Source
There is a huge amount of rumor and misconception going around when it comes to the Facebook learning phase.
Believe it or not, the learning phase is there to HELP your campaign (not to hurt it). It’s important to understand how the learning phase can help with your ads, how it works, and how you should approach it to give yourself the best chance at great Facebook advertising results.
In this guide I’m going to go over how we approach the learning phase in my Facebook ads agency and show you an example of how it can benefit you and how it works in real life.
You can see what Facebook says about the learning phase in it’s guide on the subject here.
Facebook uses a machine learning process to show your ad to the best people and places for your particular campaign in order to get you the very best results.
Remember that Facebook wants you to get good results because they want you to spend more money. It’s a win for them if they are delivering good results to you.
The learning phase is a process that can help you to get the lowest cost per action possible on Facebook.
A lot of Facebook advertisers fear the learning phase, but it’s actually a great thing because Facebook will take the information you provide on targeting, conversion events, and ad spend and optimize your campaign.
The learning phase will test out different pockets of people within your audience. It will test out the best times of day to run your ad. It will test out placements to find out if your ad works better on Facebook or Instagram or both.
During the learning phase Facebook is out on the platform tweaking your campaign to see what will bring in the most optimized results. That process can lead to quite a bit of volatility and higher prices at the beginning.
Even after your campaign leaves the learning phase, Facebook will continue to learn what works well, but the most intense period of learning is at the beginning of the campaign.
Here is an example of a traffic campaign that we did in my agency. It sent traffic to a specific blog post and it was a mixture of warm and cold audiences.
The green line on the bottom is the one that you want to pay attention to. That’s the line that shows us cost per result.
The first month is really interesting, because it takes nearly that long for the cost per click to settle down out of volatility and into a lower cost. In fact, after the first month the cost per click was about half of what it was at the beginning of the campaign. It went from 14 cents per click on many days to settling in around 7 cents per click.
Even after the learning phase is done you are going to see some variation in pricing for your conversion events, but overall the cost can dramatically settle down once the campaign gets out of learning phase.
This cost reduction and stability are why the learning phase is so important. When Facebook figures out the right combination to optimize your campaign then your costs can start to drop – sometimes pretty dramatically. The cost differential can be so big that it can be the difference between a losing campaign and a winning campaign.
But you have to be patient to get there.
Facebook says that the average learning phase is about 48 hours. That depends on how much you’re spending. If you’re not spending very much and you’re generating fewer results than the learning phase may end up being extended. If you have a massive budget then the learning phase can potentially be done in just a few hours.
With this campaign we were spending about 150 to 200 pounds a month. That’s a fairly low amount so the learning phase was extended and you see really volatile results over the first few weeks.
At the beginning of a campaign when it’s in the learning phase, you will not only see higher prices, but you will also see far more volatility.
You’ll notice that as the campaign goes on the price per click starts to settle down to a much more consistent range.
It’s often a good idea to test new audiences or creatives in your ad campaign. Testing and making adjustments can give you a much better performing campaign overall. But since making adjustments can put your campaign back into the learning phase a lot of people are afraid to make changes to their ad creatives or audiences – even when they should!
Don’t be afraid of the learning phase. You CAN make adjustments if you want to test out new things. Just don’t do it too frequently. In our campaigns we limit adjustments to about once a week and if we have multiple adjustments to do we do them all at once in a single session.
f you are spending a low amount of money then you may want to spread any adjustments that you need to make out further and adjust things no more frequently than every two weeks.
You do need to be aware that once you make adjustments your results might become erratic for a bit before they start to settle again. That’s normal.
You also want to give your campaigns a chance to get out of the learning phase after making adjustments in order to see what the effects of the adjustments actually are. Don’t rush it too much.
Once your campaign is out of the learning phase it can sometimes get stuck in the learning limited phase.
All that learning limited means is that your ad set isn’t getting enoughoptimization events. If you are optimizing for sales then you aren’t getting enough sales.
But I’ve also seen that learning limited isn’t always accurate. I’ve seen plenty of ads that are generating thousands of conversions at good prices and Facebook still says they are in the learning limited stage. That’s why the first thing I recommend is to see if the learning limited designation is even accurate. Sometimes it’s not.
And even it learning limited is accurate, I recommend that you don’t worry too much about it.
What I’ve seen is that many of the things you can do to quickly get out of learning limited can backfire and cause your cost per optimization to go up. For instance, you might triple your budget which gets you out of learning limited, but makes your cost per conversion jump as well.
In the big picture, being happy with the cost per conversion in your campaign is more important than getting out of learning limited.
There is a huge amount of rumor and misconception going around when it comes to the Facebook learning phase.
Believe it or not, the learning phase is there to HELP your campaign (not to hurt it). It’s important to understand how the learning phase can help with your ads, how it works, and how you should approach it to give yourself the best chance at great Facebook advertising results.
In this guide I’m going to go over how we approach the learning phase in my Facebook ads agency and show you an example of how it can benefit you and how it works in real life.
What Is The Facebook Learning Phase?
You can see what Facebook says about the learning phase in it’s guide on the subject here.
Facebook uses a machine learning process to show your ad to the best people and places for your particular campaign in order to get you the very best results.
Remember that Facebook wants you to get good results because they want you to spend more money. It’s a win for them if they are delivering good results to you.
The learning phase is a process that can help you to get the lowest cost per action possible on Facebook.
A lot of Facebook advertisers fear the learning phase, but it’s actually a great thing because Facebook will take the information you provide on targeting, conversion events, and ad spend and optimize your campaign.
The learning phase will test out different pockets of people within your audience. It will test out the best times of day to run your ad. It will test out placements to find out if your ad works better on Facebook or Instagram or both.
During the learning phase Facebook is out on the platform tweaking your campaign to see what will bring in the most optimized results. That process can lead to quite a bit of volatility and higher prices at the beginning.
Even after your campaign leaves the learning phase, Facebook will continue to learn what works well, but the most intense period of learning is at the beginning of the campaign.
Example of How the Facebook Learning Phase Works
Here is an example of a traffic campaign that we did in my agency. It sent traffic to a specific blog post and it was a mixture of warm and cold audiences.
The green line on the bottom is the one that you want to pay attention to. That’s the line that shows us cost per result.
The first month is really interesting, because it takes nearly that long for the cost per click to settle down out of volatility and into a lower cost. In fact, after the first month the cost per click was about half of what it was at the beginning of the campaign. It went from 14 cents per click on many days to settling in around 7 cents per click.
Even after the learning phase is done you are going to see some variation in pricing for your conversion events, but overall the cost can dramatically settle down once the campaign gets out of learning phase.
This cost reduction and stability are why the learning phase is so important. When Facebook figures out the right combination to optimize your campaign then your costs can start to drop – sometimes pretty dramatically. The cost differential can be so big that it can be the difference between a losing campaign and a winning campaign.
But you have to be patient to get there.
Facebook says that the average learning phase is about 48 hours. That depends on how much you’re spending. If you’re not spending very much and you’re generating fewer results than the learning phase may end up being extended. If you have a massive budget then the learning phase can potentially be done in just a few hours.
With this campaign we were spending about 150 to 200 pounds a month. That’s a fairly low amount so the learning phase was extended and you see really volatile results over the first few weeks.
At the beginning of a campaign when it’s in the learning phase, you will not only see higher prices, but you will also see far more volatility.
You’ll notice that as the campaign goes on the price per click starts to settle down to a much more consistent range.
What About Making Adjustments to Your Campaign and Putting it Back Into the Learning Phase?
It’s often a good idea to test new audiences or creatives in your ad campaign. Testing and making adjustments can give you a much better performing campaign overall. But since making adjustments can put your campaign back into the learning phase a lot of people are afraid to make changes to their ad creatives or audiences – even when they should!
Don’t be afraid of the learning phase. You CAN make adjustments if you want to test out new things. Just don’t do it too frequently. In our campaigns we limit adjustments to about once a week and if we have multiple adjustments to do we do them all at once in a single session.
f you are spending a low amount of money then you may want to spread any adjustments that you need to make out further and adjust things no more frequently than every two weeks.
You do need to be aware that once you make adjustments your results might become erratic for a bit before they start to settle again. That’s normal.
You also want to give your campaigns a chance to get out of the learning phase after making adjustments in order to see what the effects of the adjustments actually are. Don’t rush it too much.
What About Learning Limited?
Once your campaign is out of the learning phase it can sometimes get stuck in the learning limited phase.
All that learning limited means is that your ad set isn’t getting enoughoptimization events. If you are optimizing for sales then you aren’t getting enough sales.
But I’ve also seen that learning limited isn’t always accurate. I’ve seen plenty of ads that are generating thousands of conversions at good prices and Facebook still says they are in the learning limited stage. That’s why the first thing I recommend is to see if the learning limited designation is even accurate. Sometimes it’s not.
And even it learning limited is accurate, I recommend that you don’t worry too much about it.
What I’ve seen is that many of the things you can do to quickly get out of learning limited can backfire and cause your cost per optimization to go up. For instance, you might triple your budget which gets you out of learning limited, but makes your cost per conversion jump as well.
In the big picture, being happy with the cost per conversion in your campaign is more important than getting out of learning limited.
