Reporting SE spam -what are the odds?

[rcjordan]

I'd love to get some inside information from the SEs, but that isn't going to happen. So, I'm going to use admittedly flawed logic just to try and get a very rough idea of the scope of their editorial problem. Using Google:

From their front page: Google - Searching 3,083,324,652 web pages

Let's say a mere 1/2 of 1% of these involve questionable techniques under their guidelines (3,083,324,652 x .005 = 15,416,623)

and only 1/2 of 1% of these are reported (15,416,623 x .005 = 77,083)

Assuming an editor can keep the pace of reviewing/documenting/processing 1 page every 15 minutes and given a 2000-hour work year, you're looking at roughly 10 man-years of editorial work.

And that's at extremely conservative .005 multipliers. I'd hate to be coping with what I suspect are the real numbers.

[mivox]

So, umm.... why haven't they dropped my competitor's spammy site from the index yet? I've been waiting, like.... 5 minutes at least since I submitted my spam report. :o

[brotherhood of LAN]

Assuming an editor can keep the pace of reviewing/documenting/processing 1 page every 15 minutes and given a 2000-hour work year, you're looking at roughly 10 man-years of editorial work.

Could maybe also assume that Googles algo is less than perfect, and those on the other side of the fence as "the spammers" are sometimes unfairly penalised and making google help desk (and editors) well aware of it.

So you can double the workload of that editor straight off :)

[rcjordan]

At a $30k GROSS salary (that's not much if you're including exployer taxes and benefits -I'm allowing nothing for overhead, like office space), each 15 minutes costs $3.75.

[jeremy goodrich]

Wow. So, until we hear about Google opening up an office in someplace where they could actually get a fleet of editors for that price they prolly won't be doing that. :)

No wonder why the PR rep here always says, 'they prefer scalable spam fighting' ...it'd be impossible to fight it by hand.

...also, it reminds me why so many engines went into the toilet - they forgot they were search engines, - not directories.

[mivox]

OK, so lets assume 77,083 real spam pages are reported out of the current Google index. Then there's BOL's very good point about the flood of "Hey, I'm not a spammer!" protests they no doubt get after every update. I'd be willing to go with his suggestion that those emails would rougly double the spam report numbers...

Heavens! Up to 154,000+ reports to look into.

I'd be there are a lot of people who also report competitors' sites that may be pushing the edge, but aren't actually "bannable" spam. Probably less than the actual spam numbers though. How 'bout we slap a purely theoretical 25,000 on top of the previous number to cover false-alarm spam reports (that nonetheless need to be checked into)?

Now were talking around 175,000 "reports" to be checked out of the current index. And we're up to what... around 23 man-years of work at the previous estimate of 1 report investigated per 15 minutes, during a 2000-hour work year?

Fine. I'll give them another half hour to remove my competitor from the index in that case... ;) They'd better get cracking though.... who knows how many new spam/false-alarm-spam/I'm-not-a-spammer pages will need to be checked after the next update!

[Lots0]

And as rcjordan said, the multiplier of .005 he used is most likely low.

IMO very low - you might be able to take that 175,000 figure and quadruple it and still be on the low side.

[Brett_Tabke]

I think the time taken to surf a page is way over rated. If they spend more than 1 minute on a page, I'd be surprised. With the utils and surf toys they have available, pages can be viewed in rapid fire sucsession. They are looking for the obvious stuff first. Something such as a kw stuff page could be spotted and acted upon in 5-10 seconds and on to the next page to do the same. Browser shell programs make quick work of serial surfing.

[chiyo]

I think its pretty clear on spam reporting to Google.

If you are doing it to try to get rid of a competitor its a complete waste of time.

If you are doing it to help Google clean up a certain spam technique, it may help long term.

Up to you...