Google now lets Demand Gen advertisers opt out of AI-suggested Lookalike audiences and keep them as a hard targeting limit, Search Engine Land reported on October 12.
If you run Demand Gen for a high-ticket store, this decides whether your ads reach people who resemble your buyers or people Google predicts might convert. It touches your cost per lead, your lead quality and the phone calls your sales process has to absorb. I cover this kind of change daily at Ecommerce Paradise, and this one is easy to miss because nothing in your account looks different until the reach starts drifting.
Below is what Google published, how Lookalike segments got here, what I’d do if your orders run $1,500 and up, and the five checks to run this week. If you want a refresher first, my guide to high-ticket dropshipping covers the model this advice assumes.
Ad accounts, customer lists and card processing all sit under one legal name, so make it an LLC and not you personally. Form your LLC with Northwest, the best LLC formation service for ecommerce owners. Plaintiffs and regulators start with the address on an LLC filing, and Northwest’s registered agent address keeps your home address off that record. Form your LLC with Northwest →
Google Documents the Demand Gen Lookalike Opt-Out
Search Engine Land published the report on October 12, 2026. Writer Anu Adegbola said Google is moving Lookalike segments in Demand Gen toward “a suggestion mode by default,” according to Search Engine Land. Advertisers who want the current behavior can opt out and keep using Lookalike segments as a targeting constraint, which limits delivery to the selected audience.
Google’s own help page, titled “Use Lookalike segments in Demand Gen campaigns as a Targeting Constraint,” says the default behavior “will evolve towards a suggestion mode.” It tells advertisers to opt out through a form linked on the page. Per Google Ads Help, the page gives no date for the mode change and no opt-out deadline.
The mechanics are unchanged. Per the same help page, a Lookalike segment is built from first-party seed data: customer lists, website or app activity and YouTube engagement. You choose a reach tier. Narrow targets about the most similar 2.5% of users in your ad’s target location, Balanced about 5% and Broad about 10%.
The seed list floor is 100. The combined active matched people across all seed lists must exceed 100, and segments refresh every one to two days. If a list drops under the minimum, the segment becomes unavailable, though Google says it can take up to three days to stop targeting.
Two limits on the setup matter for planning. Segments can only be created while you build a Demand Gen campaign, and per Google they cannot be edited or deleted in Audience Manager. Google recommends creating the campaign and segment two to three days before you need ads to serve.
Search Engine Land noted that broader targeting does not automatically perform better and told advertisers to track conversion quality, acquisition costs and customer value. The article does not describe the opt-out steps and gives no rollout date. It only advises reviewing the option “before the default changes.”
I could not confirm from public pages whether an in-interface toggle has shipped yet. The February coverage said one was expected later in 2026, so check your campaign settings before assuming the form is the only route.
How Lookalike Segments Went From Hard Limits to AI Hints
This is not new. Search Engine Land reported on February 17, 2026 that Google was converting Lookalike segments in Demand Gen from strict targeting to optimization signals, effective March 2026. Under that model the tiers act as guidance, and Google’s system can reach beyond the list to find users it predicts will convert.
According to PPC Land’s February report, advertisers who wanted the old behavior could request an opt-out through a form that takes about a week to process. The same report said in-interface opt-out controls were expected later in 2026. Google also said that when Lookalike suggestions and Optimized Targeting are both on, it expects more of the performance gains to come from Lookalike suggestions.
Here is a timeline conflict worth flagging. The February reports described a March 2026 effective date. The October 12 article and Google’s current help page both still say the default will “evolve” toward suggestion mode. I can’t tell from public sources whether your account was switched in March or is still waiting, and I’d treat that as unknown until you check the audience reporting table. PPC Land said a “signal” tag in that table marks segments already running in suggestion mode.
The reaction split. Google Ads consultant Adriaan Dekker called the change another step toward AI-led targeting, similar to Meta’s direction, and noted that the opt-out form suggests Google expects many advertisers won’t want full expansion by default. Brian Lasonde made a similar point. Manoj Kumar, CEO of Rankdriven, welcomed the shift and argued that AI-led targeting is the future of performance optimization.
That is the counterpoint, and it deserves weight. For a store with a deep conversion history, wide reach can find buyers a rigid 2.5% slice never touches. The question is whether your account has the data to make that trade pay.
Demand Gen has been getting a steady stream of changes. I covered Google lowering the bar for Demand Gen view conversions. I also covered the new ability to overlay text on Demand Gen videos. Each one nudges the system toward more automation and less manual control, and this one follows the same direction.
Why Strict Lookalikes Can Protect a High-Ticket Store
This section is my read, not reporting. Everything above is sourced. Everything below is how I’d think about it.
A high-ticket store lives on a small number of expensive conversions. If your average order is $2,500 and you close 30 to 60 orders a month, your seed list of past buyers is tiny compared with a store selling $40 items. Google’s own floor is 100 matched users, so a seed list built from 300 buyers sits barely above the line. A model working from that little signal and told to expand beyond it is guessing more than it is learning.
Take a hypothetical to see the stakes. Say you spend $6,000 a month on Demand Gen at a $90 cost per lead, which gives about 67 leads. If you close 10% of leads at a $2,500 order value, that is about 7 orders and roughly $17,500 in revenue. Now suppose suggestion mode lowers your cost per lead to $70, which gives about 86 leads, but lead quality falls and you close 6%. That is about 5 orders and $12,900 in revenue. These numbers are invented to show the mechanism. The cheaper lead lost money because the close rate fell faster than the cost per lead.
That is the trap with automated expansion. Platform reporting rewards the cheapest conversion it can find. A form fill from someone browsing on a phone at midnight counts the same as a form fill from a homeowner ready to spend $4,000. Your CRM knows the difference. Google’s dashboard doesn’t, unless you feed the real outcomes back. My piece on why high-ticket Google Ads need a complete conversion system explains how to do that.
There are two scenarios with thresholds. If you have fewer than roughly 500 matched buyers in your seed lists and you sell above $1,000, I’d keep the targeting constraint and watch lead quality for 30 days. If you have thousands of matched buyers, a clean offline conversion feed and a close rate you track weekly, I’d test suggestion mode on a split budget and let the numbers decide. Those thresholds are my judgment, not Google guidance.
Lookalike audiences also depend entirely on the quality of what you upload. A list stuffed with refund requests, tire-kickers and one-time low-value orders teaches the model the wrong lesson. Clean exports from Shopify and your email platform are the cheapest improvement available. A tool like Klaviyo makes it simple to segment past buyers by order value before you export.
Where Demand Gen sits in the funnel also changes the stakes. It is a discovery channel, closer to social than search. My comparison of ad platforms for high-ticket dropshipping argues that high-intent search and Shopping carry the load, while Demand Gen and its cousins feed the top. If Demand Gen wanders into low-intent audiences, the damage shows up as wasted spend and a flooded phone line, not as a visible error.
Budget discipline matters here too. Google Ads pros I covered recently argue that budget is the new Quality Score, which means the system gives more room to accounts that feed it spend. That favors large advertisers. A store spending a few thousand dollars a month has less room to absorb a month of loose targeting.
If you are running Demand Gen alongside other paid channels, the same logic applies. Meta’s past-buyer exclusion exists because wasted reach on existing customers is a real cost. Hard audience limits are a cost control, and expansion mode trades some of that control for scale you may not need.
Managing Lookalike seeds, offline conversion imports, split tests and a CRM feed is real operator work, and it competes with sourcing, supplier calls and customer service for your week. If you would rather hand the build and the ad setup to a team that does this daily, my turnkey done-for-you service builds and runs the whole high-ticket store.
Not sure whether strict or suggested Lookalikes fit your store? Walk through your Demand Gen setup with me one on one and leave with a plan. Get the coaching details →
Check Your Demand Gen Lookalike Settings This Week
These are the five moves I’d make before Google changes the default on your account.
- Open every active Demand Gen campaign and find the Lookalike segments. In the audience reporting table, look for the “signal” tag that PPC Land said marks segments in suggestion mode, then write down which campaigns are already affected.
- Pull a 30-day comparison of cost per lead and lead-to-sale rate by campaign from your CRM, such as HubSpot, and not from Google’s dashboard. If the close rate fell while the cost per lead dropped, you have the pattern from the example above.
- Count your matched seed users. If any list sits near Google’s 100 floor, rebuild it from a cleaner export that excludes refunds, chargebacks and low-value orders, and add past buyers by order value.
- Decide per campaign: keep the constraint through Google’s opt-out form, or split budget 70/30 between constraint and suggestion mode for 30 days. Allow about a week for the form to process, per PPC Land’s February report. Review the Turnkey service details and email me with questions before ordering.
- Track profit per lead, not only cost per lead. A bookkeeping tool such as Finaloop keeps your margin visible so a cheap lead that doesn’t close doesn’t look like a win. Pair it with keyword and search-term checks in Semrush to confirm your high-intent search campaigns still carry the account.
If you are still choosing a niche and don’t have a buyer list yet, none of this applies to you today. Start with my high-ticket niches list. Then grab the free niches list and build the store first.
Frequently Asked Questions
What is a Lookalike segment in Demand Gen?
It is an audience Google builds from your first-party seed data, such as customer lists, website visitors, app users or YouTube engagement. Google then finds users with similar characteristics, per its help page.
What does suggestion mode do?
It treats the Lookalike tier as a starting point rather than a boundary, so Google’s systems can serve ads beyond the 2.5%, 5% or 10% pool if they predict better conversions, according to Search Engine Land’s February report.
How do I keep the old behavior?
Opt out through the form linked on Google’s help page and keep Lookalike segments as a targeting constraint. PPC Land said that form takes about a week to process.
Does this affect Google Shopping and search campaigns?
The reporting covers Demand Gen only. For Shopping, my guide on turning Google Shopping clicks into sales is the better starting point. Also read about price-change disapprovals.
Should I switch to TikTok or Meta instead?
Not because of this alone. TechCrunch reported on October 5 that TikTok rolled out an AI shopping assistant and one-click checkout, but one-click impulse buying suits low-ticket items better than a $3,000 purchase. Read what store owners must do after Meta’s TikTok ad ban before you rebuild a channel mix.
How do I stop cheap leads from wasting my sales time?
Follow up fast and score by order value. My post on high-ticket sales follow-up covers the process.
Can ads push chargebacks up?
Broader targeting can bring buyers with weaker intent, and that can raise disputes. See Visa’s wider chargeback evidence rules and prepare your records.
Already have a store and want it scaled the right way, with ads managed by people who track profit per lead? See the scaling service →
Google will keep moving advertisers toward more automation, and the stores that win are the ones that know their own numbers before the default changes. Subscribe to the YouTube channel for daily breakdowns. More breaking news coming through the day.
Related Articles
If this was useful, these go deeper:
- Google Ads Tests Typing Your Goal to Build Campaigns
- Google AI Mode Ads Run Through Your Existing Campaigns
- ChatGPT Ad Clicks Cost 6x Google’s, Measured Estimates
- Ecommerce Agency: How to Choose One in 2026
- Shopify Now Shows AI Agents Your Compare-At Prices
Trevor Fenner is an ecommerce entrepreneur and world traveler from Seattle, Washington, 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.
He maintains homes in Seattle and Bali, spends most of the year overseas, and usually returns to Seattle for about a month each summer.
