“In the Facebook Algorithm We Trust” – Perpetual Traffic Podcast Dives Deep Into Campaign Budget Optimization
Facebook’s campaign budget optimization (CBO) is awesome.
And it’s working really well. Basically, you set a daily budget at the campaign level and then tell Facebook what to optimize for and they go out and do the rest. They split test all your creative, and ad sets and find the least expensive action you have set as your campaign objective (i.e. lead or video view).
I experienced a 50% reduction in the cost per lead within a week of launching the campaigns. I started with a lower budget and built up to what I was spending prior. It works really well. And, now I have to do much less maintenance.
Here are my notes from this amazing two podcast series. Burns and Pittman are super excited to share their findings with their audience, which is amazing because between Tier 11 and Molly’s consulting business they are spending millions of dollars a month on advertising.
I’m not going to keep you in suspense. Ralph and Molly love CBO, despite some of the challenges it poses.
CBO was Released With a Whimper, But Now ...
Facebook released CBO in November 2017. Molly didn’t start getting results until Fall 2018. Facebook takes a while to optimize new tools.
Ralph’s agency got a beta invite. They could see the potential but overall they discovered “it sucked.”
This might be a hint to let the beta testers work out the kinks of new Facebook ad features, then swoop in right when it's still new but stable.
Facebook's description of Campaign Budget Optimization.
Campaign budget optimization is a way of optimizing the distribution of a campaign budget across your campaign's ad sets. This means Facebook automatically and continuously finds the best available opportunities for results across your ad sets and distributes your campaign budget in real time to get those results.
A campaign budget is a budget you set at the campaign level (rather than the ad set level). The amount you set can apply to each day the campaign runs (daily budget) or over the lifetime of the campaign (lifetime budget). All campaign budgets use campaign budget optimization.
We optimize your campaign budget in real time on an opportunity-by-opportunity basis. Our goal is to get you the most results possible, and for the cost of those results to align with your bid strategy.
Molly’s Results: Very Good
Caveat: Molly Pittman always gets great results. She’s a Facebook advertising savant, so it may take you more time, but the probability is high that you will get good results too. The technology is solid.
Molly was doing a launch for a client. It was a lead acquisition campaign. Again, she's awesome, but she did have three things going for her:
- Seasoned pixel. Her client had already been running ads for this lead magnet. Facebook was familiar with what they were looking for, so it didn’t take as long to hone in on the most responsive audiences.
- Ads with social proof. She used the post ID from ads that already had social proof. This is a good thing to do whenever you’re using the same ads in different ad sets. Make sure you’re not creating duplicate ads, but rather are using the same one, so you can consolidate the social proof.
- Great offer. Her client did the heavy lifting of building an offer that converts. This certainly fast tracked Molly's CBO success.
Even if you don’t have these things you can still take advantage of CBO. Well, you need an offer that converts no matter what, or you'll have trouble staying in business, but you can work up to a seasoned pixel and ads with social proof.
Both Molly and Ralph recommend not doing testing on new campaigns, offers, audiences, or ads with CBO. They promised to drop some knowledge in a future episode about how to do testing for new offers outside of CBO campaigns. I'll let you know when that's available.
I actually have a feeling that CBO might work with new offers if you willing to give Facebook more time and dollars to nail, who it's showing impressions to.
Molly had a compressed timeline because her client was doing a launch. She started with a $1,000 budget and was able to scale up to $10,000 in just a few days. Compare that to manual ad set budget scaling. She estimates it would have taken her a month to scale that much.
Back in the day if you tried to scale too quickly it would throw off the algorithm and Facebook’s impressions would get sloppy. The traffic wasn't as targeted. When they are in total control, as with CBO, it’s an entirely different story.
In fact, while scaling from $1,000/day to $10,000/day in a week, her average cost per lead only increased by .12. That’s 12 cents. Unheard of. Normally, with that much scale the leads get a lot more expensive, particularly when ramping up at that speed.
Facebook CBO Picks Winning Ad Sets Fast ... Maybe Too Fast
Molly discovered that Facebook started to favor three of her initial 27 ad sets within days. Those ad sets, subsequently, received most of the impressions and accounted for over half of the leads.
Ralph equated this with how Facebook treats ad sets with 20+ ads. It will only show between 2-6 of those ads and ignore the rest.
Not exactly a scientific split test result.
Ralph suggested fixing this issue by setting a minimum daily budget for each ad set to force Facebook to give each of them impressions. This way you can compare results and determine best performers.
You can find the minimum daily budget setting underneath budget>scheduling at the ad set level. Apparently, t’s hard to find. I haven't used it yet.
Applying a minimum daily budget to each ad set makes the process more cumbersome, but it enables you to reach the audiences that might be ignored otherwise.
Molly tested this idea on the ad sets that weren’t getting impressions in a separate campaign she was running. She said adding the minimum daily budgets to each ad set confused Facebook, and screwed up the overall performance.
She concluded you have to do the minimum daily budget at the launch of a CBO rather than midstream.
She also discovered that moving the ignored ad sets into a new CBO campaign worked.
This experiment started rough because she set her budget too high to start. But, once figured out she got good results from the previously non-performing ad sets.
And that was instrumental in helping her scale.
Now, when she wants to build new audiences she creates new campaigns with new ad sets, rather than building new ad sets in existing campaigns.
Use Campaign Budget Optimization For Scaling
Both Molly and Ralph re-iterated that they wouldn’t use CBO for new campaigns. Rather they would use it only if you have a seasoned pixel, social proof on the ads, and proven audiences.
However, they also said that if a non-CBO campaign is working very well, don't mess with it. Only migrate over if the performance starts to slip, the ads get stale, and/or the CPA’s start to rise significantly.
How to Reach Facebook Account Nirvana
Fewer ad sets and campaigns.
It’s easy to get lost when you have too much stuff going on. It’s too complicated to keep track of and manage. Especially if you’re doing everything manually. Set up your account simple enough to be able to make smart decisions at a glance.
“Looking at the campaign level for your results keeps you out of the weeds.” - Molly Pittman
The best accounts only have 3-5 campaigns. Each campaign has 3-5 ad sets. The ad sets should be about the same size because Facebook tends to give more impressions to the bigger ad sets.
Learning phase is usually 2 to 3 days. It’s less painful if you simplify the ad account as much as possible. Don’t confuse Facebook. Don’t give it too many variables.
But, don't set and forget. Things are always changing. Continuously test for what works and what doesn’t.
You Must Squish!
It used to be that to optimize your campaign for maximum results you had to run a metric ton of ad sets. Ralph still has clients with 1000s. But, those days are waning. You can build much bigger audiences with CBO by squishing more of the interests into one ad set.
“The days of super-extensive crazy split testing is not as necessary any more.” - Ralph Burns
Use less ad sets and larger audience sizes (15-30 million each).
“It’s easier. It takes less time.”
Molly and Ralph agree campaign budget optimization is a big breakthrough. It simplifies things.
It’s also great because you don’t have to worry about audience overlap. Facebook takes care of that within the campaign. Your ad sets won’t be in a bidding war to get impressions from the same audience.
This was a huge problem in the days of massive campaigns with hundreds or even thousands of ad sets.
So, get out there and build some CBO campaigns. They rock!
I'll leave you with Ralph's words of wisdom:
“What Facebook is doing within the audiences is finding the lowest cost conversions. They’re finding lowest cost conversions in the other ad sets that aren’t at high volume, but they have flexibility, because of campaign budget optimization, to not feel the need to spend your daily budget at the ad set level, which would then blow out your CPA’s.”
- I snagged the Facebook with and without CBO graphic from the great Jon Loomer's blog post on the subject.