AI Content vs Human Content: What's Better for SEO?

00:00

Welcome to episode two of the AI Search Report. Today we'll answer, does AI written content perform as well as human written content? How many clicks go to the top ranking results on Google? Should you delete your entire blog just to improve SEO? And does chat GBT Traffic Convert better than Google? I'm your host, Nathan Gooch, and this episode is brought to you by Rank Ability, an AI SEO software that helps you rank everywhere customers are searching. Let's get right into it.

00:47

So just a couple weeks ago, Graphite released this study on AI content in search and LLMs. And essentially what they found is that the AI written content is roughly, I think 31% of it shows up in the traditional search results and 33% shows up in the chat GT responses as well. But I think they found roughly that 67% of the sites were human written. It's about twice as much that were human written. So essentially the headline would be human written content is performing better than AI written content. If we just take this as is the true narrative, that is what the study found.

Now, I do want to add some nuance here. They did find have the same finding as well, which is like most of the pages that are performing well on Google are human written and not AI written. Now, I showed this maybe a couple of weeks ago about AI content detectors, and they're very easy to manipulate. Okay, so I'll just give you an example. So a lot of the content that we're producing right now is AI written content, and it is human inputs. Then we use AI to create that content at scale, and then we edit the content after it's created and it gives it the human touch, basically.

So we put it through one of these AI checkers like Originality or uh, it's called Copyleaks, and it came back a hundred percent human written. I don't think that's true. I think these detectors are just very easy to manipulate. You can humanize it, you can do certain things to essentially manipulate the algorithm. So just to be clear that this isn't necessarily the case and that clearly these pages that are ranking do have quite a bit of AI assistance in them because that doesn't, that's just a fact. I mean, most SEOs are probably using AI in some capacity.

01:46

But I think the broader point here is that if you're going to go straight human written content, you can certainly rank quite well, but as long as you have all the other variables in place. But it's also dangerous to make these conclusions in general because a lot of the pages that perform well in Google also tend to be very strong pages from strong websites, from a link profile perspective, strong third party signals. So there's a lot of reasons why those pages are ranking well. Obviously the human written content plays a role, but I don't know if we could say definitively that human written content is the main driver of these rankings. So it'd be really hard to make that conclusion.

I think in general though, you probably could say that the pages that were fully written by AI, they are probably not performing very well. Okay, and there's a lot of reasons for that. Usually people doing that are just trying to put a ton of content out and they don't really edit it or add any sort of value to it. There's going to be patterns like keyword stuffing that you see, there's going to be very low value. So there's a lot of variables why those pages are not performing well.

02:39

And as they say in writing, the magic is in the edit anyway, so some pretty good findings here, but once again, Google and the AI platforms don't care whether AI was used or not. It's more about the usefulness and the value of that page. And if you can do that via ai, then you should, they'd be insane not to use this technology.

03:29

Okay, this next one is actually from Reddit. Saw this thread pop up, and this person was asking if they should delete 700 of their articles to improve their SEO. And they said they've been running the site since 2020. They now have around 900 articles. Somebody told me to remove those articles that have had zero clicks and are not ranking on Google search as they make almost no contribution. There needs to be some nuance to this. This is an extreme action to take, but without proper nuance in systems, that's going to cause some problems.

So in general, the idea behind pruning a website is actually very valid, but you have to have proper process in place. So for example, you have to conduct an SEO audit. I would prefer to do an SEO content audit. And what you want to do is export, say, the last 16 months of traffic from Google Search Console and also Google Analytics. You want to connect your hs, like a tool like Ahrefs or Semrush, so you can actually evaluate the link profile as well.

04:19

So what we want to do is when we have that data, you're going to be able to see a few things in Google Search Console. We want to see if those pages have impressions and clicks. Then in Google Analytics, we want to see if it has traffic, and then with the HSAP, I want to see if it has backlinks, okay? Now, if you find a page that doesn't have any of those things, it has no impressions, has no clicks, obviously, if it has no impressions, has no traffic from Google Analytics, and it has no backlinks, I am going to go ahead and delete that page. We're going to 4 0 4 it. We're not going to 3 0 1 that page anywhere because it has no value, doesn't contribute anything to your site.

So, but let's say a page doesn't have impressions and it doesn't have clicks and it doesn't have traffic, but it does have backlinks. That's a good page that you may want to 3 0 1 to a relevant asset. Okay? Now there's different reasons why you might keep a page like this. And I want to run through each of them just quickly.

05:10

So if it has no lagging signals but does have some good backlinks, we will 3 0 1 redirect to a relevant page. Another thing is sometimes a page just got stale or outdated, so maybe it's a good idea to upgrade and optimize that asset and bring it back to life. There's some situations where that makes sense. And then another one would just be consolidation. So sometimes you might have a group of pages that maybe are outdated in stale and you want to bring those together to a single master page. And so those types of pages should be probably 3 0 1 redirected to that master page that you just created.

The other one would just be if a page is no longer relevant at all and it's just, well, it's just not serving users. One example is that we had a client that would have like pages that were about events. So obviously they did well at the time, but they're no longer relevant pages because they're in the past. So we redirected those or 4 0 4 those pages.

05:59

But if it's we're talking pages that are multiple years old and they don't have any positive metrics, something needs to happen to those pages, just letting them just rot on your website is not a good thing in any way. So definitely want to attack those.

Okay, next one is that he asked or whoever asked, if I remove all of those articles and redirect them, would it lower my SEO ranking score? Well, number one, there is no such thing as an SEO ranking score, so that's not a thing. So getting rid of a page is not going to lower your SEO ranking score per se. And there's no other metric that they could be referring to that would work that way. And certainly not some third party metric like domain authority or something like that. But clearly there's some confusion. Let's just assume that they're talking about domain authority. It doesn't really matter either way because again, like I said, you're not going to get penalized by Google or lose rankings by deleting pages that have no value.

07:03

So there would be literally no downside to getting rid of them in this case. But once again, assuming you went through that process I just described, this is the next question. If I redirect them to the homepage, will it cause redirect chains? And is it okay to have so many redirects pointing to the homepage?

Okay, so this is a common mistake. When you start to redirect, you should only redirect pages if they have either some KPIs, some positive signals, or they have backlinks. If you're going to delete a page, it doesn't have any backlinks, you should just 4 0 4 it. Do not redirect it to your homepage because the homepage is optimized for your brand's keywords, like Rank ability or Nike or Ford or whatever your brand is. It's optimized for that theme.

07:49

Don't just redirect to the homepage. You want to find a relevant alternative. You want to try to find a page that's relevant to that original page, and then 3 0 1 redirect. And if you don't have a page that's relevant, and then you've got a bunch of pages you want to 3 0 1 that have these positive signals, you may have to go and create that page and then do the 3 0 1 redirect or multiple redirects. But in general, redirecting to the homepage is generally not going to be effective. Okay, so once again, you should always do relevance matching when doing redirects because you want to make sure you do this right.

So for example, say you have a bunch of pages, say on your website, you have 50 pages all optimized for blue Nike shoes. And then say you had your homepage, which is about just Nike in general. Now all of a sudden you're sending all these irrelevant signals to the homepage. So you'd be really, really careful. What you want to try to find is the apples to apples page. So you have your page about blue shoes, maybe you're going to redirect it to your page about blue baseball cleats. So at least there's some level of relevance, not a complete mismatch. So just be really careful with that.

09:00

The last one here, is there a possibility of losing the DA of the website? Okay? Once again, DA is domain authority, which is a Moz metric. Moz is just a third party tool provider. No different than hfs or SEMrush or even rank ability, just third party tools. They're just third party metrics. It's not necessarily what Google uses in its algorithms, but more importantly, based on this, this is where there's obviously a clear disconnect between how these things work.

What you do on your website does not influence what happens off of your website as far as your link profile, right? If you delete a page or you redirect, it doesn't actually influence your link profile. Your link profile is like, you could think of it as assets. You could think of it as a bank account, okay? So if someone linked to a page that you deleted or you redirected or whatever, that link still exists. It just simply pointed to a page that was either deleted or redirected. Your link profile stays the same.

10:06

You just want to be careful that you don't have 4 0 4 pages that have backlinks, because otherwise you're going to have to go through 4 0 4 link reclamation and you're going to want to recover that link equity. But in general, what you do with your content on your site is not going to influence your existing link profile.

And then hres came out with a study saying that roughly 97% of clicks happen in the top 10 search results. It's always good to get confirming data. I don't really think anyone here would be surprised by that. Obviously this is probably varying a lot by commercial intent as well. So commercial based queries probably have even higher click through rate. Like something like a question might have lower. But in general, yeah, the top 10 results, specifically the number one, two, three spots get the most clicks.

11:09

But based on when we look at the sources of retrieval, when search is used in chat GBT, usually it's pulling from sometime it may only pull the top ranking result in some results, but sometimes it can go as far back as page three. We've seen anywhere between one to 30 as far as those sources of retrieval. And also when it uses the retrieval, sometimes, like I said, it'll pull the number one results. Sometimes it'll pull the top five other times, it may just, for whatever reason, grab number 23. Like it, it varies a lot in chat, GBT specifically.

11:56

So in general, you should be trying to improve your traditional search rankings because it influences basically everything influences obviously, your ability to get organic search traffic. It also increases your ability to become a source of retrieval in chat, GBT and other LLMs. So it's really, really important to improve your performance. Kind of goes without saying.

And then another study just came out talking about chat GBT versus Google as far as conversion rates for SaaS products. And basically what they found is that those who engage with chat GBT convert 4.2 times higher than those from Google organic search. So chat GT people are converting 4.2 times higher. Well that makes sense because it's a much smaller population than Google. It's a higher intent population or seems to be. But that's how that works, right? Niche audiences like LinkedIn traffic or even compared to chat GT usually do convert at higher rates.

12:56

That's pretty obvious. If you go into Google Analytics for, you can filter your traffic and you can see typically chat GBT is usually 80% of the AI search AI traffic referral traffic. But when you look at the traffic from actually AI traffic in general, it accounts for very, very little of the total traffic that goes to most websites. In fact, it's according to them, 200 times smaller than Google's organic traffic. Now, I've seen this in many of the accounts that we have access to, even our own, even our account, it's very small.

So if you see 4.2 times higher conversion rate, but it's 200 times smaller than Google, I wouldn't be comparing chat GT directly to Google organic traffic. That's like comparing Google Organic search traffic to traffic you get from LinkedIn. You can't really compare those things because very different. And when it comes to chat GT and LLMs, it's more of a branding initiative. It's more of a full funnel process that if you are visible in these platforms for commercial queries, you're much more likely to get branded searches in Google.

14:53

So it's not exactly a direct thing there. We should be tracking referral traffic, but what we should really be focusing on is the KPI, which is the visibility inside of these platforms. So you're trying to show up for as many commercial queries for your brand.

And of course, the reason why no one wants to track that is because it's hard to correlate, okay, you've got a brand recommendation in chat GBT, how did that lead to a conversion? It's very hard with attribution, but like I said, it's a much better comparison against LinkedIn or social media because that's kind of the case there too, right? You're building brand on social media and by doing that, you're going to drive branded searches in Google.

And there's another thing which is like someone maybe engaged with your brand via chat GBT, and then you can kind of assume, okay, they go to Google and search for your brand. But the original, the real attribution probably should have come from chat GBT, but that will never be the case because someone's going to Google and conducting that branded search.

15:49

So this is not so simple of a concept and people like to throw these numbers around, but just be careful not to make quick conclusions on this. But it's also, don't focus on the wrong KPI. You should be tracking the referral traffic from these platforms, but that's not the primary metric for you to be tracking. It's the visibility inside of these platforms because that's much more of an intent to what's actually going on in the background. Just like we would track our visibility inside of Google as far as what we rank for, what we don't rank for. It's the same exact thing.

So hopefully this was helpful. Thanks so much for tuning in. We'll see you in the next episode.