Conversion Lift helps measure the actual number of conversions driven by advertising on Display & Video 360 and Google campaigns. It helps you understand conversion numbers, including website visits or purchases, that your ads specifically caused and would not have occurred otherwise. Quantifying this incremental value enables more informed decisions regarding advertising spend and campaign optimization.
On this page:
- How Conversion Lift measurement works
- Eligibility and availability of Conversion Lift studies
- Request a Conversion Lift study
- Best practices
- Reporting
- Considerations and limitations
How Conversion Lift measurement works
Conversion Lift uses a scientific approach and determines your campaign incrementality. This experiment divides your audience into two groups through randomized controlled experiments.
- Treatment group: Users in this group are exposed to your ads.
- Control group: Users in this group do not see your ads. Instead, they are shown the next ad in the auction.
Conversion Lift isolates direct ad impact by comparing conversion behavior. Enhanced Conversions improve accuracy and measurement coverage.
Eligibility and availability for Conversion Lift studies
Conversion Lift supports a wide range of campaign types and platforms:
- Google Ads: Video, Performance Max, Discovery, Demand Gen, App (Android only), Search, and Shopping campaigns.
- Display & Video 360: Campaigns managed through Display & Video 360 (including YouTube & Google Display Ads line items).
Before setting up a Conversion Lift study, ensure your account and campaigns meet these requirements:
- Conversion Tracking: Your Google Ads account must track at least one compatible conversion type. Conversion tracking can be direct or through a manager account.
- Enhanced Conversions: It is strongly recommended to enable enhanced conversions. Enhanced conversions capture a broader range of conversion data.
- Conversion Volume: Campaigns should generate at least 3,000 conversions during the study period. Conversion volume helps achieve statistically-significant results. Achieving the minimum conversion volume is essential for successful studies.
- Feasibility Status: For YouTube campaigns, aim for a "medium" or "high" feasibility status during study setup. The status indicates a higher likelihood of obtaining statistically-significant results. Avoid studies that do not meet minimum conversion-volume requirements.
- Study Duration: Studies must have a minimum duration of 7 days and a maximum duration of 56 days.
Determine when to use Conversion Lift
Conversion Lift can be highly beneficial in some scenarios:
Good fit:
- If you want to understand their ad-spend effectiveness or platform performance.
- If you’ve a strong interest in understanding incrementality and value data-driven insights.
Consider other useful options (not a good fit):
- If you are already satisfied with the current performance metrics.
- If you primarily value view-through conversions as their main success metric.
- If you’ve focused exclusively on brand awareness (Brand Lift studies are excellent for this).
Request a Conversion Lift study
Request a Conversion Lift study by following the steps below, once Eligibility and availability for Conversion Lift studies criteria are met.
- Contact your Google sales representative: Initiate the process by reaching out to your dedicated sales contact.
- Verify eligibility: Be prepared to share the necessary campaign and account details to confirm you satisfy the minimum requirements.
- Define study parameters: Collaborate with your sales representative to establish the scope, duration, and core metrics for the study.
Best practices
- Optimize conversion events: Implement multiple conversion events to represent the different stages of the marketing funnel. The implementation will provide deeper insights into how your ads impact user actions.
- Choose appropriate lookback window: Select a lookback window that reflects expected time for an ad exposure to lead to a conversion. A shorter window might show higher relative lift, but not necessarily a higher absolute lift.
- Set study-end date strategically: Extend the study-end date beyond your campaign-end date (for example., 7 days after) to capture delayed conversions that "trickle in" after the main campaign period.
- Leverage optimization recommendations: Use smart bidding strategies. Examples include target CPA or maximize conversions. Use direct-response-oriented creative for campaigns included in the study.
- Avoid stacking tests: Do not run concurrent lift experiments. Applying to prospecting and remarketing campaigns simultaneously can lead to control group contamination and compromises the accuracy of your results.
- Account for wide conversion values: Adjusting your conversion pixel setup smooths widely varying values, helping to accurately calculate ROAS.
Video campaign specifics (YouTube):
- Note that conversion lag is longer for video campaigns compared to search and Google Display Ads campaigns.
- Understand potential mobile volume tracking gaps especially true for cross-device conversions and logged-out users.
- Target CPA bidding normalizes in about 7 days. Do this before starting a study. It ensures optimal campaign performance during the experiment.
Reporting
Accessing detailed reports
Your Google sales representative will provide you with the results and detailed reports for your Conversion Lift study. For eligible video and Demand Gen studies, some aggregated results are shared.
Considerations and limitations
- External factors impacting lift:
- Control group contamination: Occurs if users in the control group are exposed which means similar advertising from other channels or campaigns not included in the study.
- Environmental factors: Seasonality, industry events, or other external influences can affect consumer behavior which can impact study results.
- Lower bound of performance: Conversion Lift results represent a lower bound of actual performance due to inherent measurement coverage gaps. Examples include Intelligent Tracking Prevention (ITP) and cookie churn. Other external factors cannot be fully captured.
- iOS 14+ limitations: Tracking click-based conversions on iOS campaigns is limited due to App Tracking Transparency (ATT) restrictions. To mitigate the limitation, we recommend using Enhanced Conversions for iOS traffic. If many conversions cannot be captured, consider excluding iOS traffic from experiments.
- Proactive management:
- Avoid launching studies on the same day campaigns start. Schedule studies to begin at least 24 hours after campaign launch.
- To prevent overlap and potential bias, you should avoid running multiple experiments concurrently, especially when different targeting types are involved.