Google API Leak & E-E-A-T

Hello and welcome to. Seo is not that hard. I'm your host, ed Dawson, the founder of keywordspeopleusecom, the place to find and organise the questions people ask online. I'm an SEO developer, affiliate marketer and entrepreneur. I've been building and monetising websites for over 20 years and I've bought and sold a few along the way. I'm here to share with you the SEO knowledge, hints and tips I've built up over the years the SEO knowledge, hints and tips I've built up over the years. Hello, it's Ed here.

Welcome to this latest episode of SEO's Not that Hard. Today I'm going to talk a bit about the Google API leak that's happened recently and what we can learn from EEAT with it. Now, if you're not familiar with EEAT, that's a concept of experience, expertise, authority and trust. Now, this is a sort of framework slash model that Google use within their Google search rate guidelines thousand or so google raters that they employ around the world to analyze websites and rate them on page quality and expertise and things like that. I did a whole series of episodes on the google quality raters and the guidelines they follow, from episodes 27 to 34. So if you go back and listen to them, if you're not familiar with the Google Quality Raters and the guidelines they work to, but the core of the guidelines the raters have to think of pages in terms of experience, expertise, authoritativeness and trust. That's the E-E-A-T in the acronym. Now, experience is considering the extent to which a content creator has the necessary first-hand experience or life experience for the topic. Expertise is considering whether the extent the content creator has the necessary knowledge or skills for the topic. So a bit like experience, but slightly different. Then authoritativeness is whether to what extent the content creator or the website is like a go-to source for the topic. Doesn't have to be official one one, but they just have to be like very authoritative. And then trust is the final one and the most important one, and it's like can, um, you trust this person? They're not. They're not going to cause you any harm. Um, they are providing you good, trustworthy information. So this is the, the core thing that google essentially wants to find and provide pages that have. You know, all of these four key attributes.

Now they use the quality raters to rate a a set of pages and queries. Where it starts with queries. They've got, I think, 15 000 queries and they have the google raters review all the results that come up for those queries in terms of this EEAT and they're trying to obviously make sure that whenever they do new changes to their algorithm, they can test them against this data set that the quality raters have provided and to see whether their changes work within that context. And then, obviously, at that point point, google want to be able to have a model where they can see yeah, for any site they want to be able to judge its eeat algorithmically, because they can't go around the whole of the internet and judge it manually. With real humans, it's just an impossible task. So they're trying to take that data set that the Google quality raters have created and put it through a machine learning model and apply it to the rest of the internet.

Now, there's a lot of you know, in the past, however many years EEAT has been going on, there's been a lot of people going it's nonsense, they can't judge EEAT, it's not a thing, it's just propaganda. Eat, it's not a thing, it's just propaganda. Now, looking at this google leak that's happened recently and comparing it to the google rater guidelines, we've been able to see where there's crossover. So there's things in the google api leak that clearly relate to concepts in the google quality guidelines. So we'll go through this. We've actually done a full guide on this, a full written guide, and I will put in the show notes a link to how you can go and read the full guide We've got. It shows you all the attributes, it shows you all the quotes and images and we've pulled it all out into one thing. But this podcast is really just to discuss the same kind of core concepts. Okay, so let's look at some of the bits in the API docs compared to quotes in the Google search quality guidelines. Now, the first one we came across is where there's this quote from the guidelines.

For example, you may be able to tell someone's an expert in hairstyling by watching a video of them in action styling someone's hair and reading others comments. Commenters often highlight expertise or lack thereof. Now that's an instruction to others. Comments commenters often highlight expertise or lack thereof. Now that's an instruction to google quality raters to go and try and find these people on youtube videos who have created a website or a web page and see what people are saying about them. Now in the api docs, there are some modules all about youtube, but they don't go into full detail on all of YouTube's functionality. They only cover certain things, and one that we noticed is entity sentiment. Now, the entity that into is entities in the, we think in the google knowledge graph. So if you're a person or an organization in the google knowledge graph and it relates you to youtube videos, it will then look at the comments and look at the comment sentiment for any comments against your videos and then decide what the score of that content sentiment is whether it's good, bad, middling, whatever, which is an interesting concept. This means that possibly it's a way of google finding out what your level of expertise is based on people commenting on your youtube videos. So I know it's maybe niche not everybody creates their youtube videos and has contents comments on them like this but it appears to be something they're definitely trying to do programmatically.

Next, we looked at um, if we could find signals where they are looking at the quality of the main content on a page. Now google says in the document that the quality of the main content should be listed as the lowest. If a page is hacked, defaced or scammed sorry, spammed the page is gibberish or otherwise makes no sense. The main content is copied, auto-generated or otherwise created without adequate effort. The main content is created with subtle effort, originality, talent or skill that the page fails to achieve its purpose. So we looked at some of these. So the page is hacked, defaced or spammed. There is a spam Muppet signal and this contains hacked site signals which will be used in query time join. So this means basically, if there are hacked site signals, then they're going to be recorded and put in this attribute. So they're definitely looking for hacking there.

The page is gibberish or otherwise makes no sense. There is a gibberish score. Um, literally, a gibberish score is one of the attributes. So they obviously are qualifying content and deciding whether it's gibberish or not. The main content is copied, auto-generated or otherwise created without adequate effort. Now there's a couple here. There is one called copycat score, which is an integer score. So that's obviously where they're scoring content about whether it is copied from another source. And there's another one called content effort, which is an effort estimation for article pages. So therefore they're obviously at some point algorithmically looking at pages and saying what kind of effort has been put into this. So that's all interesting. So obviously they are at some point algorithmically working out all these things. So those eeat type signals are being scored against all documents. Now another interesting we found is they don't like page titles that are misleading, shocking or exaggerated. And there is an attribute. Well, one of the modules is actually just called bad title info, which is a score and a type. So they're definitely looking at titles and if they decide it's a bad title maybe it's too click-baity, something like that then they are definitely going to score it down to clickability, something like that. Then they are definitely going to score it down.

Next, we started to look to see if we could find anything around ads, because although the google search quality guidelines do say that it's fine for people to monetize pages with ads, it's not got to be intrusive and they deliberately say one of the things is the main content is deliberately obstructed or obscured due to ads, ads, supplementary content in interstitial pages, download links or other content that is beneficial to the website only but not necessary to the website visitor. So this is where you've gone too far with that. It's essentially the same. Now we found a whole number of places where they talk about um, these kind of things. So we've got desktop interstitials. So they're definitely recording when, on a desktop, you've got interstitial adverts coming in, they have a clutter score. So this is a comment directly from the API penalizing sites with a large number of distracting slash, annoying resources loaded by the page. So they've definitely, definitely got something there when it comes to having too much going on. They've got another one called add density interstitial violation strengths. So again, this is um and in their own comments, as they indicate, if the page is violating mobile ads, density, interstitial policy and the violation strength. So this is like saying how far you're violating whatever their policy is, how far past it you're going. And they've got another one violates mobile interstitial policy. So again, there's several modules on Promptly related to different types of interstitials and adverts, and this clutter score, which is an interesting one, penalizing sites with a large number of distracting and annoying resources. So definitely, definitely, there, at some point you can overdo it with ads, by the looks of it.

The next quote we've got france Google quality guidelines is the page is deliberately creative as such low quality main content that it fails to achieve any purpose. And in the compressed quality signals module there is an attribute, low quality, which I mean. What can it mean, apart from low quality of the, of the, of that quality signal and the final one I'm going to cover today from this quote of that quality signal. And the final one I'm gonna cover today from this quote Originality. Consider the extent to which the content offers unique, original content that is not available on other websites. If other websites have similar content, consider whether the page is the original source. Now they've got in one of the modules a original content score. Now this you know, clearly google measuring how original this content is compared to everything else out there. Um, so again, this is something they're clearly algorithmically trying to do, based on ee8.

One final thing I also want to add into the consideration here is that obviously, part of authority and trust we know comes from backlinks, page rank from backlinks and the link graph itself. So, although the things I've spoke about so far are probably machine learning based attributes, there's going to be other attributes into EEAT which are more likely to be around those link based metrics, because nothing says trust better than high quality backlinks. We've seen in other parts of the google api leak and how they attribute and the quality of links I spoke about in the podcast a couple of days ago and about whether it's a high quality link, a medium quality link or a low quality link, and you know, if you get lots of high quality links. That's obviously going to build into the trust and authoritative section as this as well. But it's interesting to see from these that google clearly attempting to measure eeat algorithmically based on the training data set that they get from their army of google quality raters. Um, now, the output from these algorithms, combined with the link graph and page rank you know, can quite easily be used as an algorithmic or algorithmicic approximation for EEAT. But will it be as perfect as a human evaluation? Well, possibly not, but it's an impossible task to human review every web page and website, whereas algorithmically these evaluations can be operated at scale to achieve the task and we can see how aspects can be related back from the Google API docs to those quality-rated guidelines. So I mean, there's probably more that can be done in research for this, but I think it's clearly giving more and more evidence to how Google are algorithmically implementing EAT.

Now I do suggest go to the show notes, find the link to the article that we've posted on Keyword People Use. That covers this in detail, because it covers everything I've just shown, talked to you about. It shows images from the API leaks where we've underlined exactly each element links to the API link, api documents so you can go and sort of give it your own thought, and there's plenty more in there, more than I've covered just talking about it for a few minutes. So please do go and look at that. Yeah, and until next time I'll see you later. Before I go, I just wanted to let you know that if you'd like a personal demo of our tools at Keywords People Use that, you can book a free, no obligation one-on-one video call with me where I I'll show you how we can help you level up your content by finding and answering the questions your audience actually have. You can also ask me any SEO questions you have. You just need to go to keywordspeopleusecom slash demo where you can pick a time and date that suits you for us to catch up. Once again, that's keywordspeopleusecom slash demo and you can also find that link in the show notes of today's episode. Hope to chat with you soon.

Thanks for being a listener. I really appreciate it. Please subscribe and share. It really helps. Seo is not that hard. It's brought to you by keywordspeopleusecom, the place to find and organize the questions people ask online. See why thousands of people use us every day. Try it today for free at keywordspeopleusecom To get an instant hit of more SEO tips tips. Then find the link to download a free copy of my 101 quick seo tips in the show notes of today's episode. If you want to get in touch, have any questions, I'd love to hear from you. I'm at channel 5 on twitter. You can email me at podcast at keywords people, usecom. Bye for now and see you in the next episode of seo is not that hard.