Hey,
I have a long post, but stick with me, as I have only one question for now
This is something I've seen many people wonder about but I've never found the answer. When starting a new campaign, whether you're targeting LAL audience or cold broad audience, what usually happens is that on the first few days campaign performs really well, and then between 3rd and 5th day it just completely dies off. The campaign in question is purchase conversion one.
And I've seen the same happening over and over again over the past few months. I've seen ROAS on the first day starting at high as 12 and then finishing at 2-3 after few days.
Recently, I've tested broad audience of around 600k people (this is pretty broad for my country as it has nearly 1.6 mil Facebook users overall) with 10€ daily budget. It started performing really well, but on day 4 it practically died out (daily CTR was literally cut in half on day 4 and CPC jumped from 0.17€ to 0.41€).
Since I've noticed that Facebook mostly showed my ad to male audience, I decided to split the audience based on gender and broaden it up a little bit (up to 800k people). I've placed CBO budget of 20€. This time different thing happened - over the first 3 days, daily CTR rose from 0.73 to 1.55% while ROAS from 2.88 to 4.16. And then on day 4 daily CTR again dropped from 1.55 to 0.52% with 0 purchases.
I've killed the campaign for 2 reasons:
- I've seen this scenario more than a dozen times over the past few months. And it's always the same - daily CTR suddenly drops, and there are no purchases anymore.
- Overall ROAS was just 3.46 which is not profitable for my business
So I've noticed the following - my frequency was already at around 1.6-1.7 for this audience, even though it is pretty broad and I've spent only 90€ on it. Cumulative frequency, taking into the consideration the previously mentioned audience of 600k, was 1.95.
I've read in several groups if frequency rises so quickly in such a broad audience, the problem was with low budget. Facebook simply doesn't have enough budget to bid on a more qualified audience and you end up recycling only a chunk of that 800k audience, resulting in ad fatigue and ROAS dropping off. Keep in mind I'm using lowest cast optimization.
So I've said - let's do one more final test. I've decided to switch off audience separated by gender and just show ad to the whole audience of 800k. I also did one more tweak - I've excluded past engagers, visitors and customers from the audience. This lowered my total audience by maybe 10k. My daily budget was 24€ for this one.
First day (yesterday) it performed really well. Although CTR was incredibly low, and CPC extremly high, the day ended with 4 purchases and ROAS of 8.14. Probably the best daily ROAS I've seen in a while when atleast 2 purchases happened. Daily Cost per 1000 people reached was also pretty low compared to the previous campaigns I've tested on this audience (1.8€ compared to the average of 2.8€).
My explanation for this results are following:
- 600k audience couldn't perform well for a longer period as the daily budget was only 10€ and Facebook didn't have enough money to bid for a more qualified audience
- Gender-splitted audience, even though it ran on 20€ daily budget, couldn't perform well for a longer period, since Facebook allocated around 12€ to the male gender and 8€ to the female gender, which again made the budget too low for Facebook to bid for a more qualified audience
- 800k audience (which overlaps 100% with 600k audience mentioned in the first dot) performed really well because Facebook had 24€ to bid on a more qualified audience. Low daily Cost per 1000 people reached could be a proof of that. Facebook had a budget to bid for a more qualified audience that wasn't available in the previous 2 tests, which resulted in 4 purchases at 8.14 ROAS even though CTR was low and CPC was high.
This conclusion is based on a presumption that higher daily budget unlocks more qualified audience. Simply put - If you spend 5€ a day for 30 days (150€ in total) you would reach much lower percentage of the total audience than if you'd spend 150€ over the 5 days with a daily budget of 30€.
Is this true and could this explain why good performing campaigns usually die off within few days?
Thank you for your answer!