A popular scaling strategy involves duplicating an ad set and setting a higher budget for the duplicate. The approach has reasonable motivations.
The thought is that this won’t restart learning on the original ad set (which could mean losing the performance you once had for good), and if the duplicate doesn’t work you can just shut it off.
That said, ad sets with identical targeting and optimization could also hurt performance due to auction overlap — which is why creating unnecessary competing ad sets isn’t recommended.
Meta also allows higher budget increases without restarting learning these days, even recommending specific safe increases in some cases.
There’s also an argument that it may be better to continue the ad set that was working rather than trying to start over.
I get why advertisers do this, but it may be an outdated strategy. I just don’t know that it’s necessary, and it’s questionable whether it is beneficial.
If you’re curious, I don’t duplicate ad sets to scale, choosing instead to increase the budget of the ad set that’s working. If I’m at all worried about restarting learning (which can be a real concern), I’ll increase the budget slowly.
That’s what I do, and I don’t think it’s necessary to duplicate ad sets for the purpose of scaling. Still, if you’ve found that duplicating ad sets is beneficial, do what works for you.
As always, there are few absolutes with this stuff.