Google’s AI Max for Search Is Forcing DTC Brands to Rethink Keyword Bidding

When Google quietly expanded AI Max for Search campaigns to all U.S. advertisers in late June 2026, the rollout landed with a thud inside some of the most active Shopify merchant Slack groups. Within days, performance marketing leads at brands ranging from mid-tier apparel labels to specialty home goods stores were sharing screenshots of blown CPCs, mismatched search term reports, and conversion rates swinging 30% in either direction. The consensus: AI Max is not a passive upgrade. It rewrites the rules of keyword-level control that DTC brands have spent years tuning.

AI Max for Search is Google’s most aggressive push yet toward fully automated campaign management. It combines broad match expansion, automatically generated headlines and descriptions via asset generation, and a new “URL expansion” feature that directs traffic to landing pages Google’s model selects — not necessarily the ones advertisers specify. For DTC operators who’ve built tightly structured Shopping and Search stacks, the implications are significant.

What exactly is AI Max for Search and how does it differ from Smart Campaigns?

AI Max is not a repackaging of Smart Campaigns or Performance Max, though it shares DNA with both. It operates within standard Search campaigns, adding an opt-in layer of AI-driven features including query expansion beyond exact and phrase match, creative asset generation, and the contested URL expansion. The key distinction is that advertisers retain their campaign structure — ad groups, bidding strategies, negative keywords — while Google’s model takes over match-type logic and creative assembly in real time.

According to Google’s internal benchmarks cited in its May 2026 advertiser briefing, early beta participants saw a median 14% increase in conversions at similar CPA. But those numbers mask significant variance. Several mid-market DTC brands running tightly managed Search campaigns have reported that URL expansion is routing high-intent branded queries to collection pages rather than product detail pages — a problem that directly tanks conversion rate on paid traffic.

“We turned on AI Max across three of our best-performing Search campaigns in early June and within 72 hours our branded CPC jumped 28% and Google was sending ‘best running shoes men’ traffic to our homepage. That’s not an upgrade — that’s a regression.” — Brett Nolan, Head of Performance Marketing, Lume Athletics (Chicago)

💡 Article Summary

Key Insights

1

What exactly is AI Max for Search and how does it differ from Smart Campaigns?

2

Which DTC brands are seeing real gains — and which are getting burned?

3

How does URL expansion interact with Shopify store architecture?

4

What are the conversion rate implications for Search traffic quality?

5

How should DTC operators and agencies structure AI Max tests going forward?

Source: Ecommerce Times

Which DTC brands are seeing real gains — and which are getting burned?

The split appears to correlate closely with catalog complexity and campaign maturity. Brands with smaller SKU counts, limited historical Search data, and newer accounts are reporting stronger early results. Google’s model has more room to optimize when there’s less incumbent structure to disrupt. A DTC skincare brand running fewer than 40 SKUs told Ecommerce Times its cost per acquisition dropped from $38 to $29 in the first 30 days after enabling AI Max — without any manual intervention.

Larger operators with 500-plus SKUs, multi-brand portfolios, and years of negative keyword lists are a different story. For them, AI Max’s query matching logic is surfacing irrelevant terms that their Search campaigns had long suppressed, effectively re-opening the funnel to low-intent traffic at premium CPCs.

“Our client has 11 years of Search history and a negative keyword list that’s 4,000 terms deep. AI Max ignored about 60% of it in the first week. We had to manually exclude 200 new terms in seven days. That’s not AI efficiency — that’s manual labor with a new name.” — Dana Okafor, Senior Paid Search Lead, Metric Theory

The agency perspective is particularly sharp. Metric Theory, Tinuiti, and Wpromote have all issued internal guidance to account teams recommending selective, monitored rollouts of AI Max rather than blanket adoption — a notable contrast to the broad Performance Max enthusiasm of 2023 and 2024.

How does URL expansion interact with Shopify store architecture?

This is where Shopify merchants are running into the most operationally specific problems. URL expansion allows Google’s AI to redirect clicks to any indexable page on the domain it deems most relevant to the query. For Shopify stores with standard URL structures — /products/, /collections/, /pages/ — the model is making routing decisions based on its own crawl data, not the merchant’s conversion-optimized funnel.

Several Shopify Plus merchants have reported that AI Max is sending paid traffic to blog posts and “about” pages when their product pages lack sufficient on-page copy — a problem that loops back directly to SEO fundamentals. Stores with thin product descriptions, minimal structured data, or duplicate content across variants are particularly exposed.

• URL expansion can be disabled at the campaign level — Google buries this toggle in the AI Max settings panel, but it is accessible and several agencies are recommending turning it off by default on mature accounts
• Page feed uploads remain the most reliable way to constrain which URLs Google can select — Shopify merchants using a supplemental feed via DataFeedWatch or Feedonomics have more control than those relying on auto-crawl
• Negative keyword lists at the account level still apply to AI Max campaigns, but campaign-level negatives are being interpreted more loosely according to multiple advertiser reports — Google has not officially confirmed this behavior
• Asset generation quality varies dramatically by product category — apparel and beauty are generating usable headlines; home improvement and technical B2B products are producing generic, off-brand copy that requires heavy override

What are the conversion rate implications for Search traffic quality?

The CRO downstream effects are proving harder to isolate than the CPC data. When AI Max expands match types and routes URLs dynamically, it changes both the traffic composition and the landing experience simultaneously — making it difficult to attribute conversion rate changes to either variable independently.

Northbeam and Triple Whale users have flagged that their attribution models are showing increased click volume from Search with flat or declining revenue attribution, suggesting that AI Max is generating engagement that doesn’t complete purchase. One apparel brand using Northbeam reported a 22% increase in Search clicks in the three weeks post-AI Max enablement, with attributed revenue up only 4% — implying a meaningful traffic quality dilution.

“The click volume looks great in Google Ads. The revenue doesn’t match. When you dig into Northbeam, you can see the Search traffic quality dropped. We’re getting more top-of-funnel searchers who weren’t looking to buy. AI Max is treating them the same as high-intent shoppers.” — Priya Mehta, CMO, Vestry Home (Austin)

Vestry Home, a direct-to-consumer furniture accessories brand doing approximately $18M in annual revenue, has since dialed back AI Max to a single low-volume test campaign while the team rebuilds its product page copy and structured data to better signal purchase intent to Google’s crawler.

How should DTC operators and agencies structure AI Max tests going forward?

The emerging playbook from agencies and merchants who’ve navigated the first 45 days of broad availability breaks into several clear phases:

• Phase 1 — Audit first: Before enabling AI Max on any active campaign, audit landing page quality scores, on-page copy depth, and existing negative keyword coverage. Pages with Quality Scores below 6 are more likely to receive misrouted AI Max traffic
• Phase 2 — Isolate a test campaign: Enable AI Max only on a campaign with limited budget ($50–$150/day) and non-branded keywords, keeping existing campaigns fully intact to maintain a clean comparison baseline
• Phase 3 — Disable URL expansion initially: Run the first 30 days with URL expansion off. Evaluate click quality and conversion rate before reintroducing it with a tightly scoped page feed
• Phase 4 — Monitor search term reports daily for the first two weeks: AI Max’s query expansion moves fast. Daily reviews are necessary, not weekly — agencies running weekly check-ins have reported significant wasted spend before catching mismatches
• Phase 5 — Evaluate creative asset performance separately: Google’s AI-generated assets should be audited for brand voice compliance. Several brands have found the model generating price-focused headlines for premium positioning brands — a tone mismatch that undermines brand equity even if the click converts

Is AI Max a signal of where Google Shopping is heading next?

The broader context matters here. AI Max for Search is not an isolated product decision — it’s part of Google’s multi-year campaign to reduce advertiser control over keyword-level mechanics in favor of signal-based, outcome-driven automation. Performance Max already removed campaign-type-level transparency. AI Max is now applying the same logic to standard Search, historically the most transparent campaign type in the Google Ads ecosystem.

For DTC brands that built their acquisition stack on granular keyword bidding — separating brand vs. non-brand, exact vs. phrase, product-level campaigns with dedicated landing pages — this is a structural shift, not a feature update. The brands most exposed are those whose competitive advantage came from better keyword architecture rather than better product economics or creative quality.

Google’s response to criticism has been consistent: the platform argues that AI-driven systems outperform human keyword management at scale when given sufficient conversion signal. That may be true for large-volume advertisers. For Shopify merchants doing $2M to $15M in revenue with 60 to 90 conversions per month per campaign, the conversion data volume required for Google’s model to optimize effectively often isn’t there — and the model fills the gap with expansion logic that may not match the merchant’s actual customer profile.

Ryan Kovacs, VP of Growth at digital agency Hawke Media, summarized the operator dilemma bluntly:

“Google wants you to trust the black box. Some of our clients should — they don’t have the internal resources to manage keyword lists properly anyway. But our sophisticated DTC clients? They’ve spent three years building keyword architecture that reflects how their actual customers search. AI Max doesn’t know what they know. The question is how long before Google takes away the option to keep doing it their way.”

That timeline is unclear, but the direction is not. Google has already sunset several manual bidding options over the past four years. For DTC operators still running tightly controlled Search campaigns, the window to extract full value from that architecture — before Google’s automation layers make it moot — may be narrowing faster than most are prepared for.