Facebook ads learning phase: the minimum budget to exit it
By Matheus Mello, founder of Ads Editor and owner of YEP Agência · Published · 5 min read
The Facebook ads learning phase ends when an ad set collects about 50 optimization events in 7 days, and that single number decides how much budget each ad set needs. Most learning limited problems are budget math problems in disguise.
Quick answer
The minimum daily budget for an ad set to exit the Facebook ads learning phase is your target CPA times 50, divided by 7, because Meta asks for about 50 optimization events in 7 days. With a $40 target CPA, that is about $286 a day per ad set. Without that budget, run fewer ad sets.
Summary
- Formula: minimum daily budget per ad set = target CPA x 50 / 7.
- The 50 events are per ad set, not per campaign.
- Without the budget, run fewer ad sets, optimize for a more frequent event or broaden the audience.
- A test floor of about 2x target CPA per day is a different rule, for reading creative signal.
- Google Ads bid strategies have their own learning period, measured in conversion cycles.
Where does the CPA x 50 / 7 formula come from?
To exit learning, an ad set needs about 50 optimization events within a 7 day window. If each event costs your target CPA, you have to spend 50 CPAs in 7 days. Divide by 7 and you have the daily budget.
The formula
Minimum daily budget per ad set = target CPA x 50 / 7
The detail that trips most people: the 50 events are per ad set. A campaign with three ad sets needs 50 in each, not 50 combined.
Meta has tested a lighter threshold. Jon Loomer reported in July 2024 that for a few weeks his accounts showed a learning phase of 10 optimized events in three days, then went back to 50 events in seven days without explanation, and that Meta's documentation still said 50. Plan with 50 until your own Ads Manager says otherwise.
What is the minimum budget for each target CPA?
| Target CPA | Budget over 7 days | Minimum daily budget |
|---|---|---|
| $10 | $500 | $71 |
| $20 | $1,000 | $143 |
| $40 | $2,000 | $286 |
| $80 | $4,000 | $571 |
| $150 | $7,500 | $1,071 |
What if the client does not have that budget?
- Run fewer ad sets. With budget for one ad set, run one ad set.
- Optimize for a more frequent event. If purchases are rare, optimizing for add to cart or initiate checkout reaches 50 events faster, at the cost of optimizing for a less qualified signal.
- Broaden the audience. An audience that is too small limits how many people can complete the event.
- Accept learning limited knowingly. Sometimes it is what is possible; just do not expect stable costs.
Is the 2x CPA test floor the same as CPA x 50 / 7?
No, and mixing them up is the most common mistake. The floor of about 2x target CPA per day is the minimum per ad set in ABO for a test to produce signal: below it, the ad set does not even spend the equivalent of one conversion a day, and the read is worthless. CPA x 50 / 7 is what an ad set needs to exit learning and stabilize cost. A test ad set can live at the floor for a few days; an ad set you want stable needs the formula. How to structure the test itself is in Facebook ads creative testing.
Illustrative scenario (not a real client)
An ecommerce brand with a $40 target CPA wants to test 4 audiences with $300 a day in total. The math asks for $286 per ad set, or $1,144 for all four. The right call is testing 1 audience at a time in ABO, or optimizing for initiate checkout, which costs less, until there is volume.
Does a budget increase get you out of learning?
It helps when budget was the bottleneck, but a large increase is a significant edit and can restart the phase. Raise in steps of up to 35%, which can be repeated in sequence: the method is in how to scale Facebook ads. Structure matters too: CBO spreads budget across ad sets, which is why it needs volume first (see CBO vs ABO).
Does Google Ads have the same learning math?
Similar idea, different unit. Google Ads Help says it can take a few conversion cycles (1 to 2 typically) for a bid strategy to calibrate, and that a new strategy, a setting change or a composition change (adding or removing campaigns, ad groups or keywords) triggers the Learning status. Changing targets every week keeps a campaign learning forever. The routine is in how to manage Google Ads for multiple clients.
Frequently asked questions
How many events does Meta need to exit the learning phase?
About 50 optimization events in 7 days, per ad set.
What is the minimum budget per ad set on Meta?
To exit learning, target CPA x 50 / 7 per day. With a $40 CPA, about $286 a day.
What does learning limited mean?
It is the status of an ad set that cannot collect enough events to exit learning, usually because the budget or the audience is too small.
How long does the Facebook ads learning phase last?
As long as it takes to collect about 50 optimization events, within a 7 day window. With enough budget, a few days; without it, the ad set can stay learning limited.
Glossary
- Optimization event
- The action the ad set is optimized for, such as a purchase or a lead.
- Learning limited
- The status of an ad set unlikely to get enough optimization events to exit learning.
- Target CPA
- The cost per acquisition you are aiming for, used to size budgets.
- Conversion cycle
- The time between an ad click and the conversion it produces.
References
- Is the Learning Phase 10 or 50 Events? (Jon Loomer). Supports: Post of July 2024: his accounts briefly showed 10 optimized events in three days, then returned to 50 optimized events in seven days, and Meta's documentation still reflected 50 events in seven days.
- Facebook Ads Edits that Trigger the Learning Phase (Jon Loomer). Supports: Quotes Facebook: during the learning phase performance is less stable and CPA is usually worse; it occurs on new ads or ad sets and after significant edits.
- Duration of the learning period for campaigns and what affects it (Google Ads Help). Supports: It can take a few conversion cycles (1 to 2 typically) for the bid strategy to calibrate; new strategy, setting change and composition change trigger the Learning status.
Links opened and checked when the article was last updated.
About the author
Matheus Mello, founder of Ads Editor and owner of YEP Agência. Runs client ad accounts at YEP Agência and built Ads Editor so his own agency could stop publishing ads one at a time. Usage numbers quoted on the blog come from the product activity log. Instagram: @theusm
