Optimizing Campaigns Across Meta.

Building a Solid Target Audience

To start off strong, let’s talk audience. A well-defined audience is one pillar of a successful campaign. If your campaign requires a custom or saved audience, Meta recommends a group of at least 100,000 people to achieve optimal results. While smaller audiences may work, they can lead to higher costs and extended or limited learning periods in the long run. 

When uploading custom lists, ensure they have at least 1,000 members, but ideally 3,000-5,000. You can also exclude specific locations to further refine your targeting. However, try not to get too granular, as this may limit Meta’s ability to optimize your ads.

Leveraging Meta Advantage+ Audience

Meta has tons of tools to help supplement performance. These are known as their Advantage+ offerings. After testing out some of their capabilities, we have found the most success with Meta’s Advantage+ Audience—which can be super helpful for optimizing campaigns with small or niche audiences. By allowing Meta’s AI to learn from user data and actions, the platform can expand a campaign’s audience based on pixel or Conversion API data and achieve significant improvements in performance. According to Meta’s research, Advantage+ Audience has been shown to deliver:

• 13% lower median cost per product catalog sale
• 7% lower median cost per website conversion
• 28% lower average cost per click, lead, or landing page view

It’s important to note that Advantage+ Audience may not be suitable for all campaigns. Meta doesn’t recommend this setting for custom audiences or retargeting if the audiences are highly specific, as they may not benefit from AI-driven optimization and expansion. However, in our tests we have seen campaign efficiencies come from enabling this feature even with custom and retargeting audiences. 

Navigating the Learning Phase

When having Advantage+ Audiences enabled, we have also seen a decrease in the time ad sets and ads spend in the learning phase. What’s that you might ask? When a new ad set or ad is created, or significant edits are made to an existing one, it enters the algorithmic learning phase. During this period, Meta’s AI is learning what works and what doesn’t. Typically, ad sets exit this phase after around 50 optimization events. To ensure a smooth learning phase:

• We avoid unnecessary edits, as they can reset the learning process
• Recommend 3-5 ad sets with 2-5 ads each
• Distribute budgets across ad sets to support at least 50 optimization events

Overcoming Learning Plateaus

Sometimes, ad sets don’t exit the learning phase. This happens when they are unable to reach 50 optimization events within 7 days of the last significant edit, and they go into limited learning. When this happens, it’s time to reach out to us—or all jokes aside—test some of the following solutions:

• Combining ad sets and campaigns to aggregate learning and boost performance
• Expanding the audience to increase ad visibility (this could mean enabling advantage+ audiences)
• Adjusting budgets and bids to encourage optimization events
• Changing optimization events to more frequent actions, such as moving from landing-page views to link clicks

Managing Creative Fatigue

Creative assets are another key factor to a successful campaign. If there are too few asset versions, creative fatigue can occur—meaning your audience is exposed to the same ad too many times. Meta recommends keeping frequency to 2, but we recommend adjusting this based on performance. To keep your creatives fresh:

• Introduce new ad creatives to maintain audience engagement
• Expand your audience to mitigate fatigue and boost performance
• Include two creative versions for each audience group you want to reach

Creative fatigue can also be avoided by adding anywhere from 2-5 creative versions per ad set. This allows Meta to pick the highest performing creatives and optimize efficiently. 

Let’s Sum It Up

Meta can be tricky to navigate as there are always automatically enabled updates that might work against your performance or brand standards. To start off on the right foot, remember that Meta optimizes best when audience size is closer to 100,000+, in-platform tools like advantage+ audiences should be used strategically (not all advantage+ tools are created equal), and creative versioning for different audiences should be refreshed depending on frequency to avoid fatigue. 

And remember, we are here to help! Feel free to reach out to info@ciceron.com with any questions about Meta or questions regarding a Meta campaign.