Facebook’s Learning Phase: What It Is and How It Works

Unstable results during the learning phase is something that is familiar to everyone. But what can you do about it? To answer this question, let’s look into what learning phase actually is.

When you create a new ad set or make a significant change to an existing ad or ad set, Facebook’s system starts learning how to optimize for your results. It looks for audiences, time, placements, and creatives that would perform best based on your optimization goals.

Your ads sets will quit the learning phase as soon as the performance stabilizes. Usually it happens after it gets about 50 conversions within a week. So make sure you have enough budget to get these 50 conversions not to re-enter the learning phase. Except for budget, there are other factors that can throw your ad sets back. They can be on

-? ? ? ? ? campaign level:

• ??changes of bid amount
• or bid strategy

-? ? ? ? ? ad set level:

• ??optimization event
• ??targeting
• ??placement
• ??adding new creative
• ??pausing for over seven days

-? ? ? ? ? ad level

• ??anything

The best way is not to make edits during the learning phase. But if you really need to, try to batch them all. This way you’ll make sure it’s only reset one time.

Okay, everything seems to be clear with the learning phase. But what about learning limited? It means that your ad sets couldn’t exit the learning phase because of one (or several) of the following:

-? ? ? ? ? too many ad sets

-? ? ? ? ? very small audience

-? ? ? ? ? insufficient budget

-? ? ? ? ? too low bid/cost control

It may also occur if some other ad sets perform better. In this case, try combining them into fewer larger ad sets.

To sum it up, although the learning phase can cause unstable results, it’s an important stage because it allows the system to optimize and deliver your ads in the best way possible.

Got any questions? Feel free to reach out!