So I have 1 adgroup with 5 variations. They have run for approximately one month. Not all of them have run the same length Stats Clicks Ctr Var 1 42 21.4% Var 2 9 20.5% Var 3 32 16.8% var 4 32 15.8% var 5 1,156 13.0% Interestingly var 2 is me trying out abbottsys suggestion to repeat the headline in line 2 and 3. Have run it for two days. It's INSANE...lol Anyway, my real question is do you drop var 5, maybe var 4 and even var 3 and then try new variations of var 1 and 2? I don't have a conversion tracking cookie for a variety of reason so please assume that's it's the same for every click thru. Roligan ps. var 6 just has the headline and nothin in line 2 and 3. Stats: Impr. 40, clicks 2, CTR 5%
I feel there may be a lesson here about serving up the adverts equally. If you get 1300 clicks per month, you should probably be testing two, or at most 3 adverts, serving each up the same number of times... You can't realistically compare an advert with 42 clicks from 200 impressions against an advert with 9 clicks from 50 impressions... Also, from your question, it sounds like the adverts have been running for different lengths of time. Again, you can't do that - it'll invalidate your results since your clickthrough rate will vary from day to day (and certainly from month to month). I would suggest this. Keep advert 1, and the original advert. Run them 50/50 for a week. Then analyse them to see which is better. Delete the worst one, then run advert 2 against the winner. Keep repeating the process...