Site Targeting: Image Ads or Simple Text?

Which performs better?

[RockSolidWes]

Curious if anyone has had better success with image ads as opposed to simple text ads for site targeting.

Trying to optimize for maximum traffic, any suggestions as to what types of banner ads would work best for this goal?

Thanks!

Wes

[europeforvisitors]

Not all AdSense publishers accept "image ads," and common sense would suggest that they're least likely to be used on professionally-run sites that have display ads from other sources. So, if you go the image-ad route, you may be limiting the quality of your audience.

Still, why guess when you can test?

[TMF_Melissa]

I am finding graphic ads are doing well on content for me.

In some cases - 65% lift in volume

The CTR is equal to Search is some cases. 250x 300 do the best of me.

[RockSolidWes]

Thanks for the advice? I found that I am getting better results with image ads. I made about five image ads and let them rotate on a few sites - I just deleted the ones that have the lowest CTR.

I want to start adding new sites to my campaign.

I have two options, and am writing to see what others think.

1) visit manually all these sites and select the ones I think would perform better for me.

2) just dump about 100 different sites into my campaign and monitor them carefully... delete sites that do not meet conversion and cost metrics.

I'm leaning toward number 2, even though at first I may loose some money, but that is the only way to find 'hidden gems' on the site targeting arena.

Wes

[jimbeetle]

Definitely visit each site. If you just dump them into your campaign your wasting everybody's time if the ads aren't correctly targeted.

As a publisher I find that I wind up blocking most of the targeted campaigns on my sites (seat of pants number probably about 75%), because it is apparent that the advertiser is taking the lazy way out by using a shotgun approach. The ads usually turn out being not really relevant and not doing any party--advertiser, publisher, visitor--much good.