Advantage+ Shopping vs Manual ASC: Cohort Quality Side-By-Side
A head-to-head look at how Advantage+ Shopping and manually-structured ASC campaigns compare on the four cohort-quality metrics that actually move contribution margin — and when the gap is big enough to rebuild.
Advantage+ Shopping vs Manual ASC: Cohort Quality Side-By-Side
A back-to-back comparison of the customers Advantage+ Shopping and manually-structured ASC campaigns actually acquire — not the ROAS they report.
Advantage+ Shopping Campaigns (ASC+) and manually-built ASC equivalents can post near-identical blended ROAS in Ads Manager and still deliver very different customer bases downstream. Cohort quality is the difference: the mix of discount users, the 90-day repeat rate, contribution-margin-adjusted LTV, and how fast CAC pays back.
This comparison isolates those four metrics on cohorts acquired through each campaign structure over the same window, then translates the gap into a rebuild decision. It is the frame Performance Managers use before deciding whether ASC+ automation is worth the cohort drag it introduces.
Reported ROAS is the wrong scoreboard for this decision. ASC+ optimises to a conversion signal Meta receives inside a 7-day window, so any structural bias in the cohort it acquires — heavier discount-code usage, lower repeat rate, thinner margin — shows up 30 to 90 days later on your P&L, not inside the ad platform.
The comparison below uses cohorts of first-time buyers acquired in the same month, tagged by campaign structure, and then tracked forward on four downstream metrics. It is designed to answer one question: is the ASC+ cohort materially worse, and by how much.
ASC+ vs manual ASC — cohort quality after 90 days (apparel & beauty, AOV €55-€95)
| Cohort metric | Advantage+ Shopping | Manual ASC | Gap |
|---|---|---|---|
| First-order discount-code usage | 58% | 41% | +17 pts ASC+ |
| 90-day repeat purchase rate | 18% | 26% | -8 pts ASC+ |
| Average first-order CM % (after COGS, shipping, discount) | 24% | 31% | -7 pts ASC+ |
| CM-adjusted 180-day LTV (€ per acquired customer) | €48 | €67 | -€19 ASC+ |
| CAC payback (months to recover CAC on CM) | 5.1 | 3.4 | +1.7 mo ASC+ |
| Reported blended ROAS (in-platform, 7d click) | 2.9x | 2.7x | +0.2x ASC+ |
The pattern is consistent across accounts running both structures side-by-side: ASC+ posts a slightly better in-platform ROAS while acquiring a cohort that discounts more, repeats less, carries thinner margin, and takes roughly 50% longer to pay back. The two campaign types are not producing the same customer.
Where the gap comes from
Three mechanisms account for most of the divergence. First, ASC+ pulls a heavier discount-seeker mix than manual ASC because its creative-selection algorithm rewards promo-led ads that convert cold traffic quickly — the same reason the discount-seeker skew shows up so clearly on high-discount apparel accounts.
Second, the existing-customer budget cap defaults to a low share, and any spend Meta classifies as new-customer is what ASC+ optimises around. That makes the cohort look cheaper to acquire in-platform while quietly loading it with lower-intent buyers. Third, manual ASC gives you exclusion control — suppression lists, wholesale-code exclusions, low-LTV lookalike removals — which meaningfully lifts the quality of who actually converts.
The existing-customer cap trap
ASC+ lets you cap existing-customer spend but reports blended ROAS including any existing-customer conversions that slip through. If you leave the cap at the default and read ROAS at the campaign level, you are systematically overstating new-customer performance by 15-25% — which hides most of the cohort gap in the table above.
When to stay on ASC+, when to rebuild manually
The rebuild threshold is not a fixed ROAS gap — it is a CM-adjusted LTV gap crossed with your payback tolerance. If ASC+ is costing you more than €15 of CM-LTV per acquired customer and pushing payback past four months, the automation savings rarely justify the cohort drag. Below that, ASC+ operational simplicity usually wins.
The pragmatic middle path is a hybrid split: keep ASC+ running for top-of-funnel prospecting on lower-margin catalog SKUs, and rebuild manual ASC for cohort-critical verticals — subscription, high-AOV, or anything where 90-day repeat drives the unit economics. A properly designed holdout test on a single vertical will surface the answer within 6-8 weeks.
Cumulative CM recovery per acquired customer — ASC+ vs manual ASC
Advantage+ Shopping
Manual ASC
Frequently asked questions
In-platform, ASC+ typically reports equal or slightly better ROAS. Downstream, its cohorts show 6-10 points lower 90-day repeat rate and roughly €15-€25 lower CM-adjusted LTV in the accounts we see. The gap is real on the P&L; it just does not show up in Ads Manager.
A common rebuild threshold is a CM-adjusted LTV gap greater than €15 per customer combined with CAC payback drifting past four months. Below that, the operational cost of running manual ASC — creative volume, audience management, exclusion upkeep — usually exceeds the cohort lift.
ASC+ creative selection rewards ads with strong short-window conversion, which disproportionately favours promo-led creative. That in turn pulls a heavier discount-seeker mix. Manual ASC lets you cap or exclude promo creative in mixes where repeat rate matters more than volume.
Only partly. The cap limits existing-customer spend but does not correct the new-customer cohort composition. Even with the cap set aggressively, ASC+ still pulls higher discount-usage and lower repeat rate than a well-structured manual ASC. Use the cap plus a holdout test, not just the cap.
Split by geography or by product line, hold structure constant for at least six weeks, tag first-time buyers by campaign source, and evaluate on 90-day repeat rate and CM-adjusted LTV — not blended ROAS. Attribution needs to be server-side or post-purchase-survey based; pixel attribution favours ASC+.
High-discount apparel is the clearest case — the cohort gap is widest where promo intensity is highest. Beauty subscription, supplements, and any category where repeat drives 60%+ of LTV also show large gaps. Low-repeat, one-shot verticals (mattresses, furniture) see the smallest gap.
Yes, and a hybrid split is often the right answer. Keep ASC+ for prospecting on lower-margin SKUs where operational simplicity wins, and run manual ASC on cohort-critical product lines. Just make sure your exclusions prevent the two structures from bidding against each other.
It narrows the discount-seeker skew somewhat but does not eliminate it. The creative-selection bias toward promo-led ads persists regardless of the seed audience. Manual ASC with the same customer file typically still outperforms on 90-day repeat by 4-6 points.
90 days minimum for repeat rate to stabilise, 180 days ideally for CM-adjusted LTV. If your product has a long repeat cycle (skincare refills, apparel seasonality), extend to 270 days. Deciding on 30-day data will systematically favour ASC+ because early conversions weight it.
Not without a signal change. ASC+ optimises to purchase events, not to margin, repeat, or LTV. Until the optimisation signal includes downstream value — which requires server-side conversion enrichment with margin or repeat data — the structural cohort bias will persist. Feeding value-enriched conversions is the closest current workaround.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.