AI Article Penalty Speed Run Case study 4000 posts

Have you managed to index the the posts yet?

No net yet. The posts have only been live a couple of days and I have scheduled out to the posts in WordPress to drip feed them. Will be monitoring it though. Rank tracker setup for all keywords so that will give an early indication as well (hopefully!)
 
No net yet. The posts have only been live a couple of days and I have scheduled out to the posts in WordPress to drip feed them. Will be monitoring it though. Rank tracker setup for all keywords so that will give an early indication as well (hopefully!)
I wasn't aware of the scheduling feature. I bought a domain 5 days ago. Now am getting keywords. So far I Have 700 keywords. I want to come up with 5k keywords then start scheduling WordPress posts. I will also do this on one of my domain with 10 DA.
 
I wasn't aware of the scheduling feature. I bought a domain 5 days ago. Now am getting keywords. So far I Have 700 keywords. I want to come up with 5k keywords then start scheduling WordPress posts. I will also do this on one of my domain with 10 DA.
Sounds a good plan. I used the WP Import plugin (no affiliate link) for scheduling. I can bulk upload all articles then set a scheduling rate in the plugin. Plugin costs $7 but worth it.
 
Sounds a good plan. I used the https://www.wpimport.com (no affiliate link) for scheduling. I can bulk upload all articles then set a scheduling rate in the plugin. Plugin costs $7 but worth it."
Hi, can I ask
"WP Import" and another plugin with same name "WP All Import", which is best for import thousand of article.
 
Hi, can I ask
"WP Import" and another plugin with same name "WP All Import", which is best for import thousand of article.
Not used WP All Import (think it's very good for programmatic SEO) but you just need a basic uploaded/scheduler which WP Import does fine. Mot tried with 100) articles but it managed fine with 70 :)
 
when you say u

that's interesting but a bit confusing. Can you please give an example.
Sorry, im not sure how else to explain it. I did post usage of low fruits to show how that is done.
I am not known for keyword research, you would be better off looking at other posts for more precise info on keyword research.
thanks again for your inspiration! I have added AI posts to one site and 70 to another. Plan is to keep adding to different sites and to keep improving the process. With my process I think I can output 70-100 AI posts per day, including the keyword research & clustering part.
I wish you luck, can I ask how your process of deciding the order of what to post is? I'm also posting at a slower rate than before right now and I wonder to myself if I should post higher traffic keywords first because they take longer to rank so need to be published earlier, or if I should post long tail first because id rank faster. Are you going broad or sort of filling out one category on each site at a time for relevance?
Thank you for updating the news. It reminds me of a new episode of the show. Very cool.
What plugin did you use to make Internal Links? I guess you did it automatically.
you are welcome, the plugin I just added for internal likes is "Inline Related Posts" as it was recommended here to me.
The thousands you see on my GSC screenshot are what the site had already, which is massive because it counts links to every page on my menu (4k links to my about me page doesn't really help but the GSC will count them ) Thats why I took the screen shot so I can do the math when the plugin links finally show up in it.
This is awesome, im preparing something like this but with several fresh sites in several niches that can pull many related stuff.
Im wondering, do you do internal linking in content, auto external links or just a plugin to add few random related articles below the content? if you are using a plugin, which?
On my old human posts and a single category that I was testing, its all done semi-manually with link whisper.
For all the other thousand posts its now done with the "Inline Related Posts" plugin. I wish you luck, if you learn anything interesting on your journey id love to hear it!
all aritcle indexed?
Yes, from day one
 
I'm also posting at a slower rate than before right now and I wonder to myself if I should post higher traffic keywords first because they take longer to rank so need to be published earlier, or if I should post long tail first because id rank faster. Are you going broad or sort of filling out one category on each site at a time for relevance?
I know you didn't ask me... but I post main/high traffic (potentially) articles first... the reason?
  • As you correctly said, they take longer to rank
  • I can use internal linking to boost them as it's easy to insert links into long-tail keyword articles
  • Once the main articles are done... you can add all you want but you can't add all you want without your main articles (a silly example... when a tree grows it starts with a trunk and then grows branches).
 
I wish you luck, can I ask how your process of deciding the order of what to post is? I'm also posting at a slower rate than before right now and I wonder to myself if I should post higher traffic keywords first because they take longer to rank so need to be published earlier, or if I should post long tail first because id rank faster. Are you going broad or sort of filling out one category on each site at a time for relevance?

I'm not generating a huge amount of keywords but my process is:

1. Run a few wildcard reports in LowFruits
2. Identify competitors in LowFruits and scrape (and add) their keywords
3. Cluster keyword in LowFruits, export report and go for everything except 0 volume or clusters where there are 0 "low fruits"

On the 2 sites I have posted to so far:
Site 1. Large amount of categories so only focussed on one (think Ford 150 on an automotive website). Was thinking this afternoon that with the outline I have generated, I can just swap out the keyword (Ford F150 -> Nissan Titan) and then rerun my Python script.
Site 2: This site was a lot more niche so I could only generate 70 keywords and I'm pretty much done. Much harder to produce AI content when you start niching down and getting technical as well I found.

For info, my Python script at the moment generates a list of article sections/H2s for a list of keywords in a CSV file (I have setup the script to show me the outline for each keyword then I can approve or regenerate) and then writes the content for each H2 and then adds and intro and summary. This usually gives me a 1200-1800 word article. The output file its HTML so I can just upload to WordPress using WP Import.
 
I know you didn't ask me... but I post main/high traffic (potentially) articles first... the reason?
  • As you correctly said, they take longer to rank
  • I can use internal linking to boost them as it's easy to insert links into long-tail keyword articles
  • Once the main articles are done... you can add all you want but you can't add all you want without your main articles (a silly example... when a tree grows it starts with a trunk and then grows branches).
No problem, hearing from anyone is good, we all learn little things from each other (or at least that's my hope when interacting besides having a little fun with things).
I understand your reasons, I just want to ask if you can answer the question about categories aswell since this slightly falls out of the tree example (I find it hard to imagine a tree growing one branch at a time lol, but In SEO this isn't nonsensical due to relevance, and interlinking).
Keep in mind I work with mid to large niche sizes only so each category by itself tend to be large and could be considered not related enough by google if you try building all at one time.

Like having a finance site and writing content on loans and content on stocks, vs running your own store. Sure they are all related at some deep level but all 3 of these things can be categorized with hundreds of posts by themself.
I personally find more success building out sections of a niche in whole at a time (however for the AI site I've just gone wild but I'm referring to my normal operations).

tldr: Do you go category at a time or build the hard articles (most reward in potential traffic) for the whole niche starting first?
I'm not generating a huge amount of keywords but my process is:

1. Run a few wildcard reports in LowFruits
2. Identify competitors in LowFruits and scrape (and add) their keywords
3. Cluster keyword in LowFruits, export report and go for everything except 0 volume or clusters where there are 0 "low fruits"

On the 2 sites I have posted to so far:
Site 1. Large amount of categories so only focussed on one (think Ford 150 on an automotive website). Was thinking this afternoon that with the outline I have generated, I can just swap out the keyword (Ford F150 -> Nissan Titan) and then rerun my Python script.
Site 2: This site was a lot more niche so I could only generate 70 keywords and I'm pretty much done. Much harder to produce AI content when you start niching down and getting technical as well I found.

For info, my Python script at the moment generates a list of article sections/H2s for a list of keywords in a CSV file (I have setup the script to show me the outline for each keyword then I can approve or regenerate) and then writes the content for each H2 and then adds and intro and summary. This usually gives me a 1200-1800 word article. The output file its HTML so I can just upload to WordPress using WP Import.

This sounds great, also nice on your python script, the increased article length compared to what I'm producing probably will help you out in the long run too. (I do think many of my articles are held back by being low if its possible for you to get traffic data from elsewhere before putting keywords in low fruits it is better. You also wont waste money analyzing keywords that dont have traffic as you would know that even if low fruits is telling you it has 0, the better tool tells you it wasn't 0.
However, if you just want speed and dont care about the possible keywords falling through the cracks then it probably doesn't matter that much.
Wishing you some good results.
 
Sorry for hijacking your thread @Jaordas but I've got another idea I would like to share with you and get your thoughts.

There is a risk that some of my websites could appear to be "AI" sites because they are obviously testing sites. But what if I made them blend in with the serps? A lot of the competition sites are Income School/Acabado theme, so if I make my site look like an "Income School" site then any manually reviewer will (may) see it and pass it off as a normal blog.

A manual reviewer MUST have seen a ton of Acabado/Income School websites and would (hopefully) assume that these are 100% white hat. So why not blend in with them :)

On another note, I've been through the M0t1on 1nv3st website listings and found a lot of sites that are on Mediavine etc that are clearly using AI content. This is just by reading a few articles, running through Originality gave 100% AI for 1 article. This website was generating $600 per month and up for sale for $17k. Traffic had dropped off massively during 2022 (like a lot of niche sites) but was steady over the last few months.
 
No problem, hearing from anyone is good, we all learn little things from each other (or at least that's my hope when interacting besides having a little fun with things).
I understand your reasons, I just want to ask if you can answer the question about categories aswell since this slightly falls out of the tree example (I find it hard to imagine a tree growing one branch at a time lol, but In SEO this isn't nonsensical due to relevance, and interlinking).
Keep in mind I work with mid to large niche sizes only so each category by itself tend to be large and could be considered not related enough by google if you try building all at one time.

Like having a finance site and writing content on loans and content on stocks, vs running your own store. Sure they are all related at some deep level but all 3 of these things can be categorized with hundreds of posts by themself.
I personally find more success building out sections of a niche in whole at a time (however for the AI site I've just gone wild but I'm referring to my normal operations).

tldr: Do you go category at a time or build the hard articles (most reward in potential traffic) for the whole niche starting first?
I work in a niche which is high competition... well, my site is old so I have a chance but I don't have their money as I (like you) neglected the site for years.
It's all about leaning from each other.
I'm in the finance niche and stupid me learned something new... use exact match keyword links on your site.
You can PM me... but I'll give you an example... let's say you run a finance site and you talk about investing in shares/stocks... In the UK you can do it through ISA or invest directly BUT... google don't understand the difference so if you link to ISA shares you use ISA in you link google will understand it but if you use shares... google will struggle to understand if you mean buying shares or want to use ISA for shares.
That's why I say linkwhisper is crap and if you want to do proper internal linking... do it manually.
 
I work in a niche which is high competition... well, my site is old so I have a chance but I don't have their money as I (like you) neglected the site for years.
It's all about leaning from each other.
I'm in the finance niche and stupid me learned something new... use exact match keyword links on your site.
You can PM me... but I'll give you an example... let's say you run a finance site and you talk about investing in shares/stocks... In the UK you can do it through ISA or invest directly BUT... google don't understand the difference so if you link to ISA shares you use ISA in you link google will understand it but if you use shares... google will struggle to understand if you mean buying shares or want to use ISA for shares.
That's why I say linkwhisper is crap and if you want to do proper internal linking... do it manually.
makes sense, link-whisper isnt the best by any means but I just use it as a tool to see across the website when I'm building the links. Im sure something cheaper probably exists.
I get where you are coming from though, sadly in the finance niche the site age probably only helps you fight the other ai sites, not the competition that was already there.
Words like shares and the context def would probably trip up google, maybe only words with clear singular meaning / intent "dropshipping" are viable for auto-linking. Since I don't think words like those have much alternative meanings.
In some ways, these last 2 weeks I've been regretting neglecting this site more and more, but on the other hand, if I didn't neglect it then id be unwilling to just drop AI content on it.
I've got another idea I would like to share with you and get your thoughts.

There is a risk that some of my websites could appear to be "AI" sites because they are obviously testing sites. But what if I made them blend in with the serps? A lot of the competition sites are Income School/Acabado theme, so if I make my site look like an "Income School" site then any manually reviewer will (may) see it and pass it off as a normal blog.

A manual reviewer MUST have seen a ton of Acabado/Income School websites and would (hopefully) assume that these are 100% white hat. So why not blend in with them :)

On another note, I've been through the M0t1on 1nv3st website listings and found a lot of sites that are on Mediavine etc that are clearly using AI content. This is just by reading a few articles, running through Originality gave 100% AI for 1 article. This website was generating $600 per month and up for sale for $17k. Traffic had dropped off massively during 2022 (like a lot of niche sites) but was steady over the last few months.

I never really thought about this since in my eyes you shouldn't be able to tell an AI site just by looks. Any site I call and AI site is because I've tested the content like you did. People can say what they want about originality but I find it gives the best detection ability for this.
I am not too familiar with manually reviewers mentality but in my mind I don't think its likely they remember and care so much about specific theme styles, thousands of themes exist so I assume they have seen thousands. Ive seen a lot of websites, I'm familiar with what income school is, but I couldn't tell you if I was looking at Acabado or just some other blog unless the person was practically doing copy paste.
If your competition is income school users then id assume your bigger issues would be if they are writing at a the quality they are supposed to (which most don't but this is just in theory), it will be harder to out rank them based the content. Not the design. You do want to keep some sort of EEAT on the site though, I think that will matter more if a manual reviewer comes than the website looking like a project 24 carbon copy.
Income school teaches you not to do black hat but I don't think their influence is so big to have manual reviewers thinking a site must be white hat when they see one that looks that way.
Also, I'm not sure if they even would use an AI content checker in any way, and AI writes decent now.
I have all my human-written articles displayed on my front page etc, so if anyone comes to my site and samples a random page from the front, or about us etc, its one that is human-written.
On the bright side, your mediavine note is probably positive in that you can monetize with them. Well, since at the end of the day we are doing these things to make money as an end end goal.
Ive seen an AI site on even Adthrive, now its debatable to me if they ad companies are aware and choose to leave them alone, or not.

So much sites fell down last year.
Im still scratching my head trying to figure out how the heck to get google to give my site back snippets. As months ago when I was neglecting the site it lost ALL snippets across all keywords even if it ranks number 1 for the keyword it wont generate a snippet. That is part of why my traffic gains might be slower, my site right now is not getting any snippets, and I've already checked the basic stuff like making sure I didn't unallow snippets on my site. If you have any idea how to fix this or what is going on with this please help a brother out. All I'm finding online is people who have the same issue but also have no idea how to fix it. My main site never had this issue so its new to me.
 
makes sense, link-whisper isnt the best by any means but I just use it as a tool to see across the website when I'm building the links. Im sure something cheaper probably exists.
I get where you are coming from though, sadly in the finance niche the site age probably only helps you fight the other ai sites, not the competition that was already there.
Words like shares and the context def would probably trip up google, maybe only words with clear singular meaning / intent "dropshipping" are viable for auto-linking. Since I don't think words like those have much alternative meanings.
In some ways, these last 2 weeks I've been regretting neglecting this site more and more, but on the other hand, if I didn't neglect it then id be unwilling to just drop AI content on it.
When it comes to age, you are absolutely right... it helps to compete with new sites but it's hard to catch up with old sites which didn't stop working and have big budgets now.
I think you have 7 re-written articles so it's easy to get internal links to them... I use google itself for interlinking. Let's say my site ranks #15 for "ISA shares" so I enter in google site:mysite.com ISA shares and google will give me pages on my site which mention ISA shares so I go to those pages and add internal links. 2 benefits: 1. I add internal links 2. I update the pages.
 
When it comes to age, you are absolutely right... it helps to compete with new sites but it's hard to catch up with old sites which didn't stop working and have big budgets now.
I think you have 7 re-written articles so it's easy to get internal links to them... I use google itself for interlinking. Let's say my site ranks #15 for "ISA shares" so I enter in google site:mysite.com ISA shares and google will give me pages on my site which mention ISA shares so I go to those pages and add internal links. 2 benefits: 1. I add internal links 2. I update the pages.
Nice, that way you can be more selective, just means you need to do it article by article once you google them.



Updates (gonna try to keep this short):

The competitor AI site now only has homepage indexed....... still no idea what going on with their site or why they got hit off the serp. The other AI site is still there but both of them were on new domains.

I practically finished publishing articles for the niche, but my domain name is brand like so I am just branching out into another niche where ill set articles to be posted over the year just to keep the site "active" (hence the drafts)
1677960718098.png

Google has indexed every single published article with no issue.

1677960907963.png

1 week comparison:
1677961021150.png
2 week comparison:
1677961118070.png
3 week comparison:
1677961219937.png


(only using data from the first 4 k AI articles since I dont have data for the newly published stuff)
articles rank by their main keyword
Last week I had
Number 1: 28
Top 10: 1499
Top 20: 2628

This week I have

Number 1:20
Top 10:1212
Top 20: 2448

So across all the first AI articles my ranks over all dropped compared to when they were just posted. May just be google bouncing but it is what it is.

86 articles have completely fallen out of the top 100 results, I checked and the few I checked are still indexed, just not ranking. Some have fallen from high, some low.

None of the articles ive touched by hand have fallen hard

1677962358786.png


3 more articles have been changed to human written seo optimized

Is it possible to change the title from
"AI Article Penalty Speed Run Case study 4000 posts" to update the post number?
 
@Jaordas - thanks again for your inspiration! I have added AI posts to one site and 70 to another. Plan is to keep adding to different sites and to keep improving the process. With my process I think I can output 70-100 AI posts per day, including the keyword research & clustering part.
Can you explain how you cluster the keywords in your script?
 
Not being able to get featured snippets on your website it really depressing, I'm reaching near traffic level before my site got site wide traffic ban but its so painful that I'm ranking "first" on so much of my main keywords but because the enter site got snippet ban (no idea why cus it was a fully white hat site at the time it got banned and never got them back).

I cant help but hope to get snippets back because it made up a chunk of my traffic. I feel like many of the articles need to work much harder just to get their traffic meanwhile sites that literally don't answer search intent are getting the clicks just cus they have the snippet.

I keep searching for answers on how to get my snippets back but I'm finding nothing. I see a lot of people got theirs back last year a while after the update that killed mine but mine never returned.

Now I'm using the site for this case study I just really wish they would come back.

Sucks, cus its like no matter what I'm doing, the site is just going to be half the size its supposed to be.
1678053862778.png
Ive probably asked this a few times before but if anyone has any info on how to get my site to have snippets again please let me know.
 
My main money site got the snipped ban, I can't think why, and not sure how to get them back. I am sure you will pick up more traffic as those articles start ranking, some can take up to a year.
 
Back
Top
AdBlock Detected

We get it, advertisements are annoying!

Sure, ad-blocking software does a great job at blocking ads, but it also blocks useful features and essential functions on BlackHatWorld and other forums. These functions are unrelated to ads, such as internal links and images. For the best site experience please disable your AdBlocker.

I've Disabled AdBlock