries rather than entering isolated keywords. 6. Maurice cites a Cologne affiliate blogger who ran informational content sites funneling to retailers like Amorelie and lost over 90% of traffic to Google AI overviews answering queries directly. 7. At the Shop Talk conference in Barcelona in June, Maurice received a Google demo showing that even in AI mode, product listings and advertising will remain integrated — contradicting early predictions that AI overviews would eliminate ad revenue. 8. Maurice states his 82-18 principle: 82% of whether ChatGPT recommends a brand comes down to standard SEO fundamentals already being in place. 9. The basic SEO checklist he itemizes includes: pages not on noindex, keyword-driven category architecture with subcategory depth, meta titles on every page, adequate page speed, and category-page descriptive text. 10. He warns against 'charlatan' GEO agencies selling fear-based products claiming SEO is dead, arguing that brands should do foundational SEO first before exploring AI-specific tactics. 11. The host describes his neoprene triathlon suit purchase: he wrote a full context prompt to ChatGPT specifying his triathlon goal, target finish time, amateur status, budget ceiling, and multi-year durability need, and received a direct product recommendation. 12. Maurice calls the difference between a keyword and a prompt 'context and demand orientation' — prompts describe a life situation and specific need rather than a product category. 13. He describes three methods for identifying prompts to optimize for: post-purchase customer surveys asking what terms they entered into ChatGPT, keyword research tools identifying search volume, and a 'prompt matrix' combining customer demographics, life situations, specific problems, and product lines. 14. Maurice recommends using prompt monitoring tools that survey ChatGPT, Perplexity, and Google AI Overviews daily to track whether and how a brand is being mentioned or recommended. 15. He warns that AI systems surface reputation data from across the internet, so brands with negative reviews on external platforms can be actively recommended against by ChatGPT. 16. For AI visibility improvements, Maurice presents a four-part framework from his K5 conference presentation: prompt research, application-oriented categories, FAQ with structured data, and external trust signals. 17. Application-oriented categories are illustrated by a workwear client creating 'work pants for electricians' and 'work pants for farmers' subcategories, and a natural bumper client creating model-specific roof tent pages for a Skoda Yeti. 18. For trust signals, Maurice notes that Reddit, YouTube, Quora, Wikipedia, and LinkedIn are among the most frequently cited sources by ChatGPT, and UGC content consistently outperforms brand-produced content as an AI reference. 19. E-commerce brand owners and digital marketing professionals who want a practical, skepticism-filtered introduction to GEO tactics that builds on existing SEO investments will find this episode directly actionable. 20. Listeners expecting a purely technical deep-dive into AI model architecture or who have already mastered foundational SEO and want advanced LLM-specific optimization strategies will find the coverage too foundational.

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Onlineshop-Geflüster - Der E-Commerce & Shop Podcast