BillyS - 11:10 pm on Oct 2, 2005 (gmt 0)

Has anyone seen a return to the "sandbox" after getting inflated page counts or a temporary decrease in traffic?

It's a bit confusing to me, but here is my logic...

Google knows the actual page count of any site in its index - it's just a calculated value (actually just a count). But why doesn't it show the real value? Is it too lazy to calculate the value and uses an estimate instead?

But how else is this estimate used? Clearly if it stored the real value, then it would show that in the site:www.foo.com query. So perhaps when a site gets large enough it uses an algorithm instead of actual count - perhaps to save CPU cycles. But if this estimate is now used as the "actual" for a site, then perhaps a penalty can kick in for getting too big too fast. Just a thought…

If I'm wrong, then of course Google should explain why it used an "inflated" page count when it could just use an actual value it already has stored somewhere. Personally, I'd rather believe the first explanation (I'm not big on conspiracy theories).