How do I measure Performance Max?
Measure Performance Max in two layers. Google Ads' channel performance and search terms reports show where it served and what it claims. An uplift experiment or a geo Conversion Lift study shows what it added. The first is a claim; the second is a test.
By Joris van Huët, Founder & CEOUpdated 7 min read
Run the numbers for your store: the free break-even ROAS calculator.
Measure Performance Max in two layers. Its own reports, mainly channel performance and search terms, show where it served and which conversions it claims. A test shows what it added: an uplift experiment, or a geo Conversion Lift study if Google gives you access. If it mostly buys your own brand searches, its claims usually overstate what it added.
What does Performance Max report, and what does it count?
The usual answer: open the campaign, read its conversions and ROAS, then check the channel report. Fair enough. Those reports are real, and they describe Google's count by Google's rules.
Google's channel performance report splits a Performance Max campaign across Search, Maps, YouTube, Discover, Gmail and the Display Network. It covers dates after June 6, 2025, and each conversion follows the attribution model set on its conversion action. Its Ad event type segment shows what came before each conversion: an engaged view, a click, an interaction or an impression.
The search terms report lists the searches that triggered Performance Max ads, with Shopping and text ads told apart. Its data starts in March 2023.
Then the windows. By default, Google Ads counts a conversion for 30 days after a click, 3 days after an engaged view and 1 day after a view.
GA4 sees less of all this. Google's help page on Performance Max in Analytics says the campaign counts engaged-view conversions, which Google Analytics leaves out by default. GA4 also files its visits under Cross-network, not Paid Search.
What one store's data shows
Store A is one store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.
| What the export shows | Share of revenue | Source cell |
|---|---|---|
| Direct, in last click, first click and touched views | 57.7% | Channels sheet, Direct row |
| Journeys with 1 touch (0.5 days to buy) | 79.5% | Journeys sheet, 1 touch row |
| Journeys with 2 to 3 touches (12.5 days to buy) | 12.2% | Journeys sheet, 2 to 3 touches row |
| Journeys with 4 to 9 touches (16.9 days to buy) | 5.4% | Journeys sheet, 4 to 9 touches row |
Read the Channels sheet first. On the Channels sheet, Direct holds 57.7% of revenue in last click, first click and touched views. Those buyers came back through a saved link or a typed address. A buyer who knows your name can also search for it. If Performance Max serves on that search, the sale moves into Google's column without one extra buyer.
Then the Journeys sheet. Journeys with a single touch hold 79.5% of revenue on the Journeys sheet, at 0.5 days to buy. One touch leaves no room for an assist. If a Performance Max click was that touch, it takes the whole sale in GA4; if not, it gets nothing.
The slow lane is smaller. On the Journeys sheet, journeys of 2 to 3 touches hold 12.2% of revenue at 12.5 days to buy. Journeys of 4 to 9 touches hold 5.4% on the Journeys sheet, at 16.9 days. With the 10 or more row, journeys of 2 or more touches hold 20.6% on the Journeys sheet (12.2% + 5.4% + 3.0%).
That fifth is where YouTube, Display and Discover would have to earn their keep, by starting journeys that other channels finish. The rest of the money arrives from people who already knew the way.
The export holds no spend, so it cannot say whether this store ran Performance Max, or what it paid for any of this.
Why do Performance Max's own numbers mislead?
Because it is allowed to stand where buyers already are. Google's help page says an exact match keyword in a Search campaign takes priority over Performance Max. It adds that Performance Max may serve on exact match terms when that Search campaign is limited by budget. Without brand exclusions, your own name is fair game.
Google is candid about what those clicks are. Its brand exclusions page talks about avoiding paying for clicks from users who are already navigating to your site. When those users buy, the campaign still counts the conversion.
It also counts touches GA4 does not. An engaged view of its video can claim a conversion for 3 days by default, and GA4 leaves those out by default. And unless you switch on a customer acquisition goal, a regular's reorder counts the same as a first purchase.
Google's own evaluation guide warns that the average return of one channel over a short timeframe does not tell the full story. True. Neither does the average return of the whole campaign.
What can the report not tell you?
What would have happened without it. That takes a control group, and Google offers two ways to get one.
A Performance Max uplift experiment splits traffic, 50% by default, between your campaigns with and without it. It reads the result in Google Ads conversions. So a buyer who would have come through organic search can still count for the arm that showed the ad.
Conversion Lift based on geography compares regions that see your ads with regions that do not. Google pitches it for the total, unattributed, incremental return of your campaigns. It supports Performance Max and needs campaigns that target a single country. Access runs through a Google account representative.
If that door stays shut, run a geo holdout of your own. Pause the campaign in some regions and compare total orders, not attributed ones.
What to do this week
- Split last month's conversions by Ad event type. In Google Ads, go to Campaigns, open Insights and reports, select Channel performance, then use Segment on the table toolbar. Pass: most conversions follow clicks. Fail: a large share follows engaged views or impressions, which GA4 does not count by default.
- Measure the brand share of its searches. Open Search terms in the Campaigns menu and choose Search terms and landing pages for Performance Max. Pass: searches for your brand hold a small share of its conversions. Fail: they hold a big share, so add brand exclusions if a brand Search campaign already covers them.
- Book a test before the next budget change. Go to Experiments in the Campaigns menu and set up a Performance Max Uplift experiment, or ask your Google representative about geo Conversion Lift. Pass: a test with a control group is on the calendar. Fail: no test is possible, so hold the budget flat and run a geo holdout of your own.
Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework
Sources, 1 October 2026: About the channel performance report for Performance Max (Google Ads Help); About the search terms report in Performance Max (Google Ads Help); About conversion windows (Google Ads Help); Performance Max reporting in Google Analytics (Google Ads Help); Default channel group (Google Analytics Help); About Performance Max campaigns (Google Ads Help); About brand exclusions (Google Ads Help); Evaluate Performance Max Results (Google Ads Help); Measure your lifecycle goals campaigns (Google Ads Help); Set up Uplift experiments for Performance Max (Google Ads Help); About Performance Max experiments (Google Ads Help); Set up Conversion Lift based on geography (Google Ads Help); Understand your Conversion Lift based on users measurement data (Google Ads Help).
Related answers
Frequently asked questions
What counts as a Performance Max conversion?
A conversion that one of its conversion actions records inside its window after an ad interaction. Google's channel report labels the event before each one: Engaged-View, Clicks, Interaction or Impression. By default a click can claim a conversion for 30 days.Should I test Performance Max against my Shopping campaign?
Google offers that as an upgrade experiment. It simulates moving a Standard Shopping budget into Performance Max and compares the two before you commit. It tells you which campaign type does better on Google's count, not whether either adds sales.Is a Performance Max uplift experiment a lift test?
Not quite. An uplift experiment compares Google Ads conversions with and without Performance Max, split 50% by default. A Conversion Lift study compares all conversions between people or regions that saw ads and those held back, ignoring attribution rules.
Go deeper: Incrementality testing, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Keep reading
Terms in this article
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- Attribution ModelAn Attribution Model defines how credit for conversions is assigned to marketing touchpoints. It dictates how marketing channels receive credit for sales.
- Control GroupControl Group is a segment of an audience intentionally not exposed to a marketing campaign, used to measure the campaign's true causal impact.
- Customer acquisitionCustomer acquisition attracts new customers to a business. For e-commerce, this means driving the right traffic to the website.
- Google AnalyticsGoogle Analytics is a web analytics service that tracks and reports website traffic.
- IncrementalityIncrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
- Incrementality TestingIncrementality Testing measures the additional impact of a marketing campaign. It compares exposed and control groups to determine causal effect.
- Landing PageLanding Page: A single web page that appears after clicking a search result, marketing promotion, email, or online advertisement.