In 2019 Facebook debuted something new and promising called Campaign Budget Optimization (CBO). It was supposed to be so good that it would soon be a default, required setting for all Facebook ads. However, Facebook is now walking it back and may completely get rid of it. Why get rid of such a glorious new feature? It doesn’t always work as intended. It sometimes hurts rather than helps your results. But fortunately, there is a solution.

What is Campaign Budget Optimization (CBO)?

Campaign budget optimization was intended to be an automated way for Facebook’s ad algorithms to spend your ad budget on the audiences (ad sets) that drive you the biggest results. It’s supposed to take into account the kind of result you want (for example, video views, or reach, or leads, etc.) and maximize those results.

CBO was a change from the old way of specifying exactly how much money you wanted to spend on each audience (ad set) per day. And that would be an improvement.

Facebook saw an opportunity to build optimization as a native feature of the platform to help advertisers see more ROAS, and eliminate the complexity of working with a third party technology/ad tech partner.

Issues With CBO Performance

In checking the patterns of spend, impression delivery and results in many campaigns for many clients over the last 6 months, we kept seeing that CBO was giving the most impressions to the wrong ad sets. There have been too many situations where an audience or creative was clearly performing substantially better, yet other audiences and creatives were receiving the lion’s share of impressions.

This is not what we want to see, and it hurt our ad performance. We kept having to correct it manually.

What DOES Work for Ad Budget Optimization?

We are able to manually optimize budgets to improve results quite successfully, with a simple process, using standard ad metrics. To optimize ad budgets correctly, you have to do it by hand. Which sounds like a big deal. And it’s no small deal…

  • You have to be able to analyze performance, delivery, audience size and ad spend to see if an ad set or ad is receiving more impressions than it deserves.
  • You have to know what date range to analyze.
  • You have to know how much data is enough for you to make an optimization decision.

All of that takes experience analyzing ads and optimizing for performance. Some of it is not intuitive. For example, you may never get the amount of data a statistics professor might want you to have. You may have to make decisions without all the info, or with less data than is ideal. What’s more, Facebook doesn’t even always give you audience sizes, and sometimes we have reason to suspect those audience size estimates are inaccurate.

Similar to medicine, clinical practice is not the same as research; optimizing real ad accounts is not like studying massive economies; optimizing smaller spends is more challenging than bigger spends.

Experience in the trenches optimizing ads is required- or you can get started and make mistakes and learn in the process, just like I did!

If you’d like help with it, reach out to us!

Brian Carter

An IBM futurist and top expert recommended by LinkedIn, Brian Carter has been a trusted influencer for over a decade. He's built programs and been a keynote speaker for companies like Microsoft, NBC-Universal, Citrix, Marketo, Salesforce as well as small businesses. He delivers motivational keynotes with practical takeaways based his 15 years of cutting-edge work. Brian is a bestselling author you may have seen on Bloomberg TV or in Inc, Entrepreneur, The Wall Street Journal and The New York Times. He has over 250,000 online fans and reaches over 3 million people per year.