Do you have a specific test process that you apply to every campaign you run?
ie: i'll start with 4 creatives + 3 angles + 2 landers + 3 offers, etc
are there any set guidelines for something like this, or is it dependent of traffic source, niche, personal preference, etc
There are no carwed-in-stone rules for this, we all have our ways of approaching this. But there are some general guidelines you can go by.
To put it simple, you want to make sure you're testing enough variations, but at the same time you don't want to overdo it and test too many variables. The more combinations there are, the more you need to spend in order to get reliable data. Because of that, your budget will be a limiting factor too.
When I'm testing something on banner traffic these days, I usually use 10 banners, at least 2-3 LPs and 2-3 offers. The banners cover several angles or styles, so it's not just one style either. But again, it's not always like this. When I see a good offer with several url variations, I test 5 or even more of them in the same campaign, I just let it run for longer. Same with the LPs, sometimes I test 5 or 6 at once. It really depends on the current situation and what I'm testing.
So if you are looking for some initial setup, try to stay under 5 ads, 2-3 LPs and 2-3 offers. You can add more if you budget allows for it, but I wouldn't go over 5 per variable in case you're just starting in AM (is that the case at all?
Hello!
Mostly it is individual cases, but some details of your campaign, like a traffic source, verticals, will help you to take advanced tips from the industry experts.
Common advice is testing at once several landing pages, creatives, any your ideas. Feel free to reach help to a support team or a personal manager of your network. They are always know different skyrocket ways.