Conversion Lift allows you to measure the incremental conversions driven by your Google Ads campaigns. With Conversion Lift, you can run a controlled experiment whereby you can separate your customers into comparable geographic groups, and measure the lift in downstream conversions that are driven by each of these two groups:
- Geographic regions that show your ads
- Geographic regions that don't show your ads
The difference in conversions between these two groups will tell you the 'lift', or the increase in conversions that are caused by your campaigns. Conversion Lift also allows you to split the exposed and control groups based on geography.
In this article we’ll be focusing on Conversion Lift based on geography.
Before you begin
Before you can set up Conversion Lift measurement, you need to create a campaign to serve on a supported Google Ads platform. The supported campaign types are App, Demand Gen, Discovery, Display, Video, Search, Shopping and Performance Max. For more information, reach out to your Google account representative.
Is Conversion Lift based on geography the correct tool for my question?
This test is recommended for you as an advertiser if you want to understand the total, unattributed, incremental return on investment (ROI) of your campaigns. Here are a few of the benefits:
- Understand the total incremental return on ad spend (iROAS) of your campaigns in order to make budget allocation decisions
- Geo-split methodology is available externally
- Measure multiple channels in one study
- Measure multiple conversion types, such as web and app, in one study
Conversion Lift based on geography eligibility
To use Conversion Lift, your Google Ads account must be tracking at least one compatible conversion action. Alternatively, you can also provide your account team with your offline data. Here are the different types of conversion that you can leverage for Conversion Lift based on geography:
- Google Ads conversion tag (formerly known as Google Ads tag)
- Firebase conversion tag
- Display and Video 360 (DV360) Floodlight
- Bring your own device (BYOD) support for any offline conversion type aggregated to ZIP or city level
In addition to conversions, campaigns must target a single country to be eligible for the geo-split methodology.
There are three levels of feasibility status:
- High
- Medium
- Low
'High' feasibility status will give you the best chance at generating statistically significant results. We don’t recommend proceeding with low feasibility status, in this scenario you should update your budget based on the recommendation and review your updated feasibility result.
Depending on your conversion goals and business type, 'Medium' feasibility status may be reasonable but we recommend proceeding with caution. For more information, contact your Account manager.
Lift isn’t guaranteed for a variety of reasons, which is why we recommend using a test and learn framework to iterate and improve your outcomes over time. To improve your estimated conversions, you can increase your budget or optimise your experiment to shallow conversions that are directionally correlated with your purchase goal. Make sure that you follow the best practices of the campaign. For more information, contact your Account manager.
Instructions
Set up Conversion Lift measurement
- In your Google Ads account, click the Goals icon
.
- Click the Measurements drop-down in the section menu.
- Click Lift measurement.
- Click the plus
button.
- Select Conversion Lift under the 'Based on Geo' section.
- Choose which campaigns you’d like to opt into Conversion Lift.
- Note: Campaigns can only be active in one study at a time. If you’re unable to select a campaign, it most likely means that it’s already being used in another study.
- Select the start and end dates for your study.
- Select the conversions that you want to test with.
- Review the feasibility status in the right-hand column for an estimate on how likely you are to get precise results based on the campaigns being measured.
- Note: A 'High' feasibility status will offer you the best chances of a successful study. Increasing your budget or including all the campaigns for a specific channel or goal will help you achieve a 'High' feasibility status.
- Click Save.
.Understanding the experimentation units: Google Marketing Areas
Google Marketing Areas (GMAs) are sub-country geographical regions designed to serve as experimental units in geo experiments. A contamination model estimates how often someone travels between different GMAs. The contamination model selects GMAs that minimise possible contamination caused by exposure to advertising in one GMA, but then convert in an unexposed GMA. GMAs are designed to minimise the effects of contamination while maintaining enough experimental units to ensure statistical robustness when the experiment is performed.
Contamination
If some users are exposed to an ad in a treatment GMA, but then perform a conversion in another GMA that is part of the control or vice versa, this would result in a 'contamination' of the control group and lower the estimated difference between treatment and control. This contamination will then reduce the reported incrementality of the ad campaign.
GMA generation
We choose elementary building blocks for the GMAs from the existing geo target hierarchy at Google. A spectral clustering algorithm generates GMAs by balancing the counteracting goals of minimising contamination using Google Ads data and maximising the representativeness and diversity of the GMAs.
