SpectreJoe's Autogenerated Content to 700$/month. [Self-coded Bot]

[Beklink]

I believe you can use NLP for context analysis and see if it matches ur keywords, etc. But I'm not really doing that. I just scrape and eliminate some sentences if they contain certain keywords. (Read more, covid, links, etc)

Nice journey spectrejoe, keep us updated

 

[spectrejoe]

Over 5k Posts live since I restarted this journey.

Getting a little over 100 Visitors per day. Unfortunately got declined in Ezoic but accepted in Adsense so will use that instead I guess.

 

[Yogesh1998]

Hey, been following your journey and it is really great!
Had one question though, did you train any model for paraphrasing or are you using the default model from huggingface with Parrot to paraphrase content? Also, if using default model, do you get unique content?

 

[spectrejoe]

Hey, been following your journey and it is really great!
Had one question though, did you train any model for paraphrasing or are you using the default model from huggingface with Parrot to paraphrase content? Also, if using default model, do you get unique content?

Using default model yes I didn't train it.

Yep, unique content according to copyscape. Sometimes it detects 1 match or 2 but it's like just a sentence and that's it. Most of the time though it shows 0 results.

 

[Yogesh1998]

Using default model yes I didn't train it.

Yep, unique content according to copyscape. Sometimes it detects 1 match or 2 but it's like just a sentence and that's it. Most of the time though it shows 0 results.

Thanks for your reply!
I tried using it today on an article and the paraphrased sentences were pretty much the same as original ones with several consecutive words exactly the same. I hardly doubt that would be unique but I believe it had to do with the niche I tested it on.

Imma give it a go for a few other niches. Will let you know if I get any better results in the coming days of testing.

 

[wheewilly]

Nice journey. Keep us updated

 

[Bambarbiakirgudu]

With What programming language Did You code Your bot and what libraries were beneficial to You?

 

[Zera Api]

@spectrejoe Do you think even if content is 90% unique according to Copyscape and readable , will it still be counted as SEO friendly for such a big blog

 

[spectrejoe]

@spectrejoe Do you think even if content is 90% unique according to Copyscape and readable , will it still be counted as SEO friendly for such a big blog

Doesn't matter, I am doing quantitty over quality, plus a lot of the keywords I'm targeting are low difficulty with no one targeting them which means that they will still rank anyways after a few months

 

[spectrejoe]

Traffic going up nicely, currently at 100+ users per day, predict we'll be at 1k+ per day within 2 months.

 

[Warmobve]

If your traffic is high quality, you can try Networks like A-ads because of your content quality.

 

[Bambarbiakirgudu]

@spectrejoe
What language did You code with and can You tell me the libraries beneficial to You? because got plans to build a simillar bot for myself.

 

[Saif14]

Following

 

[Bribebox]

@spectrejoe
What language did You code with and can You tell me the libraries beneficial to You? because got plans to build a simillar bot for myself.

If you really "got plans to build a similar bot" then move your ass and don't expect to be spoonfed.

 

[minad268]

@spectrejoe your journey is great. Can I ask how you format your post (heading, bold, italic, image,....) automatically? Thank you!

 

[Bambarbiakirgudu]

If you really "got plans to build a similar bot" then move your ass and don't expect to be spoonfed.

i don't expect to be spoonfed. i already code a python based content scraper. i asked to see if Yours got somethingg that i haven't think of. it basically got two options. either grabs rss feeds accordingly the keyword i search or clicks on the google results that are harvested and coppies the whole content. then offcourse i will spun the content within buying apis.

 

[spectrejoe]

Small update on this:
Ezoic denied the website, probably because it had a few spammy posts (old ones) along with a design mistake I made which looked quite low quality, well, it was to be expected anyways and honestly, the website wasn't even close to meeting their traffic requirements to begin with anyways.

Can only apply in 6 months, as such, setup Adsense and turned on "Auto Ads". Had 450 impressions yesterday and 1 click for 1€
Traffic is increasing daily woohoo, however, google is being slow at indexing all the posts, it's indexing about 400-700 per week only.

If your traffic is high quality, you can try Networks like A-ads because of your content quality.

Will have a look thanks!

@spectrejoe
What language did You code with and can You tell me the libraries beneficial to You? because got plans to build a simillar bot for myself.

Python, newspaper3k, beautifulsoup, Pegasus / Parrot and a few others that I made

@spectrejoe your journey is great. Can I ask how you format your post (heading, bold, italic, image,....) automatically? Thank you!

I grab the H2 separately and make sure to remove them from the text I extract, then I just do something like:

text = f"<h2>{firsth2}</h2> \n {firsth2text} \n <h2>{secondh2}</h2> \n {secondh2text}"
etc

 

[Obilosa]

s Are you using offshore hosting and domain? I also set up a wordpress site and pull products from amazon to affiliate, but it doesn't work at all.

 

[kingling07]

nice! had this idea a few years back but did nothing about it .... following tentatively .

 

[TheOverlord]

i don't expect to be spoonfed. i already code a python based content scraper. i asked to see if Yours got somethingg that i haven't think of. it basically got two options. either grabs rss feeds accordingly the keyword i search or clicks on the google results that are harvested and coppies the whole content. then offcourse i will spun the content within buying apis.

buying APIs is not really a cost effective way of spinning.. they get expensive super fast