Campaign Budget Optimization for Facebook ‘Store Visit’ campaigns and our change of tone

Feb 7, 2020 | Uncategorized

On numerous occasions Social Fulcrum has been advocates for Facebook’s Campaign Budget Optimization (CBO) tool. But in all that support we made it clear that we supported the use of CBO for most campaign types. We couldn’t recommend using CBO on ‘Store Visit’ campaigns for good reason, our internal tests told us not to.

What is Campaign Budget Optimization (CBO) anyways?

Like we do with everything Facebook rolls out, we tested CBO. Then we tested it again…and again…and probably again…and most likely again.

We tested ‘Store Visit’ campaigns using CBO vs. ‘Store Visit’ campaigns without and compared them on a cost per store visit. In every test the campaign running without CBO not only won, but won at a very high margin.

So what’s the issue? Our tests were run in two week increments. When we looked at CBO over a four week time period we noticed something very surprising.

CBO Performance
Week 1: Under performing
Week 2: Better, but still under performing
Week 3: Caught up and out performing
Week 4: Significantly out performing

So what does this mean?

Well, first off, it means our testing and opinion of CBO was a bit off. We determined that CBO likely needs at least a week’s worth of data to optimize and as a result the first week did and most likely will always perform worse than a campaign without CBO.

Our latest advice:
As mentioned, we previously supported CBO except for on ‘Store Visit’ campaigns. Now, based on our latest tests and findings, we not only recommend using CBO for all ecommerce campaigns but we recommend using it for ‘Store Visit’ campaigns as well – as long as the campaign will be live for greater than 4 weeks.

Want to chat more about CBO? Let us know!
Looking for a free audit of your current or past social campaigns? Request one here.

1 Comment

  1. Great content! Super high-quality! Keep it up! 🙂


Submit a Comment

Your email address will not be published. Required fields are marked *