Tired of watching your Meta ad budget vanish into the black hole of the learning phase? You’re not alone. Every advertiser faces this sticky situation during their advertising campaigns — performance dips and costs rise as Meta’s algorithm searches for the perfect audience. But what if you could accelerate this process even slightly? Or, at least make it so that it doesn’t drag on.
We want to help you do just that, so let’s see exactly what this learning phase is all about. Look for the sprinkle of cyberlicious® tips as well.
What is Meta’s Learning Phase?
Meta’s learning phase is a critical period when the algorithm is tasting all sorts of combinations to find the perfect flavor for your ad. During this time, Meta experiments with different placements, audiences, and bidding strategies to optimize your ad’s performance. While this process is a bit of an annoyance, it will actually help in the long-term with crafting a deliciously effective campaign.
Exiting the Sticky Situation That is Meta’s Learning Phase
To maximize your ROAS and the reach of your ads, you will want to expedite the learning phase and transition to optimization. Generally, Meta craves around 50 optimization events (like clicks, purchases, or leads) within a seven-day window to satisfy its sweet tooth and exit this phase.
There are three key factors that influence the learning phase duration:
- Ad set volume
- Ad set edits
- Conversion volume
Ad Set Volume: Too Many Flavors
The number of ad sets you run simultaneously can have an impact on the length of the learning phase. When you run multiple ad sets, each set receives fewer impressions as the system rotates through them to gather data. This limited exposure can slow down the learning process. Additionally, spreading your budget across too many ad sets can lead to sour performance, as individual sets may not receive enough funding to generate sufficient data for optimization.
Cyberlicious® sweet tip: Start with a smaller number of ad sets by consolidating them to gather data more quickly. Ad sets with overlapping targeting or objectives are great choices for doing this. Once you have enough data you can add more sets if you’d like, but keep in mind, they will likely reenter the learning phase.
Avoid Frequent Edits & Keep it Sweet
One of the most common mistakes advertisers make is making frequent changes to their campaigns. We get it. The temptation to change something that doesn’t initially seem to be working is hard to fight. Fight it though; these edits can disrupt the learning phase, delaying optimization and potentially leading to a less-than-sweet performance.
Here is a full list of changes that Meta specifically notes that typically reset the learning phase:
- Campaign-level changes:
- Budget adjustments (depending on the magnitude)
- Bid amount modifications
- Bid strategy alterations
- Ad-level changes:
- Any modifications to ad creatives
- Ad set-level changes:
- Targeting adjustments
- Placement alterations
- Optimization event changes
- Adding new creatives
- Bid strategy, budget, or bid amount changes
- Pausing an ad set for more than seven days
In the end, minimizing unnecessary edits and allowing the learning phase to run its course, will increase the likelihood of reaching delicious campaign performance.
Find Delicious Conversion Volume
We already noted above that Meta wants 50 optimization events within a seven-day period for an ad set to exit the learning phase. The longer it takes to reach those events, the longer this phase will take. Reasons your conversion volume may be low include:
- Low budget: A burnt budget can limit the number of ad impressions and, consequently, conversions. Ensure your budget is sufficient to achieve the necessary conversion volume.
- Low bid or cost cap: Restrictive bidding strategies can prevent your ad set from reaching the required number of conversions. Consider adjusting your bid or cost cap to allow for more ad deliveries.
- Small audience size: A limited audience pool can restrict the potential for conversions as your ads reach fewer people. Expand your target audience if possible.
- Infrequent conversion event: If your desired conversion event occurs infrequently, consider optimizing for a higher-funnel event (e.g., add to cart instead of purchase) to accelerate the learning phase.
Note: Ad sets with longer conversion windows (e.g., offline conversions) may require more time to exit the learning phase.
Teaching Meta AI Customer Behavior and Advance Planning
To maximize the effectiveness of your Meta advertising campaigns, you will want to recognize that you are working alongside AI. And guess what? AI doesn’t just magically know your audience and their behaviors. They need to be taught. Providing them with valuable data can help you do this and influence their algorithms to better serve your target audience.
- Teach the platform: Provide detailed information about your customers, including demographics, interests, and behaviors. For instance, if you have a loyal customer base of fitness enthusiasts, create a custom audience based on their email addresses or purchase history. This helps Meta’s AI refine its targeting and deliver your ads to similar audiences.
- Plan ahead: Consider your campaign goals and objectives well in advance. For example, if you are going to run a campaign to advertise a sale during a specific date range, you will want to also consider the time that campaign may in the learning phase. This will help you optimize your ad performance.
Overall, being proactive in teaching the platform and planning your campaigns can improve your chances of achieving your desired outcomes.
Advanced Tips for Ad Optimization
Once your ad has successfully exited the learning phase and established a baseline performance, you can start implementing more advanced optimization strategies. Don’t forget that making significant changes can reset the learning phase, so proceed with caution.
- Test different ad creatives and targeting options: Experiment with variations in your ad copy, images, and targeting parameters to identify high-performing combinations.
- Utilize A/B testing: Create multiple ad versions to compare performance and make data-driven decisions.
- Consider bid adjustments: Adjust your bids based on factors like device, time of day, or audience demographics to optimize your ad spend.
- Utilize ad scheduling: Optimize ad delivery by scheduling your ads to run during peak performance times.
- Leverage audience insights: Deepen your understanding of your target audience by analyzing demographics, interests, and behaviors. Use this information to create highly targeted ad sets.
- Track and analyze performance metrics: Continuously monitor key performance indicators (KPIs) and make data-driven adjustments to your campaigns.
Important: When testing new elements or making significant changes, consider creating separate ad sets to isolate their impact and avoid disrupting the performance of your existing, optimized ad sets.
The Final Bite
To get the most out of your ads, Meta’s learning phase is nonnegotiable. So, you might as well learn how to make the situation sweet, not sticky. When it comes down to it, optimization is an ongoing process. Don’t be afraid to reenter the learning phase as your campaigns should never be set and forgotten. There is always room for improvement. People change and grow, which means your audience does the same.
Tired of watching your ad budget melt away without results? Let the cyberlicious® experts handle the sweet and sour aspects of your campaigns. From crafting compelling ad copy to optimizing for maximum ROI, we’ve got you covered.