Meta Advantage+ Shopping Campaigns: When To Use Them, When Not To

Every few weeks someone sends me a Meta account where Advantage+ has quietly swallowed most of the budget and the ROAS in the dashboard looks superb. Then we pause it in a few regions for a fortnight, watch total revenue barely move, and the conversation changes.

Here is the short answer. Run Advantage+ when you have enough weekly conversions to feed it, a deep pool of creative, and genuine new demand to go and find. Run a manual build when your budget is thin, your buyer is genuinely narrow, or you need a clean read on which creative won. Most accounts I take over need both, and the argument is only ever about the split. Below: what actually changed with the format this year, the cases where each one wins, how to structure a fair test, and how to judge the result without fooling yourself.

Advantage+ shopping is now Advantage+ sales

Start here, because half the guides you will find on this are describing a product that no longer exists in that shape.

Meta renamed Advantage+ shopping campaigns to Advantage+ sales campaigns to reflect the wider range of advertisers using the format, and the rename came with real structural changes:

• Ad sets are back. Advantage+ shopping folded the ad set into the campaign and gave you one. Advantage+ sales lets you build multiple ad sets, so you can split by offer, product line or audience.
• The existing customer budget cap is gone. That slider capping the percentage of spend going to existing customers has been removed. You rebuild the same control with separate ad sets and exclusions instead.
• You get audience inputs. Advantage+ audience takes suggestions and, crucially, custom audience exclusions. The old format’s total absence of exclusion control was its biggest single flaw.

If you are hunting for a budget cap slider from a 2023 tutorial, stop looking. The job it did now belongs to your ad set structure.

What you actually hand over

Worth being precise about this, because “automated campaign” gets used loosely.

You keep control of the objective, the budget, the country and region, the conversion event you optimise to, the creative you upload, and your exclusions. Meta takes the audience within those bounds, the placements, which creative combination goes to whom, and how budget moves between them.

That trade is good value when Meta has enough signal to make those decisions better than you can. It is poor value when it doesn’t, and that is the whole decision.

Where Advantage+ genuinely wins

Broad prospecting on a proven product. If your buyer is “anyone in the UK who might want this”, interest targeting is you guessing at a signal Meta already has. Hand it over.

Accounts drowning in fragmented learning. Meta’s guidance is around 50 optimisation events per ad set per week to get out of the learning phase and deliver stably. Do the arithmetic on your own numbers. At a £30 CPA that is £1,500 a week, roughly £215 a day, to feed one ad set properly. An account running eight ad sets on £300 a day is starving all of them. Consolidating into Advantage+ fixes that in a week.

Deep creative libraries. The format is built to test many combinations at once. If you can keep 15 to 30 live concepts in rotation, it will find angles you would not have paired manually. If you have four ads, you are not giving it anything to work with.

Small teams. One well-fed campaign that you check twice a week beats twelve ad sets nobody has time to read.

Where a manual build still beats it

Budgets that cannot feed a single ad set. Below the threshold above, automation has nothing to optimise on and you get expensive noise. Build one tight ABO campaign and earn your way up.

Genuinely narrow buyers. B2B with a defined job title, regulated categories, high-ticket products with a small addressable market, tightly geo-bound services. Broad targeting on a 2,000-person market burns money reaching the other 60 million.

Creative testing. This is the one people concede too easily. Advantage+ optimises for outcome, not for a clean answer, so it will not tell you reliably why something won. When I need to know which hook works, I test it in ABO and graduate the winner. I have written up that sequence in testing and scaling Meta ads, ABO first and CBO later.

Heavy existing-customer overlap. A strong brand with a big returning-customer base will see Advantage+ hoover up people who were going to buy regardless. Exclude your purchasers, or split them into their own ad set, before you judge anything.

Weak lead quality signal. If you are optimising to raw form fills and half of them are junk, automation scales the junk faster than a manual build would.

How to structure a test that proves something

Most Advantage+ tests fail because they were never fair. Set it up like this.

Element	How to handle it
Audience overlap	Exclude recent purchasers from Advantage+, or you are grading it on retargeting
Creative	Same core concepts in both, so you are testing the structure and not the ads
Budget	Comparable daily spend, each side able to clear the learning threshold
Duration	Four weeks minimum, and never judge inside the first week
Edits	Change nothing mid-flight. Every meaningful edit restarts learning

Then split your account rather than betting it. Put the broad prospecting spend into Advantage+, hold your testing budget in ABO, and keep any narrow high-intent segment separate. That structure is the same principle I use across the whole account, which I set out in the Meta ad account structure that scales profitably.

Read the result outside the platform

This is the part that decides the argument, and it is the part most people skip.

Advantage+ tends to report well because it is efficient at claiming credit for conversions, particularly from people already close to buying. That is not the same as creating sales. Two checks I would run before moving serious budget:

1. Blended numbers. Track total revenue against total ad spend across the test, not the in-platform ROAS. If the campaign level ROAS climbs while blended stays flat, you have shifted attribution, not performance.
2. A geo holdout. Pause the structure in a set of matched regions, leave it running everywhere else, and compare the change in actual revenue. It is the closest thing to a straight answer you will get, and it costs you a fortnight.

Rerun that check once a quarter, and any time you make a large structural change. Between those, a Monday review of spend distribution and creative fatigue is enough. If you want a working cadence, see our Meta ads optimisation routine.

Not sure which side your account sits on

If you are staring at an Advantage+ campaign wondering whether it is genuinely driving growth or quietly taking credit for it, that is exactly the question a free account audit answers. We will look at your structure, your exclusions, your conversion signal and your blended numbers, and tell you plainly what to change. If you would rather we ran it, that is what our Meta ads management does day to day.

FAQ

Are Advantage+ shopping campaigns still available? Not as a new build. Meta renamed the format to Advantage+ sales campaigns and moved everyone onto that setup. The automation works on the same principle, but you now get multiple ad sets and custom audience exclusions, and the existing customer budget cap has been removed.

What budget do I need for Advantage+ to work? Enough to clear the learning phase, which Meta puts at roughly 50 optimisation events per ad set per week. Multiply your current cost per conversion by 50 and divide by seven for a daily figure. If that number is well above what you can spend, run a single manual ad set instead until your volume supports it.

How do I stop Advantage+ spending on existing customers? Use custom audience exclusions at ad set level. Build a custom audience of recent purchasers and exclude it, or split existing and new customers into separate ad sets so you can see and control the split yourself. The old percentage cap no longer exists.

Should I run Advantage+ and manual campaigns at the same time? Yes, and most accounts should. Use Advantage+ for broad prospecting where Meta’s signal beats your guesswork, and keep a manual ABO campaign for creative testing and any narrow, high-intent segment. Set exclusions so they are not bidding into the same people.

How long before I judge an Advantage+ campaign? Four weeks, and do not touch it in between. The first week is learning and the numbers will look alarming. Editing budget, creative or audience mid-flight resets that clock and you end up judging a campaign that never stabilised.