About ad delivery

The Meta ad delivery system uses an ad auction and machine learning to determine where, when and to whom we show your ads. These processes work together to maximise value for both people and businesses.

Each time there's an opportunity to show an ad to someone, an ad auction takes place to determine which ad to show.

As we want each person to see relevant ads, the ad auction considers predictions of each ad's relevance for the particular person. This means that ads with higher relevance can win ad auctions at lower costs.

Relevance predictions estimate how likely the person is to consider the ad high quality and take the advertiser's desired action. You can use ad relevance diagnostics to understand if the ads you ran were relevant to the audience you reached.

As we want each advertiser to meet their goals at the lowest cost, the delivery system uses machine learning to improve each ad's performance. Performance is optimised to minimise each ad's cost per optimisation event.

Each time an ad is shown, the delivery system's predictions of relevance become more accurate. As a result, the more an ad is shown, the better the delivery system becomes at determining where, when and who to show the ad to minimise the cost per optimisation event. We call the period where the delivery system still has a lot to learn about an ad the learning phase.

If you have multiple ads in your ad set (or across your account), we'll show the ad that's most likely to achieve the lowest cost per optimisation event for the given person. This means that each of your ads won't necessarily be delivered the same number of times.

Sometimes, the ad or ad set that gets the most results isn't the ad set that received the lowest cost per optimisation event. That's because the ad delivery system uses predictions of future performance to determine where to deliver next – not each ad set's past performance. For this reason, we often recommend using Meta Advantage+ placements (formerly known as automatic placements) and broad audiences to provide the delivery system with more flexibility to find lower cost conversions.

To troubleshoot ad delivery, see here.