3 ms·

Fooling Google is a different game than catering to arrow clickers. This is really more of a data problem. I think the strategy is to take the most popular th

Fooling Google is a different game than catering to arrow clickers. This is really more of a data problem. I think the strategy is to take the most popular things around this time of day, day of week, and month of year from say a pool of 5-10 year old data. Then you filter it looking for evergreen content. This would be an incremental heuristic model. You take the candidate content, use some trivial grammar model to rewrite it so it's not completely identical and then automatically put it on different styled templated sites. Register a few domains that suggest they are tech information sites Make sure your Twitter cards and url naming system is reasonable, then have a bot to post it from a handful of puppet accounts Have a much larger pool of about 250 or so puppet accounts and use a different 5% subset to seed them with upvotes over Tor in order to confuse any clustering algorithm and then keep the profits. Affiliated ad revenue is total complete shit though. Even if this all worked beautifully you're looking at what? $20/day? I mean who cares... The truly dishonest and devious thing you could do is freemium it and have a generous paywall like NYT. That's an incredibly dishonest way to make money though. It'd probably work great.

dmix 8y ago
1) You'd need 100k+ visitors to make any money off Google (1+ thousand a month). 2) Just having visitors is not enough, they have to be "niche" to an ad category and ideally looking for something to buy (not just information). The only thing that pays out is high-quality PPC traffiic or even better lead-gen for high-value products (health, debt, education, etc) with conversion-based pricing. 3) Google ads on content sites with CPM (impressions) pays pennies out. 4) Targeted blogspam sounds like it involves lots of humans beyond just mechanical turk, with some text parsing and rewriting which is a big overhead 5) This is basically the business model of Buzzfeed and other cancerous sites plus countless "niche" sites you've never heard of with high quality domains + foreign content mills + partner networks+cross domain advertising systems to convert in bulk. If you want my personal opinion there's almost no low hanging fruit left in the content + high volume + conversion game unless you get lucky with a few niches (but the advertisers find those niches for you). There are communities like http://www.leadscon.com http://www.leadscon.com and similar who do this stuff. It's really unstable business with constantly changing market + technology trends, unless you have a unique product and are developing marketing channels. Otherwise you're constantly fishing for got lead buyers and new traffic sources. Which is a constant grind and probably not a great business to get into, unless you have the heart for that stuff - then you could easily become middle class off of it. At least temporarily.
- kristopolous 8y ago
  I don't really make dishonest dollars so this is unfamiliar to me but I was under the impression content mills made easily detectable formulaic clickbait "What happened next will surprise you" "a local housemom discovered something" "7 worst celebrity hairstyles" etc... This is decidedly different. Here's some examples: http://www.informit.com/articles/article.aspx?p=1941206 http://www.informit.com/articles/article.aspx?p=1941206 http://james-iry.blogspot.com/2009/05/brief-incomplete-and-mostly-wrong.html http://james-iry.blogspot.com/2009/05/brief-incomplete-and-m... https://ask.slashdot.org/comments.pl?sid=2198700&cid=36293622 https://ask.slashdot.org/comments.pl?sid=2198700&cid=3629362... These are all at least 6 years old ... the key here is to make it look like it's legit and above board all the while you're just content farming from archives. I've had people repackage articles I've written on their own site and presented them as conference papers as just one-offs (I don't really care, whatever). This is a proposal for an automated plagerization engine, shamelessly lifting old content and repackaging it in a way suggesting it's exclusive. I've had top articles and projects on reddit, slashdot, and hn ... I tried amassing a portfolio recently and it's really hard to find the old references, blogs, and articles talking about it. The web is more of a moving river than a library. That's why I think this would work. It's incredibly dishonest but I don't think I've seen it done.

dmix 8y ago

1) You'd need 100k+ visitors to make any money off Google (1+ thousand a month). 2) Just having visitors is not enough, they have to be "niche" to an ad category and ideally looking for something to buy (not just information). The only thing that pays out is high-quality PPC traffiic or even better lead-gen for high-value products (health, debt, education, etc) with conversion-based pricing. 3) Google ads on content sites with CPM (impressions) pays pennies out. 4) Targeted blogspam sounds like it involves lots of humans beyond just mechanical turk, with some text parsing and rewriting which is a big overhead 5) This is basically the business model of Buzzfeed and other cancerous sites plus countless "niche" sites you've never heard of with high quality domains + foreign content mills + partner networks+cross domain advertising systems to convert in bulk. If you want my personal opinion there's almost no low hanging fruit left in the content + high volume + conversion game unless you get lucky with a few niches (but the advertisers find those niches for you). There are communities like http://www.leadscon.com http://www.leadscon.com and similar who do this stuff. It's really unstable business with constantly changing market + technology trends, unless you have a unique product and are developing marketing channels. Otherwise you're constantly fishing for got lead buyers and new traffic sources. Which is a constant grind and probably not a great business to get into, unless you have the heart for that stuff - then you could easily become middle class off of it. At least temporarily.

kristopolous 8y ago
I don't really make dishonest dollars so this is unfamiliar to me but I was under the impression content mills made easily detectable formulaic clickbait "What happened next will surprise you" "a local housemom discovered something" "7 worst celebrity hairstyles" etc... This is decidedly different. Here's some examples: http://www.informit.com/articles/article.aspx?p=1941206 http://www.informit.com/articles/article.aspx?p=1941206 http://james-iry.blogspot.com/2009/05/brief-incomplete-and-mostly-wrong.html http://james-iry.blogspot.com/2009/05/brief-incomplete-and-m... https://ask.slashdot.org/comments.pl?sid=2198700&cid=36293622 https://ask.slashdot.org/comments.pl?sid=2198700&cid=3629362... These are all at least 6 years old ... the key here is to make it look like it's legit and above board all the while you're just content farming from archives. I've had people repackage articles I've written on their own site and presented them as conference papers as just one-offs (I don't really care, whatever). This is a proposal for an automated plagerization engine, shamelessly lifting old content and repackaging it in a way suggesting it's exclusive. I've had top articles and projects on reddit, slashdot, and hn ... I tried amassing a portfolio recently and it's really hard to find the old references, blogs, and articles talking about it. The web is more of a moving river than a library. That's why I think this would work. It's incredibly dishonest but I don't think I've seen it done.