Dynamic Search Ads to AI Max: 2026 Migration Checklist for Indian Advertisers

Google is consolidating more Search automation under AI Max, but the timeline has an important split. Google's 11 June update says the Dynamic Search Ads sunset and automatic DSA upgrade will begin in February 2027. Automatically created assets and campaign-level broad match remain scheduled to upgrade from September 2026. Read the updated Google announcement.

That gives advertisers more time for DSA, but it does not justify waiting. The useful work is not flipping a single toggle. It is documenting what the current campaign does, deciding which AI Max controls are acceptable and testing whether the new matching, assets and landing-page selection improve business outcomes.

What is actually changing in the DSA to AI Max migration?

Google describes AI Max as an optimization layer for existing Search campaigns, not a new campaign type. It combines search term matching with asset optimization. Asset optimization includes text customization and Final URL expansion.

Google says the upgrade process ports historical settings and data. It also warns that unsupported legacy DSA target types can become read-only. That is why the pre-migration inventory matters.

Which campaigns should an Indian advertiser audit first?

Start with Search campaigns that meet any of these conditions:

• use Dynamic Search Ads as a catch-all for product or service inventory;
• have automatically created assets enabled;
• use the campaign-level broad match setting;
• rely on page feeds, URL rules or excluded site sections;
• use account or campaign tracking templates;
• pin RSA assets for regulatory, offer or brand-language reasons;
• send traffic across multilingual, marketplace, support or policy pages; or
• optimize to a conversion action that sales teams do not consider qualified.

The last point is easy to miss. Better matching cannot rescue a weak conversion signal. If every WhatsApp start, low-value lead or duplicated purchase is treated as equal, automation learns from the wrong outcome.

Use the AI bidding guide for Indian D2C brands to review the broader bidding and signal setup. This checklist focuses specifically on the Search migration layer.

What should be captured before changing the campaign?

Create a dated baseline for each affected campaign.

Google's current documentation says tracking templates should be tested with AI Max dynamic landing pages. Static URLs or nonstandard `{lpurl}` use can interfere with Final URL expansion and may cause landing-page errors. Review Google's tracking guidance.

How should Final URL expansion be controlled?

Final URL expansion can send a user to a different relevant URL on the advertiser's domain. That can improve relevance, but only if the site architecture is safe for paid traffic.

Before enabling it:

1. Exclude login, account, cart, policy, support, careers and internal-search pages unless they are deliberate destinations.
2. Exclude out-of-stock, discontinued and low-margin product groups where the business case is weak.
3. Test URL parameters, redirects, consent banners and analytics on representative dynamic destinations.
4. Review whether generated text can accurately describe every eligible landing page.
5. Document RSA pins that must always serve. Google says pinned assets may not be respected when Final URL expansion selects another page.
6. Confirm that conversion values and product margins are available to the bidding system where appropriate.

For a quick economic check, calculate the campaign's current threshold with HML's Break-Even ROAS Calculator. A migration should not be declared successful because it produces more low-margin revenue below that threshold.

How should an AI Max experiment be designed?

Google provides an AI Max experiment that divides traffic and budget inside an existing campaign. The control has AI Max off and the trial has it on. Google says this approach can reduce copy-related setup errors and shorten the learning period compared with a traditional copied-campaign test. See Google's experiment documentation.

Before starting, write down:

• the business outcome being optimized;
• the minimum useful test duration and conversion volume;
• the allowed budget movement;
• the primary metric and diagnostic metrics;
• query, landing-page and asset review cadence;
• conditions that trigger a pause; and
• the rule for applying, extending or rejecting the experiment.

Google lists limitations for the streamlined experiment flow, including campaigns using shared budgets, portfolio bidding, active experiments or certain legacy settings. Check eligibility in the live account rather than assuming every campaign can use the same test type.

What would a four-week migration test look like?

Consider a hypothetical Indian home and lifestyle brand with one Search campaign spending ₹4,50,000 per month. This is a synthetic operating example, not client data.

The campaign currently uses DSA, tROAS bidding, a product page feed and 18 URL exclusions. The team imports purchase value but does not pass contribution margin.

Week 0: establish the baseline

• Export the previous eight weeks of search terms, landing pages, assets and conversion value.
• Classify search terms as relevant, adjacent or irrelevant using a documented rule.
• Identify pages with stock, margin or claim constraints.
• Validate tracking templates against a sample of dynamic URLs.
• Record current break-even ROAS and qualified revenue.

Weeks 1 to 4: run and review

The trial should be assessed on:

• qualified conversion value, not only all conversion value;
• contribution or margin proxy where available;
• irrelevant-query spend as a share of trial spend;
• selected landing pages that fail the destination checklist;
• generated assets needing correction or removal;
• brand leakage and location mismatch; and
• spend stability when the campaign is limited by budget.

The team could predefine a decision such as: apply only if qualified conversion value improves without breaching the irrelevant-query, destination-quality or contribution guardrails. The exact thresholds must come from the business, not from this example.

If the team needs a causal read beyond platform attribution, use the incrementality testing guide and Incrementality Lift Calculator. Not every account has enough volume for a clean lift test, so state that limitation before interpreting the result.

What should be monitored after the upgrade?

AI Max adds reporting fields that make the automation more inspectable. Google documents an AI Max match type, a source column for broad or keywordless matching, a combined search-term, headline and URL view, a Selected by field in landing-page reporting, and performance metrics for optimized assets.

Use those reports as an operating queue:

• review new search terms for relevance and commercial intent;
• inspect landing pages selected by AI Max;
• remove or constrain generated assets that fail brand or claim review;
• add brand and URL controls where recurring leakage appears;
• separate incremental reach from cannibalized brand demand; and
• compare platform results with blended business metrics.

The Performance Max glossary entry explains the difference between multi-channel PMax and Search-specific automation. AI Max does not replace PMax; it changes how existing Search campaigns can match, customize and select destinations.

What should the migration sign-off include?

The owner should be able to answer seven questions:

1. Which legacy settings are changing, and on what account-specific timeline?
2. Which URL rules or DSA targets will become read-only?
3. Which pages can and cannot receive paid traffic?
4. Which brand, location, text and URL controls are active?
5. Which conversion values represent commercially useful outcomes?
6. Which reports will be reviewed weekly, and by whom?
7. What evidence will cause the team to apply, modify or reverse the change?

The bespoke tailoring performance case study shows why channel decisions need to connect with wider sales and messaging operations. Its results are not evidence for AI Max performance or for the hypothetical test above.

If your Search account uses DSA, automatically created assets or campaign-level broad match and nobody owns the transition checklist, request an AI Max migration audit. The useful deliverable is a campaign inventory, control map, tracking test and experiment brief, not a promise that more automation will automatically improve profit.

Sources

1. We're upgrading Dynamic Search Ads to AI Max - Google Ads; published 15 April 2026, updated 11 June 2026; accessed 21 August 2026.
2. Set up AI Max for Search campaigns - Google Ads Help; accessed 21 August 2026.
3. How AI Max for Search campaigns works - Google Ads Help; accessed 21 August 2026.
4. About AI Max experiments - Google Ads Help; accessed 21 August 2026.