This article talks about month #2 of the SEO case study using an expired domain.
Month 2’s publishing was a bit slower than month 1. About 140 articles were added in total, or roughly half the amount from October. Part of it was writing more thorough articles, but most of it was simply lower output.
There are about 410 articles on the site over these two months.
November 2022 Stats
It generated 18,599 page views and 16,824 sessions. So a growth of a bit more than 3x from Month 1.
Bing likes the site, Google doesn’t (so far)
One interesting thing that was noticed is that the site is getting close to zero Google traffic.
In October it got some, but then collapsed down to zero clicks and almost no impressions, as shown in Google Search Console.
Even though the domain is built off an expired domain and should have some juice in it to rank fairly quickly, the content is very different from the original version of the website.
Therefore, Google probably recognized that and stopped giving it traffic and impressions.
After all, it probably doesn’t want to give people an edge simply because they have an expired domain.
So where is all the traffic coming from?
Bing, it turns out – and search engines built off Bing (e.g., Yahoo, DuckDuckGo).
I failed to realize this the first month. But I did have an inkling considering that the desktop/mobile traffic was highly skewed toward desktop. Mobile is most traffic these days.
Bing is mostly desktop because that’s what’s built into some browsers like Firefox, Vivaldi, and others, which are not commonly used on mobile.
Desktop traffic is more valuable anyhow, so there’s that.
Bing is a simplified version of Google. It’s always at least a few years behind where Google is for the most part.
So tactics that used to work on Google years ago (e.g., keyword stuffing to get higher rankings, aged domains to get an instant traffic boost) currently work on Bing.
Nonetheless, getting around 19,000 page views with only ~40 of those page views coming from Google is kind of exciting considering that whenever the site does “wake up” it should be getting a lot more traffic.
Bing, as of the time of writing, is only about 8% of the search market share, with Google still over 80%.
That doesn’t mean the traffic should be 10x what it currently is, but it’s a way to think about things.
One site I have gets around 200,000 page views per month and almost none of that comes from Bing.
So how different search engines treat a website is interesting.
So, still waiting for Google…
So even though this is an aged domain, it basically has to be treated like a fresh domain since the content being produced is so different from the original version of the site.
For that reason, it’s important to give 6-7 months to see how Google is treating it.
If it’s still mostly crickets on Google after that time, then it’s probably best to invest resources elsewhere.
Plan for this site
In a nutshell, it’s:
- Write content that ranks
- Get up to 50,000+ sessions per month to get onto the Mediavine ad network
Mediavine ads is the main monetization goal
I think the site can be a “success” by simply getting it up to over 50,000 sessions and getting it on Mediavine even if Google doesn’t ultimately treat the site well.
Based on where pageviews currently are – close to 1,000 per day – that’s about halfway to the 50K sessions mark (50K sessions is about 60K pageviews).
I see no reason why Google will permanently sandbox the site and not be a source of traffic, but you never know.
But if 50,000 sessions with 70% desktop and 60%+ US traffic, that should still be good for $25+ RPMs I would think, or $1,000-$2,000 per month if I decide to move on to other projects (I.e., if Google doesn’t come through).
Of course, the upside is much higher if it gets good traffic in Google.
So, it’s a little over 400 posts in the two months.
I’m writing all of it rather than outsourcing as I don’t want to operate at a loss for something I can’t be sure of.
Be careful with AI writers
Using AI writers is currently in vogue and is very likely to get even more popular through time.
However, it’s important to be careful with these.
Google will have ways of sniffing out AI content. They claim they already can (though it’s probably more of a scare tactic than a reality at this point).
While they probably won’t be able to determine if the words themselves (the paragraph text) are AI-generated, they will have ways of triangulating which sites excessively rely on AI content.
First, AI content just basically mimics what’s already out there.
It doesn’t produce much in the way of original thoughts.
It also gets lots of facts wrong. Sometimes they don’t even do basic math correctly in cases where you have numbers and calculations in your post.
They do many things well. The writing quality is generally good and they are very competent at writing about many topics.
But the articles that really do well on the web are generally those that have some sort of originality to them.
After all, why would Google want to rank an article high that just regurgitates information that it already has in abundance?
Another issue they have is repeating a lot of the same phrasing over and over again within the same document.
I don’t think Google will be able to tell with certainty what’s AI and what’s not in a definitive sense. If a human can’t discern, it’s SEO-optimized, and people get value from an article, then there’s no reason why it can’t rank well.
And like many things, offense is often ahead of defense. AI writing technology, such as going from GPT-3 to GPT-4, will get better than the defense mechanisms trying to detect it.
Still, it’s important to use it carefully and not blindly, and recognize where it’s strong and where it’s weak. For example, it would have been virtually impossible to write this article with AI, as no algorithm is programmed to speak on the particulars discussed within this article.