Meta’s Advantage+ Shopping Campaigns Are Rewriting DTC Acquisition Math

For DTC founders who spent the better part of 2023 and 2024 mourning the death of Facebook advertising as they knew it, 2026 has delivered a jarring reversal. Meta’s Advantage+ Shopping Campaigns — the fully automated ad product the company began aggressively pushing to e-commerce advertisers in late 2024 — are now generating some of the lowest blended CPAs brands have seen since iOS 14 decimated signal quality three years ago. But the gains come with a catch: the same automation that’s driving efficiency is making it nearly impossible to understand why performance is improving, let alone replicate it manually.

Across conversations with a dozen Shopify Plus merchants and agency operators over the past three weeks, Ecommerce Times found a consistent pattern. Brands that fully committed to Advantage+ — consolidating their legacy campaign structures, sunsetting manual interest and lookalike targeting, and feeding Meta’s model clean, high-volume purchase signals — are reporting blended ROAS figures that would have looked implausible twelve months ago. The operators who are half-in, running Advantage+ alongside a legacy manual campaign architecture, are seeing the benefits largely neutralized by signal cannibalization.

What are the actual performance numbers brands are seeing with Advantage+ in 2026?

The range is wide enough to be meaningful. Among the brands Ecommerce Times spoke with, CPAs for cold prospecting dropped between 28% and 47% year-over-year in Q2 2026 compared to manually structured campaigns run in Q2 2025. Average order values held flat or improved slightly, which operators attribute to Meta’s model learning to find higher-intent buyers rather than simply cheaper clicks.

Jake Levine, director of paid social at Sharma Brands, which manages Meta spend for a portfolio of DTC health and home brands, said the results have been impossible to ignore — but equally difficult to explain to clients conditioned to keyword-level granularity from Google.

“We had a candle brand doing $180 CPAs on manual prospecting in March. We consolidated everything into Advantage+, gave the pixel 60 days to learn, and by June we were at $94. The CFO wanted to know exactly which creative drove that. The honest answer is: we don’t fully know, and that’s the uncomfortable truth with this system.”

💡 Article Summary

Key Insights

1

What are the actual performance numbers brands are seeing with Advantage+ in 2026?

2

Why is attribution breaking down — and what tools are operators using to compensate?

3

How does creative strategy change when the algorithm controls audience and placement?

4

What does Advantage+ mean for brands still running Google Shopping alongside Meta?

5

Are there categories or business models where Advantage+ is underperforming?

Source: Ecommerce Times

Shannon Fitzsimmons, head of growth at Chicago-based apparel brand Cove & Coast, which runs roughly $400,000 in monthly Meta spend, described a similar dynamic. Her team moved to a full Advantage+ structure in January 2026 after a pilot in Q4 2025 showed a 31% CPA reduction during peak gifting season.

“The model finds audiences we never would have targeted manually. We’re seeing purchasers come through who look nothing like our historical customer profile demographically, but they’re converting at the same rate and returning at the same rate. That tells me Meta is optimizing on something more behavioral than we ever had access to.”

Why is attribution breaking down — and what tools are operators using to compensate?

The efficiency gains are real. The attribution picture is a mess. Advantage+ campaigns operate largely as black boxes: Meta’s model decides placement, audience, creative rotation, and bid in real time, and it reports aggregate results at the campaign level with far less breakout data than manually structured campaigns. For merchants who built their measurement stacks around ad-set-level ROAS from Ads Manager, that loss of granularity is operationally painful.

The platforms absorbing that gap are media mix modeling and incrementality testing tools. Triple Whale’s Incrementality Pulse, Northbeam’s cross-channel attribution layer, and Rockerbox’s multi-touch model are all seeing increased adoption among mid-market Shopify brands specifically because Advantage+ has made last-click attribution from Meta’s own reporting less trustworthy, not more.

• Triple Whale Incrementality Pulse: Runs geo-based holdout tests to measure true lift from Meta spend, independent of platform-reported conversions. Brands using it report 15–25% discrepancies between Meta-reported ROAS and incrementally measured ROAS — a gap that matters significantly at scale.
• Northbeam: Its cross-channel model is now the primary ROAS source of truth for several eight-figure Shopify brands, with Meta’s Ads Manager relegated to a signal input rather than a final reporting source.
• Rockerbox: Particularly popular with brands running simultaneous Meta, Google, and TikTok spend, where cross-channel de-duplication is the core problem.
• Meta’s own Conversion Lift studies: Several agency operators told Ecommerce Times they’re running formal lift studies every quarter now rather than annually, as a sanity check against Advantage+ reporting.

Ryan Moran, VP of performance at Pilothouse Digital, which manages nine-figure annual Meta spend across its DTC client base, put the measurement problem bluntly.

“Advantage+ is the best prospecting vehicle we’ve had on Meta since 2020. It’s also the least transparent. You have to build a parallel measurement stack to trust what you’re seeing. Brands that are just reading Ads Manager ROAS and making budget decisions off that are flying blind, even when the numbers look great.”

How does creative strategy change when the algorithm controls audience and placement?

One area where operators retain significant control — and where the performance delta between brands is widening — is creative. Because Advantage+ handles audience and placement decisions, creative quality has become the primary lever brands can pull. The brands outperforming benchmarks in 2026 are not necessarily spending more on creative production; they’re feeding the algorithm more creative diversity.

The winning formula that’s emerged across several high-performing accounts, according to agency operators interviewed for this piece, looks like this:

• Eight to twelve distinct creative concepts per Advantage+ campaign, not variations on one concept
• A mix of creator-shot UGC, brand-produced video, and static product-forward images — not a single format
• Hooks tested at the three-second mark, with brands using tools like Foreplay and Motion to audit thumb-stop rates before creative enters the campaign
• Catalog-connected dynamic creative running alongside static assets, so the model can serve personalized product combinations based on browsing behavior
• Aggressive creative refresh cycles — top performers are rotating new concepts every 10–14 days, not monthly

Levine at Sharma Brands said the agency now treats creative volume as a media investment, not a production cost. “We used to spend 80% of our time on audience architecture and 20% on creative. That’s completely flipped. The algorithm handles the audience. Your only job is to give it better creative than your competitor is giving theirs.”

What does Advantage+ mean for brands still running Google Shopping alongside Meta?

The rise of Advantage+ is not happening in isolation. Most DTC brands are running concurrent Google Shopping spend, and the interaction between the two channels is creating new budget allocation debates at the leadership level. The core tension: Advantage+ is increasingly capturing customers earlier in the funnel, which is compressing the population of incremental demand that Google Shopping can claim credit for at the bottom.

Fitzsimmons at Cove & Coast said her team noticed Google Shopping ROAS inflating in Q1 2026, which initially looked like a win. “Then we ran a holdout and realized Meta was doing so much of the heavy lifting in awareness and consideration that Google was harvesting demand it didn’t create. Our Google Shopping ROAS looks great on paper, but its incrementality score is lower than it was two years ago.”

The practical response from several brands has been to pull back Google Shopping budget modestly — particularly on branded and generic non-branded terms — and reallocate that spend into Advantage+ creative testing budgets. The brands that have done this are reporting higher blended ROAS across total paid media, though the shift requires a leadership team comfortable moving away from channel-level ROAS as the primary KPI.

Are there categories or business models where Advantage+ is underperforming?

The results are not uniform. Operators identify several scenarios where Advantage+ has failed to deliver the efficiency gains seen in broader DTC categories.

• Low SKU count brands: Brands with fewer than 20 SKUs give the algorithm too little to optimize across, limiting its ability to personalize at scale.
• High AOV, long consideration cycles: Luxury and considered-purchase categories — furniture, fine jewelry, high-end outdoor equipment — are seeing more modest CPA improvements, as the model struggles to optimize for purchases that happen two to four weeks after the first ad exposure.
• Subscription-first brands: Brands where the primary conversion event is a subscription start rather than a one-time purchase have reported mixed results, because Meta’s purchase signal training data is weighted heavily toward single transactions.
• New brands under 500 purchase events per month: The minimum viable signal threshold for Advantage+ to meaningfully learn is roughly 500 purchases per month in the account. Below that, manual targeting with broad audiences tends to outperform.

Moran at Pilothouse flagged the subscription problem specifically. “We have a couple of subscription health brands where Advantage+ is generating cheap first orders but the LTV cohorts on those customers are 20% worse than our manual prospecting cohorts. The model optimizes for conversion, not for the customer you actually want. That’s a meaningful distinction.”

What should operators do right now to prepare for Q4 2026?

With peak season twelve weeks out, the operators Ecommerce Times interviewed offered a consistent set of actions for brands that haven’t yet fully committed to Advantage+.

• Run a 30-day parallel test now, before Q4 traffic inflates CPMs and makes clean data harder to read
• Audit your Meta pixel and Conversions API implementation — Advantage+ is only as good as the signal quality feeding it, and CAPI gaps are the most common performance ceiling operators find during audits
• Build your creative library before September — production timelines compress badly in October, and campaigns that enter Q4 with thin creative sets underperform consistently
• Establish a media mix model or incrementality testing cadence before Black Friday so you’re not interpreting Q4 results through Ads Manager alone
• Define your threshold for Advantage+ success in terms of blended CPA or new customer CPA, not platform-reported ROAS

The broader picture heading into Q4 2026 is that Meta has rebuilt meaningful trust with DTC operators who wrote it off after iOS 14 — but the terms of engagement have changed fundamentally. The brands winning are the ones that have accepted the automation trade-off, invested in parallel measurement infrastructure, and redirected their operational energy toward the one variable they still control: creative. The brands still trying to manually manage audience architecture inside Advantage+ are, by most accounts, getting the worst of both worlds.