testing google

injoke

i'm going to see how good google is at targeting ads by listing increasingly obscure references and seeing if google can provide a relevant ad at the side. yes i'm bored.

starter for 10:

net curtains

injoke

damn, it came up with an advert relating to the word bored but not to net curtains. perhaps if i:

net curtains

net curtains

net curtains

net curtains

NET BLOODY CURTAINS

staplejack

their dartboard - like targeting depends on quite a few factors (much like being able hit a dartboard with some darts), not least is the relevancy of the content and the number of times that you mention the keyword (dartboards have numbers too)...

injoke

net curtains is clearly failing, but then i've always been crap at darts.

how about:

cleft pallette

cleft pallette

cleft pallette

staplejack

In the interests of achieving anything, how about "Finsbury Park locksmiths?"

Pete

My turn, my turn!

Pneumoconiosis compensation
Pneumoconiosis compensation

andy

I've tried this. It doesn't work. <http://www.stroudgreen.org/discussion/1132/hosts-toggle/>;

David

Hosts toggle. That's an interesting phrase. Hosts. Toggle. It's a shame you never got a result from Hosts Toggle. Did you try toggles host? Host toggle.

tosscat

An ad for pest control is coming up - maybe it is quite effective.

Pete

Aaaaah, no, it do work, but it scans for word over the site as a whole. That advert is the "same" advert inserted into the stylesheet, yes? So all the links to google from the same code analyses everything the code is next to.

In short, if you had lots of different google ad codes and had them on a random loop, the ads would be more associated to fewer words. Meaning the weirder keywords would come through.

injoke

i have failed