[Journey] 1 million UVs/month in 12 months using AI generated content. Let's do it!

[prvncqr]

does your cloaker have a bot detector? My friend just got banned for getting bots on Adthrive by accident. I think somebody did him.

If your site authority is good and you have an excellent Google analytics report, you may show them to adthrive. They will consider your request legit and reactivate your account. Maybe they will debit the money you made from the bot traffic before they reactivate your account.

 

[Sartre]

If your site authority is good and you have an excellent Google analytics report, you may show them to adthrive. They will consider your request legit and reactivate your account. Maybe they will debit the money you made from the bot traffic before they reactivate your account.

thanks I will pass this on

 

[jimbo-jump]

Very interesting journey. Thanks for sharing. Are you building any PBNs to help these sites rank quicker?

 

[wizzardz01]

unrelated but has anyone experimented with manipulating G's rankings by faking CTR in SERPs?

It's not that easy, you need to have a deal with telecom to provide you with enough residental proxy, you need to check in kw planner what is percentage of mobile and desktop search for your queires and to have extactly the same while you doing manipulation.
Also you have to be in top10 at least, its ideal if you are in top3 and then doing CTR manipulation to goes on page 1. But when you start and you achieve #1 position you need to keep doing the same all the time, in 90% cases when you stop you lose your rankings. It works and it costs. It can be watched as PPC on some way. But it's all about good infrastucture and proxy servers.

 

[LuckyGamble]

does your cloaker have a bot detector? My friend just got banned for getting bots on Adthrive by accident. I think somebody did him.

I can't talk about my cloaker outside the marketplace due to that I would break the BHW Rules :/

But, yeah when you block bots and only leave real users on your site, the ad networks won't ban you and might even get you a nice bonus on earnings

 

[adonthenet]

Anyone scraped enough websites using newspaper 3K to see and realize that sometimes even if you add some of your own cleanups to the HTML you may still end up with content coming from the MENU or FOOTER or Author BIO etc… because each person can write the code a little different on a site - you may always end up getting stuck with non-relevant and bad content in the output texts…. Anyone familiar with what I’m saying and managed to find a solution or just running manually reading outputs and making sure you didn’t got by accident all those unrelated texts?

 

[Shaybib]

Anyone scraped enough websites using newspaper 3K to see and realize that sometimes even if you add some of your own cleanups to the HTML you may still end up with content coming from the MENU or FOOTER or Author BIO etc… because each person can write the code a little different on a site - you may always end up getting stuck with non-relevant and bad content in the output texts…. Anyone familiar with what I’m saying and managed to find a solution or just running manually reading outputs and making sure you didn’t got by accident all those unrelated texts?

If you scrape only h2 and the text after you are quite safe from menu and footer. Other than that you add a blacklist of words like "bio" and "covid" and dont add text containing them.
Also i dont know why people use newspaper instead of beautiful soup

 

[Niffo]

Anyone scraped enough websites using newspaper 3K to see and realize that sometimes even if you add some of your own cleanups to the HTML you may still end up with content coming from the MENU or FOOTER or Author BIO etc… because each person can write the code a little different on a site - you may always end up getting stuck with non-relevant and bad content in the output texts…. Anyone familiar with what I’m saying and managed to find a solution or just running manually reading outputs and making sure you didn’t got by accident all those unrelated texts?

I have my own bs4 function with a list of filters. Works on most websites. Only weird sites with only divs (so no p,h2's,etc) won't be scrapes.

 

[Sartre]

Very interesting journey. Thanks for sharing. Are you building any PBNs to help these sites rank quicker?

I haven't used PBNs for a few years now but I know others do. I use aged domains with high DR for half of my sites and getting organic backlinks later.

It's not that easy, you need to have a deal with telecom to provide you with enough residental proxy, you need to check in kw planner what is percentage of mobile and desktop search for your queires and to have extactly the same while you doing manipulation.
Also you have to be in top10 at least, its ideal if you are in top3 and then doing CTR manipulation to goes on page 1. But when you start and you achieve #1 position you need to keep doing the same all the time, in 90% cases when you stop you lose your rankings. It works and it costs. It can be watched as PPC on some way. But it's all about good infrastucture and proxy servers.

I'm not scared of difficult setups, but have you personally tried it and can confirm it works?

Anyone scraped enough websites using newspaper 3K to see and realize that sometimes even if you add some of your own cleanups to the HTML you may still end up with content coming from the MENU or FOOTER or Author BIO etc… because each person can write the code a little different on a site - you may always end up getting stuck with non-relevant and bad content in the output texts…. Anyone familiar with what I’m saying and managed to find a solution or just running manually reading outputs and making sure you didn’t got by accident all those unrelated texts?

Maybe readability will work better for you? Give it a try with the sites that give you problems: https://github.com/buriy/python-readability

 

[Archemike]

I got over 100 WP sites on a $20 Linode droplet + cloudflare + wprocket

relevancy of content on page 1 + user generated content

nah it took me 1 evening to write a better algo than Ahref's KD

Our own model now.

Stop words, honestly I don't remem right now, The app is so huge. Readability - we discard snippets that are over 12 graded readability.

requests/ SERPs API

Nah i don't believe this happens.

google serp api, or a custom one that you pay per request and they do the crawling?

 

[wizzardz01]

I haven't used PBNs for a few years now but I know others do. I use aged domains with high DR for half of my sites and getting organic backlinks later.

I'm not scared of difficult setups, but have you personally tried it and can confirm it works?

Maybe readability will work better for you? Give it a try with the sites that give you problems: https://github.com/buriy/python-readability

It's not difficult setup, but enough residental proxy servers could be. And as I said when you start doing it, and achieve your position, you must keep going all time if you want to keep that position. I didn't work on it personally but have good friend of mine that work on service like this, and charging good amount of money to his clinets on it. It's just like paid search. But you have fixed price.

 

[hehe2444]

It's not difficult setup, but enough residental proxy servers could be. And as I said when you start doing it, and achieve your position, you must keep going all time if you want to keep that position. I didn't work on it personally but have good friend of mine that work on service like this, and charging good amount of money to his clinets on it. It's just like paid search. But you have fixed price

Do you need to rotate proxy after each search/click on your site from google?

 

[Sartre]

google serp api, or a custom one that you pay per request and they do the crawling?

I use both.

 

[hehe2444]

@Sartre any of your sites got deindexed recently?

 

[prvncqr]

Only the biggest sites are affected.

@Sartre any of your sites got deindexed recently?

 

[Sartre]

@Sartre any of your sites got deindexed recently?

nope, as my friend likes to say "sites only go up"

 

[Sartre]

I can officially say I consider this journey 100% successful now. This is my biggest PAA site:



The funny thing is that the last update everyone is crying about didn't seem to affect PAA sites.

Our app is multithreaded now with full proxy support and I'm in the process of generating this stuff on around 150 domains with my partner this month. We invested in a bunch of strong/aged/high DR domains.

I will still update this thread with new findings and ideas whenever I get the time. Cheers!

 

[prvncqr]

I can officially say I consider this journey 100% successful now. This is my biggest PAA site:

View attachment 213452

The funny thing is that the last update everyone is crying about didn't seem to affect PAA sites.

Our app is multithreaded now with full proxy support and I'm in the process of generating this stuff on around 150 domains with my partner this month. We invested in a bunch of strong/aged/high DR domains.

I will still update this thread with new findings and ideas whenever I get the time. Cheers!

Great. I have one final question. I think you monetize the site with Google. How much CPM do you get per 1k visitors?

 

[b00bz]

I can officially say I consider this journey 100% successful now. This is my biggest PAA site:

View attachment 213452

The funny thing is that the last update everyone is crying about didn't seem to affect PAA sites.

Our app is multithreaded now with full proxy support and I'm in the process of generating this stuff on around 150 domains with my partner this month. We invested in a bunch of strong/aged/high DR domains.

I will still update this thread with new findings and ideas whenever I get the time. Cheers!

Impressive results, Ive created few of PAA sites myself, although on fresh domains, since I dont have the funds for aged/high dr ones. Hopefully, I can get asmall slice of the pie

 

[spectrejoe]

I can officially say I consider this journey 100% successful now. This is my biggest PAA site:

View attachment 213452

The funny thing is that the last update everyone is crying about didn't seem to affect PAA sites.

Our app is multithreaded now with full proxy support and I'm in the process of generating this stuff on around 150 domains with my partner this month. We invested in a bunch of strong/aged/high DR domains.

I will still update this thread with new findings and ideas whenever I get the time. Cheers!

That's awesome! How old is this domain and how many articles?