How to Conduct an Incrementality Study on Meta's ASC Campaigns (Advantage+ Shopping Campaigns)

Introduction

Measuring the true impact of Meta's Advantage+ Shopping Campaigns (ASC) presents unique challenges. Unlike other campaign types, ASC does not allow advertisers to exclude specific locations, making traditional geo holdout studies more difficult. To accurately measure ASC's incremental impact, marketers must navigate these constraints using alternative methodologies.

This guide outlines the best approaches for conducting an incrementality study on ASC campaigns, exploring workarounds and advanced methodologies, and showcasing how Stella’s incrementality testing and Marketing Mix Modeling (MMM) can provide the most reliable insights.

The Problem: Why Isolating ASC’s Impact is Challenging

Limitations of Meta ASC Campaigns:

• No Location-Specific Exclusions: Unlike standard campaigns, ASC does not allow advertisers to turn off certain geographic regions within the campaign.
• Channel-Level Controls Only: To disable ASC in specific regions, you must adjust settings at the account level, which impacts all campaigns.
• Confounding Factors: If other campaigns run simultaneously, they may interfere with isolating ASC's impact.
• Meta’s Learning Phase Considerations: Turning off locations at the account level can disrupt Meta’s learning phase, affecting performance beyond the test period.

Given these challenges, traditional holdout methodologies must be adapted.

Meta's Incremental Attribution & Conversion Lift Studies

Meta offers built-in solutions for measuring incremental impact, including Incrementality Optimization and Conversion Lift Studies.

Incrementality Optimization

Meta has introduced an Incrementality Optimization feature for sales campaigns, which is designed to:

• Use historical holdout test data to optimize ad delivery toward truly incremental conversions.
• Reduce Cost Per Acquisition (CPA) by optimizing for incremental sales rather than standard attribution-based conversions.
• Provide a more transparent, performance-driven approach to Meta's automated campaign structures, such as ASC.

Conversion Lift Studies

Meta's Conversion Lift Studies provide another method for measuring incrementality:

1. Audience Split: Meta randomly assigns users into test and control groups.
   • The test group sees ads; the control group does not.
2. Measurement Period: Typically runs for 3-4 weeks.
3. Results Analysis: Meta calculates the lift in conversions driven by the campaign, measuring:
   • True incremental revenue.
   • The effect of Meta campaigns beyond last-click attribution.
   • Brand impact and long-term customer engagement.

While Meta’s tools provide valuable insights, they are often limited to Meta's ecosystem and do not account for cross-channel effects. To gain a more complete picture of incrementality, combining Meta’s tools with third-party measurement solutions like Stella's incrementality testing and MMM analysis is ideal.

Take a quick demo of Stella's incrementality tool below:

Method 1: Inverse Holdout Test

An inverse holdout test measures incrementality by turning off ASC in specific regions and comparing results against regions where ASC remains active.

Steps:

1. Identify Test and Control Regions
   • Use Designated Market Areas (DMAs) or commuting zones to define test and control groups.
   • Ensure regions are similar in audience behavior and purchasing trends to minimize bias.
2. Turn Off ASC in the Test Regions
   • Use Meta’s account-level settings to exclude locations.
   • Ensure ASC remains the only active campaign in control regions.
   • Pause other campaign types across all regions to isolate ASC’s effect.
3. Collect Data
   • Track conversions, revenue, and CPA in both test and control regions.
   • Monitor any organic sales uplift to control for non-paid influences.
4. Analyze Results
   • Calculate the incremental lift by comparing conversion rates between the two groups.
   • If ASC delivers significantly higher conversions in active regions, it indicates a positive incremental impact.

Considerations:

• Potential Learning Disruptions: Since location exclusions are applied at the account level, other campaigns may be affected.
• Spillover Effect: Test and control regions should be geographically distinct to avoid audience overlap.

Method 2: Creating a Separate Meta Ad Account for Testing

Another option is to create a dedicated ad account for incrementality testing. This method avoids the risks of disrupting account-level learning while allowing for more controlled experiments.

Steps:

1. Create a Secondary Meta Ad Account
   • Ensure it is properly configured and compliant with Meta’s policies.
   • Share the same pixel and conversion tracking data between accounts.
2. Run ASC in the New Ad Account
   • Keep all other campaign types paused in this account to ensure ASC’s impact is isolated.
3. Compare Performance Across Accounts
   • Analyze the incremental lift by comparing conversions, revenue, and CPA across the test and primary accounts.

Considerations:

• Compliance Risks: Meta sometimes flags multiple accounts under the same business.
• Data Transfer Issues: Pixel sharing can mitigate data loss, but attribution modeling may still be affected.

Method 3: Causal Impact Analysis

A Causal Impact test measures incrementality by turning off all ASC campaigns for 15-20 days while keeping all other campaigns constant. This method evaluates the loss in incremental revenue during the period when ASC is inactive.

Steps:

1. Turn Off All ASC Campaigns for 15-20 Days
   • Keep all other campaigns active to maintain consistency.
   • Ensure that no budget is reallocated to other campaigns to isolate ASC’s true impact.
2. Monitor Key Metrics
   • Track revenue loss, cost per acquisition (CPA), and conversion volume.
   • Calculate incremental Return on Ad Spend (iROAS) and Incremental Cost Per Order (iCPO).
3. Analyze Results
   • If revenue significantly declines when ASC is turned off, the difference represents the incremental value ASC contributes.
   • Use statistical methods to determine confidence in the results.

Considerations:

• Potential Revenue Loss: Since ASC is paused, businesses may experience a temporary drop in sales.
• High Confidence Measurement: This method provides one of the most accurate assessments of ASC’s true impact but requires careful planning to minimize risk.

Conclusion

By incorporating Meta's Incrementality Optimization and Conversion Lift Studies alongside robust third-party methodologies like Inverse Holdouts, Causal Impact Testing, and MMM, advertisers can gain a true understanding of ASC’s incremental impact.

Stella provides the most comprehensive approach by allowing advertisers to measure ASC’s impact accurately while maintaining overall campaign stability. By leveraging Stella’s tools, brands can make data-driven budget allocation decisions and maximize ASC’s effectiveness without disrupting broader marketing efforts.

‍