Does Google’s Spam Reports Work?
A few days ago, Matt posted a request on his blog for people to file spam reports, especially for keyword stuffing and Asian spam sites. After reading that, I had a chat with another webmaster friend who thought spam reports were useless. Her point was that spammers injected hundreds of new spam domains every day, and reporting each one, even if Google acted on those reports, was a waste of time. Her business also doesn’t rely much on Google traffic, so that does skew her perspective a bit.
On the other hand, I file spam reports often. I think spam reports are useful and they’ve worked in the past to knock out hundreds of domains. The dissatisfied? link can be used to report spam in batch mode. I also hardly doubt Google just takes spam reports, verifies them, then black list individual sites. I’m sure they’re taking note of spam tactics used to improve their spam filters.
Anyway, a few days after Matt posted the request on his blog, there are over 60 replies on it. Most are your average gripes and views like “spamming is bad”, “adsense created spam”, etc. Still, here’s a list of some comments that stuck out in my mind:
Google does absoluetly nothing against hidden noscript content (or content in hidden divs). I reported about 5 examples to Google and Matt about hidden div spam ages ago and nothing was done, so now I do it on all my sites and am making an absolute killing. I’ll keep doing it until I see the original spammer I reported get axed (that’s www.ambergreeninternetmarketing.com and its clients).
I admit after reading that I wanted to see a concrete example to test out on this site.
Rawalex adds:
Harith, the response time between “spam in the index” and “spam removed from index” is very critical to how profitable spamming can be.
PhilC writes:
If I ran a search engine, and I wanted to deal with spam, I definitely wouldn’t remove sites that were reported as spamming. I’d want to write algos that would deal with them, and others that contain the same sort of spam, and I’d want them in the index so that I could see if the algos worked.
This line made me chuckle:
You’re sending Matt unsolicited job offers through spam reports? That’s sort of ironic isn’t it?
Just tracked this down…
I just wanted to add this little piece of math to your discussion:
If a single spam domain can generate $100 per day, every day that it remains active and listed in the SERPs adds another $100. If it stays for 30 days, it means $3000 (all example numbers, don’t ask me how spam makes this money…). If the costs to buy the domain, put up the pages, and work the links to get it into the SERPs took 1 man day (and it probably takes less) add up to, say $200 (value) then you are looking at a spammer making $2800 per day. (assuming he or she gets 1 new spam domain per day into the SERPs) Not too shabby, right?
Now, if you filter the spam out quickly (say less than a week) he goes from $2800 a day to $500 per day. Filter it out even faster, and the profit drops even further.
More importantly, if Google was active in scanning the the top SERPs for spam, you would see that number drop even further, as the lifespan of the most profitable SERP spams would be very small indeed.
If Google worked a little harder at fighting spam QUICKLY as opposed to strictly by algo changes, they would kill off the big profit motive from spammers, and as a result, there would be less spam to remove because it would be less profitable to start with.
Think of it as the “broken windows” theory of spam. :)
RawAlex said this on April 20th, 2006 at 12:13 am