Advantage+ Campaigns: When They Work, When They Waste Budget and How to Stay in Control

A founder in Manchester messaged us last month with a screenshot and one line: "Meta says 4.1 ROAS, Shopify says nothing changed." He'd moved every sales campaign into Advantage+ in spring. Reported purchases climbed. Revenue on the store didn't. When we switched on the audience segment breakdown, close to four in ten pounds were going to people who had already bought from him.

That's the most common Advantage+ campaigns complaint we hear heading into 2027, and it isn't a bug. Meta's automation does what it's asked: find the cheapest conversions it can attribute. Past customers and people about to buy anyway are cheap conversions. This guide covers what has changed in the product, what the independent data says, how to diagnose leaks with real Ads Manager menus, and how to keep the parts of automation that work.

What changed with Meta's Advantage+ campaigns?

If you learned Advantage+ as "ASC", the shopping campaign with almost no settings, forget that mental model. On 29 May 2025 Meta began rolling out a single setup for sales and app campaigns, and its developer blog of 3 June 2025 spelled it out: advertisers no longer choose between Advantage+ shopping and manual. A campaign simply shows as Advantage+ on or off.

According to Meta's Marketing API documentation, three levers decide that state:

• Budget: Advantage+ campaign budget set at campaign level, with lowest cost, cost cap, bid cap or minimum ROAS bidding.
• Audience: Advantage+ audience used on at least one ad set.
• Placements: no placement exclusions at ad set level. Account-level placement controls are allowed.

Turn any one of them off and the campaign becomes a manual campaign with the same objective. The same framework now covers Leads campaigns too, which return the state ADVANTAGE_PLUS_LEADS.

The legacy ASC shutdown and the end of the customer cap

The old Advantage+ shopping campaign API was retired in stages. From Marketing API v24.0, released on 8 October 2025, new ASC campaigns couldn't be created on that version. From v25.0 (18 February 2026), creation failed on every version and un-migrated campaigns were frozen apart from migration actions. Meta's documentation says that at v26.0, legacy campaigns still using existing_customer_budget_percentage are paused.

That field was the existing customer budget cap, the one setting that stopped ASC from spending most of its budget on past buyers. It isn't available in the new setup. Meta's own suggested replacement is two ad sets, one targeting existing customers through a custom audience and one excluding them. Jon Loomer describes the same workaround, adding ad set spend limits if you use campaign budget.

Andromeda, creative volume and fewer manual levers

Meta's engineering team described Andromeda, its ad retrieval engine, in December 2024. It narrows tens of millions of candidate ads to a few thousand for ranking, and Meta reported a 6% recall gain and an 8% ads quality improvement on selected segments. The post also says Meta expects the number of creatives in the system to grow sharply. Meta hasn't published how Andromeda weighs individual creatives, so the popular advice to feed it many genuinely different concepts is practitioner observation, though it matches what we see in accounts.

Controls keep thinning. In August 2026 Loomer reported Meta testing the removal of ad set placement exclusions for Sales and Leads objectives, replacing them with value rules that raise or lower bids by placement. PPC Land reported bid cuts capped at minus 90%, and noted Meta had made no announcement. Treat it as a rollout in progress, not settled policy.

Commercially, the direction is clear. Mark Zuckerberg said on Meta's Q3 2025 earnings call that its end-to-end AI-powered ad tools, Advantage+ included, had passed a $60 billion annual revenue run rate.

Do Advantage+ campaigns actually perform better?

It depends whose numbers you read and what they measure. Meta's figures are strong. In April 2025, Meta said advertisers using its AI-enabled Advantage+ tools generate on average $4.52 in revenue per dollar spent, "an increase of 22% compared to business as usual campaigns". For Advantage+ audience specifically, Meta's documentation cites results from a 2023 experiment, collected in Loomer's Advantage+ audience guide.

Notice the pattern. The biggest saving is on clicks, leads and landing page views, which are the easiest results to buy cheaply. The saving on website conversions is much smaller.

Independent tests are less flattering. Haus, an incrementality testing company, published an analysis of 640 Meta experiments in July 2025. Advantage+ showed 2.4% higher platform-reported ROAS than manual campaigns, yet in head-to-head tests it averaged 12% lower DTC incremental ROAS. It was ahead at the midpoint of tests and behind by the end once the post-treatment window was counted.

Two caveats keep this honest. Haus's customers are large, averaging $14 million a year on Meta, so the results may not transfer to a £5,000-a-month account. And 42% of brands did see Advantage+ win. Our read: Advantage+ isn't a waste by default. It's a tool that needs measurement Meta's default columns won't give you.

Is Advantage+ wasting your budget? A diagnostic checklist

Before changing anything, work through these checks in order. Each takes a few minutes in Ads Manager or Events Manager, and together they explain most cases we audit.

The EMQ and deduplication checks deserve their own afternoon. If they look shaky, our guide to fixing Pixel and Conversions API mismatches walks through it, because no amount of campaign restructuring fixes bad input data.

What causes Advantage+ to waste budget?

These are ranked by how often they turn up in the accounts we review, starting with the most frequent.

1. Spend drifts to existing customers

People who've bought before convert at higher rates, so an algorithm told to minimise cost per purchase will find them. Without audience segments defined, you can't even see it happening. Loomer notes that segments use custom audiences from website, customer list, app, offline, catalog, lead form and shopping sources, and that page or video engagement audiences don't count.

2. View-through credit inflates ROAS

The default 7-day click, 1-day view window credits Meta for purchases from people who scrolled past an ad and bought within a day. For brands with strong organic demand or heavy email, many of those sales would have happened anyway.

3. Weak or wrong conversion signals

Optimising for the wrong event, or sending purchases with poor customer information, means the system learns from noise. This is the root cause behind a lot of "Advantage+ stopped working" complaints after a site migration or checkout change.

4. Creative too thin for the system to explore

Ten near-identical ads are one ad to a person scrolling Reels. When every variation shares a hook, the campaign saturates the same pocket of people and frequency rises. Our sister post on recovering from a Meta ads performance drop covers creative diversification in depth.

5. Not enough conversion volume

Meta's guidance is that an ad set needs roughly 50 optimisation events in a 7-day period to exit the learning phase. Splitting a modest budget across several Advantage+ campaigns starves each of them.

6. Unreviewed automation on brand and placements

Creative enhancements can add music, rewrite text or adjust visuals. Placement automation can push spend to inventory that drives clicks but few sales. Neither is always bad, but both should be choices, not defaults you never opened.

When do Advantage+ sales campaigns work, and when don't they?

Advantage+ sales campaigns tend to earn their keep in accounts that look a certain way. Here's how we decide, account by account.

Honestly, the question isn't "Advantage+ or manual" any more. It's which of the three levers you keep on, and how you check the result. A campaign with Advantage+ campaign budget and Advantage+ placements but a manual audience with exclusions is perfectly reasonable.

How to fix an Advantage+ campaign that's wasting spend

1. Define your audience segmentsGo to Advertising Settings (via All Tools, or Ad Account Settings) and open Audience Segments. Add a customer list or purchase-based website audience to Existing Customers, and website, lead form or catalog audiences to Engaged Audience. A 730-day purchase audience plus your CRM list is a sensible start.
2. Read the segment breakdown before actingGive it a few days, then use Breakdown › By Demographics › Audience Segments to see spend, purchases and cost per result for New Audience, Engaged Audience and Existing Customers. Compare cost per purchase by segment, not the blend.
3. Rebuild the customer cap with two ad setsIn one ad set, add your customer audience to the Advantage+ audience controls as an excluded custom audience. Exclusions are hard limits; suggestions are not. In a second ad set, target existing customers only. Under Advantage+ campaign budget, set an ad set spend limit on the retention ad set so it can't take more than the share you choose.
4. Check click-only resultsAdd Compare attribution settings to your columns and look at 7-day click alone. If ROAS collapses without view-through credit, report on click-only internally. Meta also offers an incremental attribution model in the ad set attribution setting, described by Social Media Today in September 2025. Haus's July 2026 editor's note says a re-run found it now outperforms standard attribution in its tests.
5. Fix the signalOptimise for Purchase if you have the volume. Make sure Purchase events carry email, phone and external ID where you have consent, and that Pixel and Conversions API share an event_id for deduplication.
6. Widen the creative, not the audienceAim for 10 to 20 ads built on clearly different concepts: a founder video, a UGC review, a static price offer, a comparison, a problem-solution demo. Review the Advantage+ creative enhancements panel and switch off any enhancement that doesn't suit your brand.
7. Set placement guardrails at account levelIf certain inventory never converts, use Advertising Settings › Account Controls › Placement Controls for account-wide limits. Where ad set exclusions have been removed, value rules can reduce bids for a placement instead.
8. Use bid controls when CPA matters more than volumeCost cap or minimum ROAS bidding works within Advantage+ campaign budget. Set the cap near your real target, not a hopeful one, or delivery stalls.
9. Prove it with a testUse Experiments in Ads Manager to run an A/B test of Advantage+ against your manual structure, or a conversion lift study if you qualify. Run it at least two to four weeks and judge on new-customer revenue in your own store data.

Mistakes that make Advantage+ worse

Some fixes do more damage than the original problem. These are the ones we see most often.

• Editing every day. Significant edits to budget, bidding or creative can restart learning. Batch changes weekly unless something is broken.
• Judging it on blended ROAS. A high ROAS driven by past buyers is a retention result dressed up as acquisition.
• Stuffing suggestions with narrow interests. Audience suggestions are only a starting point. If you want a hard limit, it has to be a control: location, minimum age, excluded custom audiences or language.
• Running three Advantage+ campaigns on one small budget. They compete for the same people and none gathers enough data.
• Turning off Advantage+ the week sales dip. Check seasonality, stock and site speed first. A broken checkout looks exactly like a bad campaign.
• Ignoring the landing page. Automation can find buyers, but it can't fix a slow product page. Our roundup of high-converting landing page examples is a good benchmark.

How to monitor Advantage+ campaigns every week

Automation needs supervision on a schedule, not constant fiddling. This is the routine we run for clients on our Meta ads management accounts.

Worked scenario: a Manchester skincare brand (illustrative)

The numbers below are illustrative, built to show the method. They aren't a client result.

A direct-to-consumer skincare brand in Manchester spends £18,000 a month on one Advantage+ sales campaign, shipping across the UK. Ads Manager shows a 4.1 ROAS on the default 7-day click, 1-day view setting. The founder's worry is simple: Shopify shows first-time orders flat since the spring, even though the campaign has scaled.

After defining audience segments, the breakdown shows Existing Customers taking 38% of spend, Engaged Audience 15% and New Audience 47%. Compare attribution settings shows about a third of purchases credited to 1-day view. Purchase EMQ sits at 5.2 because the Conversions API isn't passing phone numbers.

The fix takes an afternoon. The team passes hashed phone and external ID through CAPI, which lifts EMQ above 7 within two weeks. The campaign is split into two ad sets under Advantage+ campaign budget: a prospecting ad set with the customer list and 730-day purchasers as excluded custom audiences, and a retention ad set capped at £2,500 a month for replenishment offers. Eight new creative concepts go live, including two videos filmed by real customers and a static ad built around dry, wind-chapped winter skin, timed for the first cold snap.

Reported ROAS fell, and the business grew. That gap is the whole lesson. A competitor in Toronto or Sydney running the same structure would see the same trade: lower in-platform numbers, more genuinely new buyers. If you can't stomach the dashboard drop, run a lift test first so you have proof before you restructure.