Does Ad Scheduling Work With Quality Score Algorithm?

Lower Positioning = Lower CTR = Lower Quality Score

[LukeC]

OK bear with me - I am hoping this is Adwords paranoia!

I employ ad scheduling, so for instance on monday I bid 10% less and on friday 10% more.

By bidding less on monday your position drops, then your CTR drops which affects your overall quality score and therefore when friday comes around you can't make up the ground.

Has anyone else seen this?

I am beginning to think that I should never bid less, just increase on hot converting days, otherwise you end up with this yoyoing that messes up your quality score and a scheduling system that never actually does as it should.

Any comments greatly appreciated.

Luke

[netmeg]

I've never had occasion to use ad scheduling myself. I think that's something you would only know if you did some testing.

[LukeC]

thanks for your reply netmeg.

for the industry I work in fridays convert better than mondays, while mondays convert better than sundays etc, so I adjust the bids according to that days conversion rate.

However I think it may recquire some pretty sophisticated maths to factor in the effect of the quality score.

Perhaps a solution would be to kill any ads that drop below a certain position in order to preserve CTR, I'm not sure if you can do that I'll check.

[patrickSMC]

I just found this on eWhisper's Blog:

The CTR is normalized by position. AdWords understands that an ad in position 1 is going to receive (on average) a higher CTR than an ad in position 5. Therefore, they aren’t going to penalize the advertisers in position 5 for having a lower click through rate than the advertiser in the first position.

[bgtheory.com ]

Now this has kind of stoked my curiosity and I'm wondering about the relationship between match types, CTR and quality score.

[LukeC]

wow that is a fantastic find and a very helpful link, thank you patrickSMC.

It's a huge relief to hear they normalise the CTR by position, I can carry on ad scheduling without fear.

[LukeC]

This raises (at least) 2 new questions:

1. If CTR is normalised by position then what is the point in rejecting traffic from lower positioning?

Is there ever a case to select positions 3 to 6 only? Surely it should always be 3 to 10+?

2. If Google normalises by position it is then certainly essential to schedule your ads to show at the best converting times, say cutting out 10 to 20% of poor converting times.

[edited by: LukeC at 9:33 am (utc) on July 7, 2009]

[patrickSMC]

I definitely agree with question number 2. For example, I use Google Analytics to look at conversions on an hourly basis. I found that in one campaign, I had basically only made one conversion in six months between the hours of midnight and 2 AM, so I cut out advertising in those hours.

I would agree with question number 1 as well, that it should always be 3 to 10, but I don't have anything to back that up. I think I just have a very strong personal bias against position preference.