Matt Britton argues ChatGPT and similar AI models are pulling 50-60% of searches away from Google. He introduces "Geo," optimization for AI, and says the key is hyper-specific content: a model will surface your product if you target exact scenarios like "best pickleball racket for seven year old boy."
Chapters
How many people in the last 30 days have used ChatGPT for a query instead of saying you used to use for Google? Wow, that's about 50, 60%. This is a big deal. Think about how many times in the customer journey people are finding your brands through Google through SEO. The big term right now is Geo, which basically optimization for AI models. If you want to optimize to have ChatGPT recommend your brand, the specificity of that content matters. So for example, the model is going to know that somebody's child is 7 years old and if the they're searching for a pickleball racket, ChatGPT knows that your son is 7 years old and what it really wants to find is best pickleball racket for seven year old boy. And if it finds that, that's what it's going to push out. So if you actually extrapolate that towards the content you create, the specificity in content is what large language models are looking for because they have specificity and context and I think that's how you optimize moving forward. And next year 100 people are going to be raising their hands.