So I made a few groups with some incredibly general KWs to try to see how long it took them to be slowed and disabled (how many people have that as their goal?).
The first month, I didn't use partner listings, but as the KWs were not getting disabled, I expanded it to parther listings to see if there were higher standards which would get them slowed or disabled. None of them use conversion tracking or content syndication.
They all have max CPCs of at least $2.
The average position is 4.4 - only one was lower 8 - so all first page listings.
One KW is now at 71 clicks from 33,423 impressions (0.2% CTR)and listed as moderate.
The at risk ones are:
16 clicks in 7981 (0.2%) imp
4 in 3594 (0.1%)
This one I find amusing, its for 'online widget'
"online widget" 15 of 2680 - strong
[online widget] 6 of 1803 - at risk
online widget 2 of 571 - strong (based on the number of impressions - I think this is that case of a broad match being changed to a phrase/exact match by the system)
widget 0 of 945 - moderate
The only KW to get disabled was 13 clicks in 7,034 impressions (this was also the only KW to have a non first page listing - it was 9.6)
The worst offender is: 17 clicks in 5649 (0.3%) and listed as strong.
There are several more KWs where there are 2500+ impressions, less than 0.4% CTR, and labeled moderate.
Only one of the KWs has been slowed or disabled - at worst, they are listed as at risk. Some have been 'at risk' for a month, and still nothing has happened to them.
The entire campaign is now 523 clicks in 109,725 impressiosn - very poorly preforming (as expected for such general keywords).
It seems there is some other 'behind the scenes' factors that get KWs disabled. The entire account has an alltime CTR of 7.2%.
It's hard to tell if its the max CPC, higher positions (some of these KWs have 15+ competitors) or maybe the overall account CTR that keeps these poorly targeted KWs afloat - but for some reason, they just are not being disabled.
If I'm in position 2 with 15+ competitors and getting a 0.3% CTR, I have a hard time believing that more than 1-2 of the competitors are getting the necessary CTR to keep those words active.
Any theories?
So it would seem reasonable that low CTR ads can still survive provided they perform better than other ads nearby. How good is the ad copy? Have you written it to discourage clicking, to ensure that your ads have lower CTR than their neighbours? Or is the copy in your normal style, and probably far better than the ads nearby?
Does every KW need to hit 0.5% CTR?
Some seem to need less
eWhisper, as always, you ask great nitty-gritty questions!
This is almost certainly an example of something that has come up before: the fact that minimum CTR is normalized for the ad's actual position, each and every time it appears. The 0.05% CTR you're referring to is for the first position, and lower postions will each be slightly lower.
AWA
the fact that minimum CTR is normalized for the ad's actual position, each and every time it appears. The 0.05% CTR you're referring to is for the first position, and lower postions will each be slightly lower.
I don't think that's been stated before, or if it has, I missed it - exactly what I was looking for.
As usual, always informative, thanks AWA.
I don't think that's been stated before, or if it has, I missed it...
Heh! I'm almost certain that this came up in answer to another of your questions eWhisper, a couple of weeks ago. I just can't seem to put my hands on it.
Later, I'm sure I'll remember the context of the thread, and be able to find it. For now: no luck!
Frustrating!
<added> I just remembered the thread. It was the one in which you suggested a workaround to transplant an entire list of keywords into an Ad Group, etc. I'll look for it later, if you're not able to dig it up. As if that would happen. :)</added>
;) AWA
you're right about the statement:
the CTR would automatically be normalized for your new position. This process of normalization for postion is occuring 24/7, every single time your position changes.
being stated in this thread:
[webmasterworld.com...]
I didn't realize "normalization" referred to min CTR rate vs position.
So now its all clear. Thanks again.