If you have multiple ad sets that attempt to enter the same auction, the algorithm will choose the ad with the highest total value. This is Auction Overlap. The other ads won’t be considered.
This is most likely to happen if you have multiple ad sets targeting the same people. Auction Overlap is meant to prevent you from competing against yourself in the auction.
When is This a Problem?
On a small scale, this is unavoidable and won’t lead to noticeably bad results. But, it’s a problem if it happens often.
Auction Overlap is good for the surviving ad set because it won’t compete against another one of your ad sets in the auction. But the competing ad set that is repeatedly prevented from entering auctions might struggle to spend your budget. Or your budget will spend, but it will cost more to show your ads.
Too much auction overlap will make performance less stable and predictable, especially when scaling. It may make it more difficult to exit the Learning Phase. This, obviously, is not preferred.
What Should You Do?
First, when Auction Overlap becomes an issue, you should see Delivery Recommendations to combine ad sets. You should consider doing so when it makes sense.
Second, you can create an Automated Rule to either combine ad sets dynamically or alert you when Auction Overlap is detected.
The alert may be preferred to give you the option of combining ad sets or not.
Bottom line: Avoid Auction Overlap by limiting the number of competing ad sets that target the same people. This is far more common now because of Advantage+ Audience and Advantage audience expansion. Even if you don’t intend to create ad sets with overlapping audiences, they often will.
This is why I continue to recommend limiting the number of unnecessary ad sets. It’s what Meta wants, and it will often lead to better and more efficient results.