make reviews that stick

seo_seo_seo said:

Read more about VPNs and proxies.
Also, yes, you need to use different IPs.

Click to expand...

MatthewGraham said:

There are many factors that go into anti-spam filters. Google Reviews have more complex anti-spam algorithms than most other review websites. Anti-spam algorithms may look at factors including:
* IP address
* Type of IP (residential, data center, etc.)
* Location of the IP address (is an IP in Dallas, TX reviewing a business in Indiana?)
* Review content
* Content being flagged as spam
* When reviews are posted (are all posted at midnight GMT? Exactly one review per day?)
* History of the business (has the business had many reviews flagged as spam before? It may be on a blacklist/greylist to get taken down more easily)
* Account details for the reviewer (are they new accounts? Never used for anything but the review?)
* How the user traversed to the page (is everyone who is leaving a review coming to the page by clicking a direct link to the review page? Or do some come directly, some through a search for [company] reviews and clicking the link there, etc?
* Other reviews posted by the same account (do all of the accounts used for reviews on business X in New York also have a review for business Y in New Jersey?)
* And many other factors
And in addition to automated anti-spam algorithms, reviews may also be subject to manual review. Google has peons that personally look at potential spam.

Which of these factors are and are not used is of course dependent on what review site you're posting on. And of course companies are secretive about their anti-spam algorithms, so it's hard to tell which factors are used and to what extent. When you try to game/get around/cheat/trick algorithms, instead of thinking about it as a basic "oh I just have to do these two easy steps and it will work" you have to think about it from the perspective of the anti-spam developers. If you were coding an anti-spam algorithm, what would you include? These bullet points are all things that a single programmer could implement fairly easily, meaning that multi-billion dollar companies could certainly implement them as well.

Click to expand...

MatthewGraham said:

There are many factors that go into anti-spam filters. Google Reviews have more complex anti-spam algorithms than most other review websites. Anti-spam algorithms may look at factors including:
* IP address
* Type of IP (residential, data center, etc.)
* Location of the IP address (is an IP in Dallas, TX reviewing a business in Indiana?)
* Review content
* Content being flagged as spam
* When reviews are posted (are all posted at midnight GMT? Exactly one review per day?)
* History of the business (has the business had many reviews flagged as spam before? It may be on a blacklist/greylist to get taken down more easily)
* Account details for the reviewer (are they new accounts? Never used for anything but the review?)
* How the user traversed to the page (is everyone who is leaving a review coming to the page by clicking a direct link to the review page? Or do some come directly, some through a search for [company] reviews and clicking the link there, etc?
* Other reviews posted by the same account (do all of the accounts used for reviews on business X in New York also have a review for business Y in New Jersey?)
* And many other factors
And in addition to automated anti-spam algorithms, reviews may also be subject to manual review. Google has peons that personally look at potential spam.

Which of these factors are and are not used is of course dependent on what review site you're posting on. And of course companies are secretive about their anti-spam algorithms, so it's hard to tell which factors are used and to what extent. When you try to game/get around/cheat/trick algorithms, instead of thinking about it as a basic "oh I just have to do these two easy steps and it will work" you have to think about it from the perspective of the anti-spam developers. If you were coding an anti-spam algorithm, what would you include? These bullet points are all things that a single programmer could implement fairly easily, meaning that multi-billion dollar companies could certainly implement them as well.

Click to expand...

MisterF said:

Great reply.
People do fall into the trap of automating everything and leave footprints like this; When reviews are posted (are all posted at midnight GMT? Exactly one review per day?)

I've seen people do the same with backlinking and also social media signals, where they add the exact same numbers ( 100 G+, 100 FB etc ) and they do them all in one blitz.

Click to expand...

Mihai Stancu said:

And does it work?

Click to expand...