Google Is Force-Upgrading Your DSA Campaigns to AI Max

The Timeline: What’s Actually Happening and When#

Google has revised the DSA sunset timeline twice — the current deadline gives advertisers until February 2027, but two other legacy settings upgrade in September 2026.

On April 15, 2026, Google moved AI Max for Search out of beta and announced that all eligible DSA campaigns would be auto-upgraded by the end of September 2026. Six weeks later, on June 12, Google reversed course: the DSA-specific forced migration was pushed to February 2027 after advertisers requested more preparation time. But this extension only applies to DSA campaigns. Automatically Created Assets and campaign-level broad match settings will still be auto-upgraded starting September 2026.

The practical implication is that the September deadline isn’t gone — it’s narrower. If your WooCommerce store’s Google Ads account uses ACA or campaign-level broad match, those features convert to AI Max in September whether you’re ready or not. If you’re running DSA specifically, you have until February 2027 — but Google is actively encouraging manual migration before then to retain control over the transition.

Google postponed the DSA-to-AI Max forced auto-upgrade from September 2026 to February 2027, while ACA and broad match upgrades proceed in September as planned.

Once the auto-upgrade happens, you can’t create new DSA campaigns through Google Ads, Google Ads Editor, or the Ads API. DSA becomes a permanent legacy campaign type. Google’s one-click upgrade tools convert dynamic ad groups to standard ad groups with AI Max features enabled, preserving historical data within the same ad groups. Waiting for the forced upgrade means inheriting default settings instead of configuring them yourself.

You may be interested in: The Cumulative Tracking Loss Stack: When Ad Blockers Plus ITP Plus Consent Plus App Browsers Compound to Hide 50% of Your WooCommerce Customers

Google’s 7% Number vs What Independent Testing Shows#

The gap between Google’s reported performance and what independent researchers measured tells a more complicated story than the headline suggests.

Google reports that AI Max for Search campaigns see an average of 7% more conversions or conversion value at a similar CPA/ROAS when using the full feature suite — search term matching, text customization, and final URL expansion — compared to search term matching alone. For campaigns still tied to exact and phrase match, the reported uplift reaches 27%. These numbers are real, but they carry two caveats: the 7% figure comes from non-retail advertisers, and “similar CPA/ROAS” is never defined with a specific percentage.

Independent testing paints a different picture. An analysis of 250 AI Max campaigns found the real median outcome was 13% more revenue while CPA climbed 16%. Across a larger sample of 600 accounts, Smarter Ecommerce found that AI Max delivered 35% lower ROAS than traditional match types. In that same dataset, AI Max accounted for just 0.57% of total ad spend — indicating that advertisers kept the budget minimal.

Source	Sample Size	Conversions / Revenue	CPA / ROAS Impact
Google (internal, non-retail)	Undisclosed	+7% conversions (full suite)	“Similar” CPA/ROAS
Google (exact/phrase match)	Undisclosed	+27% conversions	“Similar” CPA/ROAS
Smarter Ecommerce (250 campaigns)	250	+13% revenue	+16% CPA increase
Smarter Ecommerce (600 accounts)	600	Mixed	-35% ROAS
Lunio (900 campaigns)	900	N/A	+35% invalid traffic increase

An independent analysis of 250 AI Max campaigns found 13% more revenue while CPA climbed 16%, revealing a gap between Google’s headline and real-world outcomes.

The Lunio data adds another dimension: across 900 Search campaigns, invalid traffic rose from a 3.7% baseline to 5% with AI Max enabled — roughly a 35% relative increase. A wider targeting net catches more queries, but not all of those queries carry genuine purchase intent. For WooCommerce stores where the average order value matters more than click volume, that traffic quality shift can erode ROAS even when conversion counts look stable.

AI Max Needs Clean Signal — Most WooCommerce Stores Don’t Have It#

The algorithm optimizes against whatever data it receives — and for most WooCommerce stores, the data it receives is 30–40% incomplete.

AI Max requires a minimum of 100 monthly conversions reaching Google Ads for the algorithm to calibrate effectively. Below that threshold, Google’s documentation acknowledges the system lacks sufficient training data for reliable predictions. It doesn’t stop optimizing — it starts guessing. And advertisers pay for the guesses.

But volume alone isn’t sufficient. The conversions reaching Google Ads need to be accurate and complete. WooCommerce stores running client-side-only tracking — browser pixels, GTM tags, standard JavaScript snippets — lose between 30% and 40% of their purchase events before those events ever reach the platform. The causes are structural: ad blockers, Safari ITP cookie caps, Consent Mode failures, and in-app browser referral stripping each remove a slice of the conversion data. The compound effect is that AI Max builds its optimization model on a systematically biased sample of your actual customers.

Let that sink in. AI Max doesn’t know the signals are missing. It treats the 60–70% of conversions that survive the browser-side gauntlet as the complete picture. It optimizes your bids, your audience targeting, and your creative distribution based on a partial view of your customer — and it does so with full confidence.

The result is predictable: bid decisions calibrated to the wrong conversion rate, audience segments built from an incomplete customer profile, and budget allocation driven by the customers who were easiest to track rather than the customers who were most valuable.

What the Forced Upgrade Changes in Your Account#

The auto-upgrade preserves your old settings on paper, but AI Max behaves differently from DSA — and the accounts that benefit are the ones that configured before the switch.

The structural change is fundamental: DSA matched queries to your website’s page content. AI Max matches queries to a broader set of intent signals that go beyond your landing pages. Where DSA crawled your site and generated ads based on what it found, AI Max uses your content as one input among several — combining it with real-time intent data, existing ad creative, and audience signals to match queries you never explicitly targeted.

For WooCommerce stores, this expansion carries specific risks. AI Max can surface product pages for informational queries that will never convert. It can match category pages to queries with purchase intent but the wrong product context. The brand, location, and text controls that AI Max provides are powerful — but they only work if you configure them before the algorithm starts learning.

Google’s one-click upgrade tools port your DSA settings into standard ad groups with AI Max features enabled. Historical reporting and campaign learnings are preserved. But the default AI Max configuration enables all features simultaneously. If your conversion tracking isn’t clean before the upgrade, the algorithm inherits your data problems at the same moment it gains expanded reach.

The Conversion Signal Gap That AI Max Amplifies#

Automation amplifies whatever signal it sees — clean data produces better optimization, broken data produces worse optimization at higher velocity.

The question isn’t whether AI Max works. The question is what AI Max works with. Every algorithm-driven ad product — AI Max, Performance Max, Smart Bidding, Advantage+ — is an amplifier. Feed it clean, complete conversion data and it optimizes toward your actual customer profile. Feed it partial, biased conversion data and it optimizes toward a distortion, faster and with more confidence than manual campaigns ever could.

For WooCommerce stores, the conversion signal gap has specific, measurable components. Ad blockers remove the tracking script entirely for roughly 30% of visitors. Safari ITP caps first-party JavaScript cookies at seven days, fragmenting repeat-visitor attribution. Consent Mode failures silently suppress conversion events. Each layer removes a non-random slice of your customer data — the most privacy-conscious, technically sophisticated, and often highest-spending segment.

WooCommerce stores on client-side-only tracking lose 30–40% of purchase events before they reach Google Ads, starving AI Max of optimization signal.

When AI Max arrives in your account — whether you upgrade manually or wait for the forced migration — it begins learning from this incomplete signal immediately. Fixing your tracking after AI Max has already calibrated means the algorithm has to unlearn its initial model and recalibrate, extending the learning period and depressing performance during the transition.

Fix the Signal Before the Algorithm Arrives#

Server-side tracking closes the data gap that AI Max will amplify — and the fix needs to happen before the upgrade, not after.

The order of operations matters. If you fix your conversion tracking after AI Max has already started optimizing, you’re asking the algorithm to unlearn everything it calibrated during the initial learning period. That recalibration adds weeks of suboptimal spend on top of whatever data you were already losing.

Server-side tracking captures the conversion at the WooCommerce order hook — a PHP event that fires on the server, independent of ad blockers, Safari ITP, consent banner failures, and in-app browser quirks. The event travels from your server to Google Ads through an authenticated API call, bypassing every browser-side failure point. The Transmute Engine™ collapses the six intermediate hops between a WooCommerce checkout and the ad platform into a single controlled server hop.

The practical impact for AI Max is straightforward: instead of optimizing against 60–70% of your actual conversions, the algorithm sees 95–100%. The learning period shortens because the training data is complete. The bid decisions are calibrated to your real conversion rate, not a discounted version of it. And the audience model reflects your actual customer profile, including the high-value segment that browser-side tracking systematically misses.

You may be interested in: Your First-Party Cookies Still Die in 7 Days: Why Safari ITP Caps JavaScript Cookies and Only a True First-Party Server Survives

Key Takeaways#

• Check your timeline: DSA auto-upgrades moved to February 2027, but ACA and broad match settings still convert to AI Max in September 2026. Verify which legacy features your account uses.
• Don’t trust the headline number: Google’s 7% conversion improvement comes from non-retail advertisers with clean data. Independent testing shows 13% more revenue with 16% higher CPA.
• Audit conversion tracking before the upgrade: AI Max calibrates on whatever signal it receives during the initial learning period. Fixing tracking afterward forces a costly recalibration.
• Close the browser-side data gap first: If your WooCommerce store loses 30–40% of purchase events before they reach Google Ads, AI Max optimizes against an incomplete picture.
• Implement server-side tracking now: A server-side event pipeline captures conversions at the WooCommerce order hook, giving AI Max complete signal from the moment it starts learning.

References#

• Search Engine Land. (2026). Google delays Dynamic Search Ads migration to AI Max. https://searchengineland.com/google-delays-dynamic-search-ads-migration-to-ai-max-480049
• Google for Business. (2026). We’re upgrading Dynamic Search Ads to AI Max. https://business.google.com/us/accelerate/announcements/dsa-upgrade-to-ai-max-2026/
• Google Blog. (2026). Google’s Dynamic Search Ads are upgrading to AI Max. https://blog.google/products/ads-commerce/dsa-upgrade-to-ai-max-2026/
• Deeper Insights. (2026). Google Ads Replaces DSA with AI Max During Sep 2026. https://deeperinsights.com/ai-blog/google-ads-replaces-dsa-with-ai-max/
• SEOquick. (2026). Google AI Max in 2026: Should You Switch From Performance Max. https://seoquick.com.ua/en/google-ai-max/
• Coinis. (2026). Google AI Max Replaces Dynamic Search Ads in September. https://coinis.com/blog/google-ai-max-replaces-dsa-september-2026
• Search Engine Land. (2026). Why your brand campaign may not be ready for AI Max. https://searchengineland.com/brand-campaign-not-ready-ai-max-479663

If your WooCommerce store is heading into the AI Max transition with browser-only tracking, talk to Seresa about implementing server-side conversion tracking before the algorithm starts learning from incomplete data.