When you Split-Test your Ads.

do you give them "Equal" Time or Let Google "Optimize"?

[bigdealioo]

I heard that its better to give equal number of impressions to each ad so it gets a fair shake. But some say better, let Google do its thing with optimization. What do you think? Is Google "auto-optimization" pretty good, math-wise?

[chief72]

If you genuinely want to split test your adcopy you need to give all ads equal exposure. You want to be able to see how the different adcopy performs under the same match conditions. This way you can make an accurate comparison and improve the underperforming adcopy.

To use a bad analogy, you wouldn't send two 100m sprinters out on two different tracks in different parts of the world, at different altitudes and with different weather conditions, time them and declare the guy with the faster time the winner.

[poster_boy]

But some say better, let Google do its thing with optimization.

Google only optimizes based on click-trough rate. Copy testing should test conversion rate as well as CTR.