The Infinite Loop of Self-Corroboration

Kasra Dash: Today we're going to be talking about the infinity loop of selfcorroberation. I am joined with James and this is probably one of the most commonly used words that James uses nowadays. So let's uh let's kick start it off for for anybody that that has never heard of it, doesn't know what it is, explain it. James Dooley: Yeah. So the infinite loop of self-corroboration was actually I picked it up from Jason Barard at Calube. So he does a lot with regards to knowledge panels and what he's saying is the messaging that you kind of present on your own website and who you are and what you do and who you serve is on your website and it's served up within the schema and obviously on social media and all the platforms that you control. But then what you need to try to do is get that's on firstparty sources. you need to try to get that information on second party sources and also on third party sources. So what that is is that's trying to get other people to say the same thing and confirm those assumptions or that bit of information. And the more times it's on trusted sources, the more it changes from being an assumption to being a fact. Now, what I love about this is not only is it great for the knowledge panels, it's unbelievable for AI ranking and ranking in the LLMs. So, I don't know if you want to expand on that a little bit more with regards to the infinite loop of self collaboration. Kasra Dash: Yeah. So, one thing that's really important to understand is you've obviously got first party sources, second party sources, and third party sources. So first party sources would be things that you actually control like your website, your Google business profile, social media accounts. So you would want to make certain that there's a consistency across all of those things. And what I mean by that is if say for example you are mentioning that you are an awesome baker in Manchester, you want every single social profile saying a very similar um description about you. As soon as you start changing it and it's not aligned with your brand, that's when there is doubt and that's when Google and also all of the LLMs don't know is this the social media of James Dooley? Is it not? So you want to make certain that everything across the board is being consistently said in the same manner. James Dooley: Yeah, for sure. I mean, I think the the main part on this is to do with um we're calling it now internally semantic triple factoids and we got that from Luke Baston who's a very advanced semantic SEO and I at first was uncertain what a factoid even meant and it was getting a a piece of unreliable information and when it's repeated so many times it becomes a factual piece of information within the algorithms and that's kind of how the LLMs are working. So they're not using page rank and link juice. Now it's to do with how many times is that messaging being repeated and when it's repeated so many times on different trusted sources the LLMs are seeing that as being facts. So it's great for rankings in perplexity in chat GPT in Gemini as well as then using those micro semantics with the semantic triple factoids to also be feeding to get that KGM ID and that a KGM ID stands for a knowledge graph machine ID which means it turns your actual website into a known entity. That's when you're a brand. If if if you do not have a KGM ID set up against your website, you are not a known brand in Google. And this trust signal is in my opinion where a lot of sites got hit with a helpful content update as well. I know it's kind of a broad kind of thing to to say. I'm not saying it's the only reason, but having that trust and being a known entity and building that brand is so important. Getting those knowledge panels is so important, but actually feeding them LLMs as well. I think it's absolute crucial for now for every single website, whether you're a local site, an e-commerce, an affiliate site, getting that infinite loop of self collaboration is so important. Kasra Dash: Yeah. And then we've obviously got second and third sources as well. So, do you want to dabble into this to the second party sources? James Dooley: Yeah. So, the first ones are obviously easy to get because that's on your own website. The second party sources could be on guest posts where you're physically writing the guest post. It's on a third party website or it could be a mixed opinion. It could be a Reddit forum where you and others are kind of explaining who you are and what you do and they're giving feedback. You're able to respond and stuff like that. Now, it's better to then get if you can then third party. These are people that these are people talking about you when you're not in the room. So, these are people on their own websites basically saying who you are, what you do, how amazing you are, and actually corroborating to make certain that you're going, "Yes, James Dooley is 23 years old. Yes, James Dooley does live in Manchester. Yes, James Dooley is an entrepreneur." And it's kind of stating those facts and you're laughing about me being 23 years old, but obviously I might be a year or two older than that. But the uh but it's stating those facts and when that gets repeated so many times it becomes fact and now because I see it's not fact about me being 23 because it's only me that says it and everybody else is that you're not. But you get where I'm coming from. These third party sources where you can try to get other people to kind of stack and compound those statements and make them assumptions of you're the founder of this business that you do this that you do that. this is where you're based, this is your date of birth. All this information then kind of molds around, okay, we now have a higher confidence score with who you are. And that's when not only creating a knowledge graph u machine ID, but actually improving that confidence score as well is what builds the trust of the brand. Kasra Dash: Yeah. So, that's been our video on the infinity loop of self-corroboration. If you guys do have any questions, feel free to drop a comment down below. There will also be a link in the description to a full-on article that James has spent probably 62 hours writing. Um, so if you want to learn a little bit more, head over to that link and get all of the information from that. Thanks for watching.