one of the most frequent mistakes I see in the ad accounts that stopping brands from profitably scaling is a bad testing structure but luckily I found the optimal way it can be slightly different depending on number of available concept on a weekly basis overall budget and the AO but here's the exact process I use to test creatives on accounts spending both $500 a day up to 50K a day let's first cover the technical stuff all of the testing is done with the best performing audience in the ad account in most cases it's a broad audience
in maybe one out of 10 ad accounts that's an interest stack if that's the case then I do all of my tests with that audience I always exclude existing buyers because my goal is to reach new customers so then I exclude both purchase list 180 days audience as well as the sync list of buyers from clavia it's like double the protection and I'm want to exclude all of the buyers because none of my clients want to spend at least to acquire someone who they already acquired or CBO this is the most frequent question I'm doing
all of my tests in an AO campaign where each new test is a new asset I want each of my tests to get a fair chance of getting spent if I test in a CBO Facebook would probably favor one or two concepts either based on click rate or average watch time which is not often connected with the lowest CPA also my goal is not to beat that winning concept and get one ultimate main winning ad my goal is to get few additional winners that are supporting that one winner which are still above my target Ross
and they are acting as a support to those couple of main winners just because if I have a Target Ross of two and I have an ad that's crushing on 2.5 doesn't mean that I don't want an ad that has 2.2 Ross still running in my ATA another reason why I'm not using CBO is the fact that often in my ad accounts testing campaign is actually the scaling campaign no matter if it's labeled as testing that means I'm scaling each of the individual good performing assets or tests so in my campaign it could be happening
that one asset is at $100 and the other is 1K just because it's performing great if I use CBO and I want to scale that testing CBO I might get a necessary spend on the tests that are actually not performing and I don't have control over where Facebook would spend the budget and I want my budget to always be connected to the performance and headset setup my tests are usually grouped either by angle let's say anti-aging and anti- acne or concept let's say product demonstration video or unboxing or per Creator let's say girl a or
girl B and each of the test is in its own asset it's important to mention that I'm not testing just one variation of the concept because that's a hitter Miss approach and that way you're forcing Facebook to spend money on that variation no matter how good or bad it is which makes no sense to me that's why every concept that I'm testing has three to five different variations of that same concept when it's a video that variation is mostly hook but sometimes it is a length of the video or a thumbnail when it's an image
it's either a different layout different design or different text on the image itself this is maximizing my chances of finding a winning creative because I'm giving Facebook choice to pick which out of these were iation could be the best Facebook is usually pretty good at predicting which ad can be the winner and which ad can handle the scale I say usually because sometimes you would see Facebook spending money on a specific variation that's not getting the results in those rare cases I would turn off the ad that's getting most of the spend and let other
ads get some spend to see if I can get some winner from those so we covered the ad set and the ad setting here are some other important things to mention all of the ads in the ad set use the same ad copy same headline same landing page and same call to action basically the only variation is the creative itself the reason for that is because I'm testing multiple variation of the same creative is because Facebook told us that creative has over 60% impact on the performance of the app so it makes more sense for
me to test that variable first I would not know if I'm testing multiple ad copies multiple thumbnails and creatives if something works or not I would not know whether that ad worked because of the one element or because of the other timing I'm launching all of my tests on the same day of the week in most cases it is Wednesday just because we usually find that Monday and Tuesday are the worst performing days for most of the brands so it doesn't make sense to have a disadvantage when launching tests I want my ads to get
the most chance of succeeding so I let them run through weekend which is usually the best performing period Then on Monday I do analysis of all of the tests and I determine the next steps it's may be cutting some of the ads it's may be cutting some of the assets it's maybe increasing the budget on good performing adsets as I mentioned I have some ad accounts where my testing campaign is actually scaling campaign but you should know that this is not all to the process it's just a tip of the iceberg so if you're interested
in video which I'm explaining how I'm coming up with the test or analyzing the results or actually scaling the winners comment below if you are on the other side brand that's spending at least 50k a month on Facebook ads and you want me to implement this structure in your ad account so we can actually scale your ad account and your brand book a call below cheers
