Google Opens Conversion Lift Tests to Search and PMax

Google now lets eligible advertisers run Conversion Lift studies on Search and Performance Max without a Google representative, per Search Engine Land on October 7.

If you run Performance Max or Search for a high-ticket store, this is the first time you can test how many of your attributed sales Google actually caused. The catch is the gate: Google requires 1,000 observed conversions and a $5,000 campaign budget, and a store that sells $3,000 sofas or $8,000 spas may not reach 1,000 on purchases alone. I cover high-ticket ad economics at Ecommerce Paradise, so here is who qualifies, what the test costs, and what to do if you don’t.

Below: what Google announced, how it got from six-figure minimums to $5,000, what the thresholds mean for a store with long sales cycles and phone orders, and a checklist for this week. If you are new to the model, start with my pillar guide to high-ticket dropshipping.

A $5,000 ad test is real money, and real money deserves a real entity: form your LLC with Northwest Registered Agent, the best LLC formation service for ecommerce owners. The address on an LLC filing is where plaintiffs and regulators look, and Northwest’s registered agent address keeps yours off the public record. Form your LLC with Northwest →

Google Conversion Lift Now Runs Self-Serve on Search and PMax

Search Engine Land reported on October 7, 2026 that Google Ads is extending user-based Conversion Lift to Search and Performance Max campaigns. Eligible advertisers can set up these studies themselves. According to the same report, Search and PMax studies previously had to go through a Google representative.

The requirements, per Search Engine Land and Google’s Conversion Lift help page: at least 1,000 observed conversions, not counting conversions that rely on supplementary data, at least one conversion action compatible with Conversion Lift, and a minimum campaign budget of $5,000. Google says the tool is not available to every account, and alpha and beta campaign types may still need a representative.

Here is how the measurement works. Google builds a holdout group that does not see the measured campaigns, then compares that group’s behavior with the users who can see the ads. The gap estimates incremental conversions, which is a different thing from the conversions an attribution model credits to the campaign.

You choose the holdback percentage, from 1% to 50%, per Google’s documentation. A larger holdout gives a better sample and costs more in withheld ads. A smaller holdout needs a longer study to detect a difference.

Studies can run as short as 7 days, though Google recommends at least 14. Google says studies with long conversion lags that run under 14 days may show reduced lift. The setup screen shows a study power from 50% to 95%, and below 90% Google treats the results as directional.

A few limits matter for planning. Each campaign can sit in only one lift study at a time, and a study cannot be restarted once it ends. Google also says supplementary conversions, such as store sales and offline leads, are modeled statistically rather than directly measured.

Geography-based Conversion Lift is a separate version. Google’s help page covers Search, Shopping and Performance Max for it, but labels it beta, says access is limited and says you still need an account representative. It requires campaigns to target a single country and states no minimum budget.

The announcement landed in a busy week for Google’s merchant tools. Google separately turns on automated promotions for advertisers, which I covered here, starting October 12.

How Google Got From $100K Incrementality Tests to $5,000

The $5,000 floor is not new. PPC Land reported on November 13, 2025 that Google lowered its incrementality testing minimum to $5,000, from thresholds the outlet described as approaching $100,000. John Chen, Senior Director of Product Management for Ads Measurement at Google, said the goal was to make incrementality testing easier to implement and more deeply integrated into measurement strategies.

That November update also promised statistical improvements that Google said deliver up to 50% more conclusive results, plus feasibility ratings tied to budget recommendations. Google said high-rated studies have a 60% to 90% chance of a conclusive result and low-rated ones have 0% to 30%. At the time, Conversion Lift for Video, Discovery and Demand Gen worked without a representative, while Display, Search, Shopping and Performance Max still required one.

So the October 7 change closes the gap for Search and Performance Max. Google also cited a January 2025 survey of 567 senior marketing analytics professionals at advertisers spending more than $500,000 a year, in which 80% said applying incrementality insights had a high impact on revenue growth. That is a Google-cited figure, not independent research.

Testing inside Performance Max is not new either. In a February 23, 2023 post, Tim Frank, Senior Director of Product Management for Google Ads, described PMax Experiments, including an uplift test and a head-to-head test against Standard Shopping. Google said Dutch and Belgian marketplace Bol.com reported a 64% revenue increase over Standard Shopping after the holiday season using that feature.

Google also lowered the bar for Demand Gen view conversions, which I broke down in my Demand Gen view conversions post. On the Merchant Center side, Google opened a checkout hub the same week, covered in my UCP hub write-up.

On the demand side, the Conference Board reported on September 29 that its consumer confidence index fell to 81.9 in September, down 6.7 points from August, according to GV Wire’s coverage. Economists polled by Reuters had expected 89.2, per the same report, and the reading is the lowest since 2014. Dana Peterson, chief economist at the Conference Board, said consumer appraisals of current business conditions became negative for the first time since September 2024.

I found no published criticism of this specific rollout as of Thursday morning. Coverage so far is limited to Search Engine Land’s report and Google’s own documentation, and Google has not attached a named spokesperson to the change in anything I could read.

What 1,000 Conversions Means for a High-Ticket Store

This section is my read, not reporting.

My read is that incrementality is the most valuable number a high-ticket advertiser never sees. Performance Max and branded Search are good at showing up where a buyer was already headed. A buyer researching a $6,000 hot tub types your brand name, clicks your ad and gets credited to the campaign, even if the order was coming anyway. Attribution reports cannot separate that order from one the ad truly created, and a user holdout can.

The first problem is volume. Take a hypothetical store with 60 Google-attributed orders a month. If only completed purchases counted, it would need about 17 months of orders to reach 1,000. Google’s page, as I read it, does not state a lookback window, so check eligibility in your own account before you plan around any of this.

The workaround is a higher-volume conversion action, such as a lead form or a tracked call. That can get you over 1,000, and then you are measuring leads, not margin. A lift study on form fills tells you Google caused form fills. It does not tell you those fills closed at the rate your phone team assumes, so pair the result with your close rate from my high-ticket sales follow-up guide.

Phone orders are the second problem. Google says offline leads are modeled statistically and conversions with supplementary data do not count toward the 1,000 minimum. Plenty of high-ticket stores close a large share of orders on the phone. If your calls never make it into Google Ads as clean conversions, your holdout study will understate what the ads do. That is an argument for fixing tracking before you test, and the pieces are laid out in my breakdown of a complete conversion system.

Now the cost, with hypothetical numbers. Say one campaign shows $20,000 in attributed revenue over a 14-day study and you hold out 20% of users. At most, $4,000 of attributed revenue sits in the group that never sees the ads. If half of those orders were truly incremental, you give up roughly $2,000 in revenue, which at a 25% gross margin is about $500 in gross profit. That is a small price for knowing whether the other $18,000 to $20,000 was bought or borrowed.

Here is the rule I’d set before launching, and it is mine, not Google’s. If the study shows that under half of the attributed conversions were incremental, I’d cut that campaign’s budget by about a third and retest. If most of them were incremental, I’d hold the budget and look for more volume in the feed, as I explained in my feed-only PMax guide. And if the power reading lands under 90%, I’d treat the result as a hint and extend the study, not rewrite the budget.

The honest caveat is the conflict of interest. Google sells the ads and runs the experiment, so I’d treat one study as one data point and not as a verdict. Google’s documentation does not address that tension, and I have not seen independent validation of the self-serve version yet. One study is also one season, and the Conference Board numbers say household demand is softening heading into the holidays.

Softer demand makes a wasted dollar more expensive. The New York Fed’s tariff pass-through research, which I summarized in my NY Fed tariff post, points to price effects that take up to a year to land, which squeezes margin from the other side. A campaign that looked fine at a 5x attributed return can look very different once you strip out the orders it did not cause.

There is a smaller upside too. A clean lift result gives you a number you can use to size every other channel, from CTV to email. Pair it with higher average order value and you move the whole math, not one line of it.

Many stores do not have the tracking, the feed and the phone workflow wired well enough to pass a lift study on the first try. If you’d rather not build that yourself, my team can build and run the entire store through the turnkey done-for-you service.

Want a high-ticket store with the conversion tracking, feed and phone workflow built to clear Google’s lift thresholds from day one? My team builds and runs it for you. See the turnkey done-for-you service →

Conversion Lift Checklist for High-Ticket Google Ads Budgets

Do these five things this week, in this order:

Count your conversions. In Google Ads, pull the last 12 months of your primary conversion action for your biggest Search or PMax campaign. If it is under 1,000, plan on not getting the self-serve study, and read steps 2 and 5 anyway.
Fix call and lead tracking. Put a tracked number on the site with Dialpad, then log every closed phone order in a CRM. My HubSpot setup guide shows the free version for a high-ticket store.
Pick one campaign to test. It needs the $5,000 budget minimum, and it should be a campaign where you suspect brand searches are soaking up credit. Pull your branded versus non-branded query split in Semrush first so you know what you are testing.
Freeze everything else. Run at least 14 days, and I’d go 21 for products with long research cycles. Do not change prices, bids, the feed or your Klaviyo flows mid-study, because any of those muddies the comparison.
Write your decision rule before you launch, then judge the result on margin, not platform ROAS. Pull contribution margin from your books in Finaloop. Then have Claude compare the lift numbers against your real cost per order.

If you cannot qualify or you would rather not run this alone, book a call at my discovery page and I’ll tell you where I’d look first.

Stores that want the whole ad program run for them can look at the scaling service. If you are still choosing your Google Shopping approach, my guide to turning Google Shopping clicks into sales is the next read.

Frequently Asked Questions

What is Google Ads Conversion Lift?
It is a measurement tool that holds back a share of users from seeing your ads, then compares their conversions with users who saw them. The difference estimates the conversions your ads caused, per Google’s help documentation.

Who qualifies for the self-serve Search and PMax studies?
Per Search Engine Land and Google, you need at least 1,000 observed conversions (not counting supplementary-data conversions), a compatible conversion action and a $5,000 campaign budget. Availability varies by account, so check inside Google Ads.

Does a holdout test cost me sales?
It can, because users in the holdout do not see your ads. Google lets you pick a holdback from 1% to 50%, so a smaller holdout limits the cost but needs a longer study.

Is Shopping covered by the October 7 change?
Search Engine Land’s report names Search and Performance Max only. Google’s November 2025 update listed Shopping as needing a representative, and geography-based Conversion Lift, which covers Shopping, is still beta and limited.

What if my store only does a few hundred orders a year?
Then you will not hit 1,000 on purchases, and your best move is tighter tracking and a bigger-ticket product mix. My free high-ticket niches list is a good place to find categories with more room to scale. The full breakdown sits in my niche pillar guide.

Will this replace my Shopify or analytics numbers?
No. Keep reconciling Google’s results against the actual orders in your Shopify admin, and use the lift study as a check on attribution, not a replacement for it. For the post-click side, an email platform like Omnisend shows what happens to buyers who did not convert on the first visit.

Should I run a study before the holidays?
My read is yes, if you qualify, because a 14 to 21 day study started now finishes before peak spend ramps. Compare options in my performance platforms roundup if you want a second channel to test next.

Want to work through your own lift test with other store owners and me inside the community? Join the Skool community →

Run the count first. It takes ten minutes and tells you whether this test belongs on your calendar. Subscribe to the YouTube channel for daily breakdowns. More breaking news coming through the day.

If this was useful, these go deeper:

Trevor Fenner

Trevor Fenner is an ecommerce entrepreneur and the founder of Ecommerce Paradise, a platform focused on helping entrepreneurs build and scale profitable high-ticket ecommerce and dropshipping businesses. With over a decade of hands-on experience, Trevor specializes in high-ticket dropshipping strategy, niche and product selection, supplier recruiting and onboarding, Google & Bing Shopping ads, ecommerce SEO, and systems-driven automation and scaling. Through Ecommerce Paradise, he provides free education via in-depth guides like How to Start High-Ticket Dropshipping, advanced training through the High-Ticket Dropshipping Masterclass, and fully done-for-you turnkey ecommerce services for entrepreneurs who want a faster, more hands-off path to growth. Trevor is known for emphasizing sustainable, real-world ecommerce models over hype-driven tactics, helping store owners build scalable, sellable, and location-independent brands.