Multiple Facebook Ads managers are saying that March was a really bad month in Facebook. Many FB mangers that I have talked to say the biggest drops they have seen were right after the outage last month.
What is going on?There are many theories. I even heard that the best engineers are no longer part of the ad algorithm team. They are working on other projects while less talented engineers are left adjusting the ad algorithm.
I doubt Facebook is allowing a bunch of junior engineers play with their most valuable part of the business.
We have some other insight which seems to make more sense.
Only one of our clients that spends around $200,000 per month has seen a major drop in March. We have been investigating why and we have a theory…
It might be due to changing ads often.
I know. Changing ads often is general practice to keep ads from getting stale, but hear me out.
For most of our clients, there are a few key evergreen ads that we run. These ads never change and just keep getting better because there are so many likes and comments.
These evergreen ads grab attention and engagement. People are more willing to engage in an ongoing conversation rather than comment on something that has few comments.
Yes, there is the argument of the stale ad. We do test fresh ads on a small scale and allow the comments to build as the main evergreen ads carry the account. These main ads don’t go away for month until another ad has enough engagement to take its place.
When you launch new ads, Facebook algorithm sees only a few comments and less engagement so they throttle the reach for these new ads. These ads don’t engage Facebook users compared to ads with a lot of comments from other advertisers.
Ultimately, you are competing against many other advertisers for a limited inventory and Facebook wants the best engaging ads there.
Which ads are the best engaging? The ads with a ton of comments and likes… Social proof.
Facebook’s reasoning makes sense if this is actually what they are trying to do, but it is having an impact on the accounts that change ads often.
The more you try, the more it declines
That chart shows that volume and CPA became worse the more we tried to fix things. We kept trying new ads, new campaigns with old ads, new Ad Sets, etc. We tried everything.
Focus majority of ad spend on old ads
Now, we are starting to recover CPA as we are focusing most of the ad spend on old ads from old ad sets which have the most comments. I think the reason for this was Facebook gave ads with a lot of comments the best treatment.
So, we looked at the ads that had the most comments and these are what we are focusing on. The ad with 1914 comments and the ad with 464 comments are the ads that we want to spend the most on. The ad with 464 would have 4640 comments if we spent ten times the ad spend which is what the first ad spent. At least in theory. So, those are the 2 we are focusing on and it is working.
Focusing on engagement has always been the case, right?
Yes, but it seems Facebook’s algorithm is leaning more and more this way. If this is the case, which we really think it is, you need to make sure your ads scale slowly before they can get a lot of budget. Once an ad has plenty of comments and engagement, then you can open up the budget. Until then, run old ads that already have a lot of likes and engagement.
Of course, there are many strategies that might be able to solve this problem such as manual bidding strategies. We are exploring these this week to see how they do
We are getting better results now
Finally our top campaigns are getting back in line so our solution is working.
Having the same issue?
If you have experienced issues with your accounts, comment below. We want to see how performance has been over the last month and if anyone can confirm they are seeing what we are seeing.