Meta ad set stuck in Learning limited

Why it happens

The ad set gets fewer than about 50 results a week. Meta's help says an ad set exits the learning phase once it can deliver stably, which "usually occurs after about 50 results in the week after the ad set's last significant edit". An ad set "becomes learning limited when it is unlikely to receive about 50 optimization events" in that week. Meta adds that this status is not a penalty: it means the delivery system cannot optimise with the current setup.

Meta does not see all your purchases. If the ad set optimises for Purchase, only purchases that reach Meta count toward those 50. A store with 70 sales a week from Meta ads can still be Learning limited if the pixel reports 40 of them, because the browser never reported the rest: ad blockers, consent choices, or a buyer who left before the thank you page loaded. Meta shows fewer purchases than Shopify covers why.

Purchases arrive without enough detail. Meta has to tie each purchase to a person who saw the ad. A purchase that reaches Meta with few matching details is harder to match and may not count as a result for that ad set. See low Event Match Quality in Meta.

Too many ad sets share too few sales. Meta lists the usual limits: small audience, low budget, low bid or cost control, high auction overlap, an infrequent optimisation event, or too many ads running at the same time. Ten ad sets that share 60 purchases a week each get 6. Meta's help says that combining similar ad sets also combines what they learn.

Edits keep resetting the clock. The 50 results are counted from the last significant edit. Meta's help says editing an ad, ad set or campaign during the learning phase resets learning, and that frequent budget changes can cause an ad set to re-enter it.

The optimisation event is rare. For a store with a high price and few orders, purchases may never reach 50 a week per ad set. Meta suggests choosing an event that happens more often, for example add to cart instead of purchase.

Check it yourself

1. 1In Ads Manager, open Columns, then Customize columns, and tick Optimization events. Meta's help shows this as the way to see whether an ad set gets enough of them.
2. 2Add the column Last significant edit. Count the results since that date, not since the start of the campaign.
3. 3In Events Manager, read the Purchase count for the last 7 days on your dataset's Overview tab. Compare it with your Shopify paid online orders for the same days.
4. 4If Events Manager shows far fewer purchases than Shopify, fix that first. A higher budget does not help an ad set learn from sales it cannot see.
5. 5Look at Event Match Quality for Purchase in Events Manager.
6. 6Count your ad sets that optimise for Purchase and add up their weekly results. If the total is above 50 but each ad set is far below, combining them is Meta's first suggestion.
7. 7Check whether the Facebook & Instagram app, a theme tag and another app all send the purchase. Meta removes copies it can match by event ID. Copies it cannot match count twice, and then your results say more than you sold. See Meta counts purchases twice.

Run the free diagnosis

FAQ

Is Learning limited bad for my ads?

Meta says it is not a penalty. It means the delivery system has less to learn from, so results are less stable and costs often higher.

Should I switch the optimisation event to add to cart?

Meta suggests it when purchases are too rare. Do it only after you know purchases are reaching Meta. If they are missing, an earlier event hides the problem instead of solving it.

Will raising the budget fix it?

Sometimes. Meta lists a budget that is too low as one cause. But if your budget already buys enough sales and Meta does not see them, more budget only costs more.

How long until the ad set leaves Learning limited?

Meta says an ad set moves from Learning limited to Active once it gets enough optimisation events since the last significant edit.

Does a new ad set always start in learning?

Yes. Meta's help says the delivery system is in the learning phase when you create a new ad or ad set, or make a significant edit to one. Learning limited is only the case where it is unlikely to get enough results to finish.

Does server-side tracking help with learning?

It can raise the number of purchases Meta receives, when the browser pixel misses some. Meta still needs to tie each purchase to an ad before it counts as a result.