iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Google Ads Conversion Lift estimates how many conversions an campaign caused, beyond the conversions that would likely have happened without ad exposure. A self-service, user-based study can be set up in Google Ads, but only if the account is eligible, the campaigns and conversion actions qualify, and the study has enough conversion volume to produce a readable result. This guide covers what to check before launch, how to set up the study, and how to read the output without confusing incremental impact with attributed credit.
What Conversion Lift measures
Conversion Lift is a controlled experiment. In a user-based study, one group of users is eligible to see your ads (the treatment group) and a second group is held back (the control group). Because both groups are compared over the same study period, the difference in their outcomes is intended to estimate the conversions your advertising directly drove.
That is a different question from the one standard conversion reporting answers. Attributed conversions show which conversions receive credit under the tracking and attribution settings in your account. Lift does not reassign credit; it compares what happened across the two groups.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchLift is also not a perfect count. Google uses modeling when a conversion cannot be linked directly to an ad interaction, for example because of browser restrictions or cross-device behavior. The result is therefore an estimate with a stated level of certainty, not an exact tally of every caused conversion.
#1 Best Overall
Check eligibility before you plan a study
Conversion Lift is not available to every account. Google’s setup guidance says, “Conversion Lift isn’t available for all Google Ads accounts,” and directs advertisers to contact their Google account representative to confirm access. Once access is confirmed, the setup guidance lists these requirements:
- Observed conversions: at least 1,000 observed conversions.
- Budget: a minimum campaign budget of US$5,000.
- Conversion actions: a conversion action compatible with lift measurement must be selected.
- Campaign types: Display, Search, Video, Demand Gen, App Campaigns, and Performance Max are listed as supported. iOS-targeted App campaigns and Travel Ads are unsupported.
- Study overlap: a campaign can be in only one Brand Lift, Search Lift, or Conversion Lift study at a time.
These thresholds are current as of Google Ads Help guidance checked on 8 October 2026. Eligibility is account-specific and can change, so confirm it with your Google representative rather than assuming a published threshold applies to your account.
How to set up a user-based Conversion Lift study
The study is created from the Lift studies area in Google Ads. Menu labels can change, so use the wording shown in your account if it differs.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Rank #2
- Open the Lift studies area in your Google Ads account and select the option to create a new study.
- Choose Conversion Lift as the study type.
- Select user-based groups as the study design.
- Name the study so it can be identified in reporting later.
- Add the eligible campaigns that will run during the study.
- Review the Study Power feasibility estimate before launch, and select a compatible conversion action.
- Confirm that your measurement implementation is ready (see the next section) before you start the study.
Prepare conversion measurement before launch
Google recommends enhanced conversions for web and lead conversions, along with consent mode and related measurement improvements. Their purpose is to recover more observed conversion data and strengthen the modeling that supports the study. They do not guarantee a particular lift result, and they do not remove every measurement limitation. A study built on incomplete tagging will still have less observed data to work with, so treat this step as part of the study design rather than an optional extra.
Feasibility, certainty, and study design
What Study Power tells you
Study Power estimates the chance that the study will produce a conclusive result. Google says the estimate depends on:
- the conversion actions you select
- your daily budget
- the study duration
- the holdback percentage (the share of users kept in the control group)
- your account’s historical data
- the lift you expect to see
The estimate is shown as a range from 50% to 95% in 5-point increments. Google recommends aiming for 90% certainty. If the estimate is low, you can still run the study, but you should treat its outcome as directional and weigh that against your own risk tolerance.
How to read certainty
Google defines lift certainty as 1 minus the p-value. The reported certainty determines how a result should be read:
| Reported certainty | How Google reports it | How to read it |
|---|---|---|
| Below 50% | “No lift” | The study did not reach the threshold for a positive lift result. Google states this does not necessarily mean your ads were ineffective. |
| 50% to 90% | Directional result | Useful for direction, but not the level of confidence Google recommends for decisions. |
| 90% and above | Recommended certainty level | The level Google recommends aiming for when planning the study. |
Design choices that improve a readable result
- Include all eligible campaigns. This reduces the chance that control users see ads from your other campaigns, which would blur the comparison.
- Choose conversions that ads directly influence. Conversions with a weak connection to ad exposure dilute the signal.
- Extend the study up to 56 days where appropriate. Google suggests longer durations when you need higher certainty.
- Repeat with more conversion volume if needed. A study that returns a low-certainty result may need more conversions to be conclusive.
Do not use certainty as a ranking of business value across segments. Segment sizes differ, and confidence intervals can overlap. A segment with higher certainty is not automatically the best performer.
Metrics and how they differ from ordinary reporting
Lift metrics are calculated from the difference between the treatment and control groups. They are not the same as the attributed numbers in your standard reports.
| Metric | Calculation | Compare it with | Caution |
|---|---|---|---|
| Incremental conversions (absolute lift) | Treatment conversions minus control conversions | Attributed conversions, which use your attribution rules | Measures the difference between groups, not credit assigned to ads |
| Relative lift | Incremental conversions divided by control conversions | Relative lift from other studies | Can become very large when the control group has few conversions; use caution across studies with different control volumes |
| Incremental CPA (iCPA) | Total ad spend divided by incremental conversions | Attributed CPA | Reflects only the incremental conversions, so it is typically higher than attributed CPA when lift is smaller than attributed volume |
| Incremental conversion value | Treatment conversion value minus control conversion value | Attributed conversion value | Requires value tracking on the conversion action |
| Incremental ROAS (iROAS) | Incremental conversion value divided by ad spend | Ordinary ROAS, which uses attributed conversion value divided by spend | Not interchangeable with ROAS from attribution reports |
Some conversion types can be modeled through Supplementary Conversion Reporting, including store sales and offline leads, according to Google’s setup help. If you report a specific outcome, state whether it was directly observed or modeled.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Methodology: frequentist today, Bayesian in transition
Google’s methodology help says Conversion Lift studies are beginning to use Bayesian statistical methodology, with a gradual transition that began in 2025. Most accounts currently use frequentist methodology, and Google advises asking your account manager about transition timing for your account.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesThe two approaches report uncertainty differently. The Bayesian approach combines study data with historical campaign information such as campaign type, performance metrics, and product vertical, and it reports credible intervals with a probability interpretation. Frequentist results use confidence intervals. Label the method used by your account whenever you explain interval language, because the same phrasing does not carry the same meaning across both methods.
Delayed incremental conversions
Delayed incremental conversions are currently available only for Demand Gen-only studies. Google models conversions expected after the official study end date, based on the conversion lag observed during the study, and reports them as delayed incremental conversions where that feature is available. Do not assume this applies to studies that include other campaign types.
Scope of this guide
This guide covers user-based studies. Google also offers a geo-based Conversion Lift design, but its current requirements are not covered here. Check the Google Ads Help setup pages for geo-based studies before choosing a design, because the two designs differ in how groups are formed and in the eligibility rules that apply.
Because Google’s setup guidance can change, confirm eligibility, thresholds, methodology status, and menu labels in your own account before you commit budget to a study.
Free tools Windows power users keep installed
One-click scans. No signup required.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
