I am stuck on whether I should split test stuff or just run each version through a number of tests.
I create small sites based on Affiliate income, and run them through AdWords.
While we are on the subject... What is a good indication of whether a site will be productive or not. How many clicks until you get a good picture? I have about 40 clicks so far on this new site, and not 1 sale.
I guess tweaking the ad can really make a difference, but my current ad does describe what my site does which is compare different things in that niche. All the KeyWords I use are targeted too.
Thoughts?
C.K.
Design 3 or 4 ads for the same adgroup and set them to serve evenly (turn off ad optimising in the Adgroup settings). Give each one a different tracking code and (if you can) try to find out which one reaps the most revenue.
When you've narrowed it down to one produce a few improved versions of that one and run them through the same process. Really small changes can make a big difference in Adwords creatives.
40 clicks isn't much at all - I don't know what kind of industry you're in/budget you have but I'd give it a good run of a month or so before deciding.
Thanks edit_g for your response to both posts. I take it you are saying to leave the site for now, and work on split testing ads?
I have up until now run only 2 ads at once with the optimization turned off. I can definitely track everything, I just always thought that more than 2 ads was too much.
If I do 4 ads at once, I think a good approach would to be 2 ads that use one angle, just worded differently/whatever and then the other 2 with a different angle and worded a little differently from each other.
Thoughts?
Thanks,
C.K.
2 ads at once with the optimization turned off/ more than 2 ads was too much.
It might be too much - I just always find that I have at least 3 or 4 approaches that I want to test and I want to test them all at once. ;)
If I do 4 ads at once, I think a good approach would to be 2 ads that use one angle
I usually start off with completely different ads - and then hone in on the best one and refine. I try to see what works best: price based messages, savings, calls to action, service messages etc.
If I do 4 ads at once, I think a good approach would to be 2 ads that use one angle, just worded differently/whatever and then the other 2 with a different angle and worded a little differently from each other.
If your ads with completely different angles have similiar conversion rates, then you need to look at your keyword list - generally you'll see 1 of the following.
1. These are broader keywords, so searcher intent is difficult to predetermine, and using multiple angles is a good thing.
2. You have too many different keywords in your group, and splitting them into their own group based on your ad angles means you can have two very good & different types ads running all the time for those keywords.
3. Your products and/or storefront has several types of specials (free shipping, 25% on orders over $100, etc), so the angles are based on the offer and not necessarily a wide variety of keywords, thus this type of angle testing and refining is the best approach.
If it's 1 or 3, then testing angles and refining ads based on conversions/ctr is a good thing. If it's number 2, then you need to redefine your adgroups.
