Optimization Events

Add to Cart Optimization

By the AdFlint research team · Last reviewed November 2026

Aims delivery at shoppers likely to put an item in the cart, a mid-funnel proxy used when purchase volume is too thin.

Carts happen several times more often than purchases, so this event gives the algorithm enough weekly signal to leave learning on a small budget. Reach for it during launch, after a pixel reset, or on low-volume ad sets. The misunderstanding is assuming cart-optimized traffic converts at the same rate: Meta will find serial cart-fillers, so watch purchase counts, not cart counts, and move up as soon as sales allow.

Key takeaways

  • Add to Cart optimization trains Meta's delivery on people who add a product to a cart, which fires more often than a purchase and therefore gives small ad sets enough weekly results to leave the learning phase.
  • The event only works if the pixel or Conversions API fires it exactly once per add, so check Events Manager before trusting the results.
  • A cost per result goal on this event means cost per cart add, not cost per sale, so judge the campaign by cost per purchase and cart-to-order rate as well.
  • Treat it as a temporary step: once purchases reach a steady weekly volume, test Purchase optimization against it.

In practice.

Add to Cart optimization is a choice you make in the ad set of a Meta campaign: you tell the delivery system that the result you want is the AddToCart event, not a purchase or a click. The event itself comes from your Meta Pixel or Conversions API setup. Meta's pixel reference says AddToCart fires when a product is added to the shopping cart, and its typical trigger is a person clicking an add to cart button. In Ads Manager you pick it by choosing a conversion location (usually your website), then selecting the event from the list of events your pixel has received. If the event has never fired, it will not appear as an option.

Under the hood, Meta's system looks at the people who recently triggered that event on your site and learns what they have in common: behavior on Facebook and Instagram, device, time of day, past shopping activity. It then bids in each auction for people who look like them. The reason advertisers choose this event is volume. Meta's help center says an ad set usually exits the learning phase after about 50 results in the week after its last significant edit. Carts happen more often than orders, so a small store can reach that count with AddToCart when it cannot with Purchase. The trade-off is that you are now buying cart adds, and the system does not know which carts turn into orders.

This event sits one step above View Content and one step below Initiate Checkout and Purchase on the intent ladder. It works with any budget setting, but it interacts most with your bid strategy. With Highest volume, Meta spends the budget to get as many cart adds as it can. With a cost per result goal, the goal now refers to cost per cart add, so a $8 goal means $8 per cart, not $8 per sale. Meta's own bid strategy page recommends 50 to 100 or more weekly conversions for goal-based strategies, so a cart goal is easier to support than a purchase goal on a small account. Attribution settings also apply: carts are counted inside the window you choose.

The common mistake is leaving the ad set on Add to Cart long after it stops being necessary. Cart-heavy audiences include people who browse, fill a cart to see shipping, and never buy, and the system will happily find more of them if that is what you reward. A second mistake is firing the event twice per add, or on page load of the cart page, which inflates results and teaches the system the wrong behavior. Check in Events Manager that one add produces one event. A third mistake is judging the campaign on cost per cart add alone. A cheap cart is only good if the cart-to-order rate on your site stays healthy.

This week, open the ad set and compare three numbers over the same dates: spend, cart adds, and purchases from your store or Ads Manager. Divide spend by each to get cost per cart and cost per purchase, and divide purchases by carts to get the cart-to-order rate. If purchases are coming in at roughly 30 to 50 a week, test a duplicate ad set optimized for Purchase and let both run. If purchases are still in single digits, keep Add to Cart, avoid edits that reset learning, and focus on the checkout and the offer, since those decide whether the carts you are paying for ever close.

Worked example

A small store deciding when to switch from carts to purchases

Suppose a candle shop spends $60 a day on one ad set, which is $420 a week. Optimized for Purchase, the shop gets about 1.2 orders a day at a $50 cost per purchase, roughly 8 orders a week. That is far short of the 50 results Meta's help center gives as the usual benchmark for leaving learning, so the ad set shows Learning limited. The owner switches the optimization event to Add to Cart. Carts now come in at about $6 each, which is 70 carts a week.

The owner then checks the store: 70 carts produced 6 orders, a cart-to-order rate of about 8.6 percent, so cost per purchase is $420 divided by 6, or $70. That looks worse than $50, but the Purchase ad set was unstable and delivered unevenly. The owner keeps both numbers on a sheet. When the carts ad set reaches 25 or more orders a week for three weeks in a row, the owner will duplicate it, switch the copy to Purchase, and compare the two on cost per purchase before moving the full budget.

Add to Cart Optimization compared with

The settings this gets confused with, and how to tell them apart.

Common questions.

Is Add to Cart a good optimization event for Meta ads?

It is a good fallback when purchases are too rare to give the ad set enough weekly results. It is a weaker choice once you have steady sales, because Meta will keep finding people who fill carts without buying.

Why does my Add to Cart event not show up in Ads Manager?

Meta only lists events your pixel or Conversions API has already received. Fire the event on your site, confirm it in Events Manager, and then it will appear in the optimization event list.

Does changing the optimization event reset the learning phase?

Meta treats changes to the optimization event as significant edits that can send an ad set back into learning. Make the switch once, on purpose, and then leave the ad set alone for about a week.

Should I optimize for Add to Cart or Initiate Checkout?

Initiate Checkout carries stronger intent but fires less often. If your ad set can reach roughly 50 weekly results on Initiate Checkout, it is usually the better step up from carts, and if not, stay on Add to Cart.

Sources

You should not need to know this to advertise.

AdFlint handles the settings for you, inside the Google and Meta accounts you already own.

Next step: How AdFlint tracks and optimizes these numbers

Try AdFlint free
All terms and comparisons