Hottest topic in my limited marketing circles is how do you get your solution or company to come up as part of an answer with LLM's like ChatGPT, Claude, etc.? I have heard thoughts as to the content on websites needing to be more of a narrative for the LLM's to learn from, but that is about it. What are you hearing these days?
I'm hearing from many that if you are using LLM for company efficency, that you need to keep a human in the mix in order to monitor voice accuracy and quality control. Seems like a logical idea to me.
ChatGPT got me jumping through all sorts of hoops like getting added to wikidata.org. Having good structured data apparently helps.
EEAT. I read something a few months ago about adding a LLM.txt at the root of a website to bait or attract AI robots. There's 0 proof this works. I might be wrong here, but think about something simple : where do AI scrappers get their data ? Are websites indexable by AI robots or does the scraping technology depend on search engines ? If they indeed depend on search engines like Bing, Google, etc... how do they select their sources ? SEO, is still the springboard in the end. EEAT + how to's + FAQ + entity-focused content are the best leads, in my humble opinion. I don't trust the GEO, LLM consultants or whatever, who are all day and night on linkedin. No one knows the perfect method for GEO and LLM. The same truth applies for Google's algorithm. We just need to keep trying, because I also think that niches are important here. I dare anyone on Linkedin to prove me that he can make any site, in an ultra high competitive niche, to appear as recommendations on chatgpt or whatever. It's impossible.
@EAsports your points are all very valid. For a long time now, I have been telling anyone who will listen that it is a fool's game to base a big chunk of one's business on SEO. It is too volatile and that is by design. That said, when search engines first came out, you could "find the angles" in the then immature technology and sometimes do quite well. I am not talking about black hat, but just wondering if these newborn LLM's have any growing pains when it comes to developing their answers and if an online presence might be optimized to benefit from any such quirks.