Understand your Conversion Lift based on users measurement data

In this article, we explain the meaning of metrics within your study and users-based Conversion Lift reporting.

How do we measure lift?

Conversion Lift based on users measures the causal, incremental impact of your campaigns by measuring conversions from users who saw ads in the exposed group and users who were held back from finding ads in the control group. Cumulative results may be available during the study, but we recommend waiting until the end of your study to get the most accurate results.

Conversion Lift based on users metrics and statuses

Here’s how you check your Conversion Lift based on users data:

In your Google Ads account, click the Goals icon .
Click the Measurements drop down in the section menu.
Click Lift measurement.
Here you can view incremental conversions and Relative Lift for all biddable conversions in table view, alongside campaign impressions, cost, and Brand or Search Lift metrics.
Click on the study name to view detailed study results.

Metrics

Incremental conversions

Incremental conversions = exposed conversions - control conversions

The most fundamental metric of Conversion Lift is “Incremental conversions”, also known as "Absolute Lift". This metric reflects the incremental conversions or incremental conversion value you were able to generate due to your ad campaigns over the duration of the experiment. It’s computed as the difference between the exposed metric, which is the conversions or conversion value, from the baseline (control) metric.

Note: Incremental conversions are different than typical conversions you see in Google Ads reporting. Standard conversion reporting follows the conversion tracking settings and attribution rules set for the given conversion action whereas incrementality, or Conversion Lift, intentionally ignores these settings. For example, conversion actions might include click or view-based attribution settings such as a 7-day click-through window. In contrast, incrementality measures all unattributed conversions after an ad impression for the full study duration.

Incremental conversion value

Incremental conversion value = exposed conversion value - control conversion value

The incremental conversion value, in local currency, driven by your campaigns in the experiment period.

Note: Conversion value can vary widely. For example, someone purchasing a product or service for a company team event might purchase in bulk leading to a high conversion value when compared to typical everyday consumer behavior. This can make incremental conversion value more volatile than incremental conversions which can lead to larger confidence intervals for a Conversion Lift result.

Relative lift

Relative lift = Incremental conversions / control conversions

Relative lift reflects the percent growth in reported conversions between the exposed and baseline (control) arms. It’s computed as the incremental conversions, divided by the total metric of the baseline. As a result, relative lift can theoretically scale between -1 and infinity. For example, if 90% of your conversions are incremental, you’ll have 90 incremental conversions for every 10 baseline conversions, resulting in a relative lift of 90 / 10 = 900%. Very large relative lifts can easily occur for brands with low baseline conversion activity in the control group.

Note: Relative conversion lift can be misinterpreted when comparing across different studies or publishers. For example, it will be much easier to get high relative lift for a study with a very low baseline as each incremental conversion will have a larger impact than a study with a very high baseline. Additionally, comparing across publishers is challenging as there are different experiment and attribution methods which will make the same relative lift percentage on two different publishers appear to be the same when they're not.

Confidence interval

This metric is the estimated range in which your Conversion Lift estimates fall. For example, you may find that your Relative Lift is 10%. This is your point estimate. In brackets, you’ll find the confidence interval from at least 8% to at most 12%.

Note: Google Ads uses modeling when a subset of conversions can’t be directly linked to ad interactions, typically due to browser restrictions and cross-device behavior. Make sure you're taking steps to capture conversions across devices so your conversion lift metrics accurately measure your campaign performance. To learn more about the impact on your specific study, reach out to your account manager.

Statuses

“Significant Lift Detected”

Lift is significantly positive and there are greater than 150 exposed conversions.

“Not enough data”

Lift is significantly positive but there are less than 150 exposed conversions or there was no significant lift detected and confidence interval half width is less than 2%.

“No Significant Lift Detected”

There’s no significant lift detected and the confidence interval half width is greater than or equal to 2%.

Was this helpful?

How can we improve it?