Measuring Meta Advantage+ properly means looking beyond in-platform ROAS. Because Advantage+ automates audiences, placements and creative delivery, it tends to find people who were already likely to buy, which can inflate reported results. A reliable measurement setup combines Meta’s reported conversions with new customer share, blended business metrics, and at least one incrementality test such as a Conversion Lift study or geographic holdout.
Advantage+ is Meta’s suite of AI-driven automation for ads on Facebook, Instagram, Messenger and the Audience Network. It includes Advantage+ sales campaigns, formerly Advantage+ shopping campaigns, along with Advantage+ audience, placements and creative enhancements. Meta has been merging these options into its standard campaign setup, so names and toggles shift over time.
Why is Advantage+ harder to measure than manual campaigns?
Manual campaigns separate prospecting and retargeting by audience, so you can see what each does. Advantage+ blends them. The algorithm decides who sees the ad, and that often includes past visitors and existing customers who convert at high rates.
The result is a ROAS figure that may look excellent while adding little new revenue. Meta does provide controls, such as defining existing customers and capping the share of budget spent on them, but you need to set them up deliberately.
Which KPIs actually show whether Advantage+ works?
- New customer acquisition cost: spend divided by first-time buyers
- New versus returning customer share of attributed purchases
- Blended MER: total revenue divided by total ad spend across channels
- Incremental ROAS from a lift test, not attributed ROAS
- Contribution margin after ad spend, not revenue alone
- Creative fatigue signals: frequency and click-through trend by ad
Attributed ROAS still belongs on the dashboard as a directional signal. It should not be the number that decides budget on its own.
How do you set up tracking correctly first?
Measurement quality depends on signal quality. Advantage+ optimises on the conversion events you send, so weak events lead to weak optimisation and weak reporting.
- Install the Meta Pixel and the Conversions API, with deduplication by event ID
- Check Event Match Quality in Events Manager and pass hashed email and phone where consented
- Send purchase value and currency on every purchase event
- Upload or sync a customer list so Meta can identify existing customers
- Configure the existing customer definition in Advertiser settings
- Confirm consent handling meets GDPR and KVKK requirements in your markets
How do you test incrementality?
Incrementality asks what would have happened without the ads. Two methods are practical for most advertisers.
Conversion Lift studies
Meta runs randomised tests that hold out a control group from seeing ads, then compares conversions between the groups. Availability depends on spend level and account access; ask your Meta representative or check Experiments in Ads Manager. Results give incremental conversions and cost per incremental conversion.
Geographic holdouts
Pause or reduce Advantage+ spend in a set of comparable regions for several weeks while keeping it running elsewhere. Compare sales in your own store or CRM data. This works without Meta’s tools and can be analysed with open-source packages like GeoLift.
Either way, plan the test before launch, run it long enough to cover your purchase cycle, and avoid changing other channels in the test regions at the same time.
What should the Advantage+ dashboard show?
Build the dashboard in Looker Studio, a BI tool or a sheet, pulling Meta data alongside store or CRM data. Keep it to one screen.
- Top row: spend, blended MER, new customer CAC, contribution margin
- Second row: Meta-attributed purchases split into new and returning, with ROAS for each
- Third row: latest incrementality result and the date it was measured
- Creative panel: top and bottom ads by spend, with frequency and click-through rate
- Trend line: weekly new customer CAC against the target
Label attributed figures clearly as Meta-reported so nobody mistakes them for finance numbers.
How Alien Road applies Meta Advantage+ measurement
In a typical engagement we begin with a signal audit: Pixel and Conversions API setup, deduplication, Event Match Quality, purchase value accuracy and whether the existing customer list is current. Fixing those usually changes reported results before any campaign changes.
Setup then covers the existing customer definition and budget cap, campaign structure with a clear naming convention, and a dashboard joining Meta data with the client’s store or CRM. For measurement we schedule a lift study or geographic holdout within the first quarter, then repeat it periodically. Reporting leads with new customer CAC and incremental results, with attributed ROAS shown beneath, so budget decisions reflect what the ads added rather than what they claimed.
Common measurement mistakes
Comparing Advantage+ ROAS with a manual prospecting campaign’s ROAS is misleading, because one includes warm buyers and the other does not. Compare like with like, or compare incremental results.
Changing attribution windows mid-test breaks comparisons. Pick a setting, such as seven-day click and one-day view, and keep it for the reporting period.
Judging creative too early is another trap. Advantage+ needs time to explore placements and audiences. Decisions on individual ads after a day or two of spend often reflect noise.
How often should you rerun incrementality tests?
Results age. Audience saturation, creative changes, seasonality and shifts in competitor spend all change how much Advantage+ adds. Many advertisers repeat a lift test or geographic holdout every quarter or after a major change in budget or structure.
Between tests, watch new customer CAC and blended MER as early warnings. If new customer CAC rises steadily while attributed ROAS stays flat, the campaign may be leaning more on existing customers, and a fresh test is due.
Keep a simple log of each test: dates, method, budget, result and confidence interval. Over a year, that log becomes the most trustworthy record of what Meta spend actually delivers.
Where to start this week
Open Events Manager and check the Event Match Quality score for your purchase event, then confirm deduplication is working between the Pixel and Conversions API. Next, pull last month’s Advantage+ purchases and match them against your store data to see how many were first-time buyers. That single new customer share figure will tell you more about real performance than the ROAS column.
Keep reading: Meta Ads · AI Advertising Optimization