Google Ads Shift from Dynamic Search Ads to AI Max: What Marketers Should Check in 2026

Google Ads is moving Search campaign automation further beyond the traditional Dynamic Search Ads model, but the change is not an overnight shutdown of DSA. As of August 2026, Dynamic Search Ads can still run, while Google provides a direct upgrade path to AI Max and states that campaigns using DSA are scheduled to begin automatic upgrades from February 2027. That makes 2026 a preparation year for advertisers rather than a deadline to switch every campaign immediately. The practical task is to understand what AI Max changes, identify which DSA controls will carry over, and make sure conversion data, landing pages, brand rules and reporting are strong enough before more automation is introduced. A careful migration should preserve the useful coverage that DSA provided while giving marketers clearer oversight of how Google expands queries, selects pages and creates ad text.

Why the Move from DSA to AI Max Requires More Than a Simple Switch

AI Max is not a separate Google Ads campaign type. It is an optimisation layer for existing Search campaigns that combines several features that previously sat in different parts of the account. Search term matching can extend reach beyond the advertiser’s existing keywords by using broad-match and keywordless signals. Text customisation can create additional headlines and descriptions from the advertiser’s ads, landing pages, keywords and other assets. Final URL expansion can then direct a user to a different relevant page on the same domain when Google predicts that page is a better fit for the search. This wider combination is the main strategic difference from DSA: instead of relying primarily on website content to match searches and select landing pages, AI Max can coordinate query expansion, creative adaptation and landing-page selection inside the same Search campaign.

For existing DSA campaigns, Google’s upgrade process changes the campaign structure as well as the serving logic. Dynamic ad groups are converted into standard ad groups, and existing Dynamic Search Ads are transformed into responsive search ads. Google creates the minimum static assets needed for those responsive search ads through text customisation. Advertisers can also use Final URL expansion to retain the ability to match search intent with relevant pages across a site. The important point is that marketers should not treat the upgrade as a cosmetic change in the interface. The new setup replaces dynamic ad units with responsive search ads and changes the controls used to manage page selection, query matching and creative generation. That means a DSA campaign that performs well today still needs an audit before it is upgraded.

Google also uses a transition period after a DSA upgrade. Legacy Dynamic Search Ads can continue serving temporarily while newly created responsive search ads are generated and checked against advertising policies. Once the new ads are eligible and active, the old DSAs are paused and moved into a view-only state. This process reduces the risk of an immediate traffic break, but it does not remove the need for account-level preparation. Some legacy DSA targets are not supported in AI Max, and Google warns advertisers about them during the upgrade. Older URL rules based on conditions such as page title or page content can become read-only, which means they may remain visible but can no longer be added to or edited after the change. Marketers should therefore record how current DSA targeting works before converting it.

Check Conversion Measurement and Bidding Before You Expand Reach

AI Max depends heavily on conversion signals, so the first audit should be measurement rather than keywords. Google specifically notes that features such as search term matching, Final URL expansion and text customisation require conversion-based Smart Bidding to work effectively. A campaign with duplicated conversion actions, weak lead-quality signals or transactions that do not pass meaningful values can give automated bidding the wrong objective. Before migration, marketers should confirm which actions are marked as primary, whether revenue or lead values reflect business importance, and whether the same conversion is being counted through more than one source. For lead-generation accounts, it is also sensible to check whether the campaign optimises for a meaningful lead event rather than an easy but low-value action such as a page view or form start.

A useful 2026 baseline should include more than total conversions. Record cost, conversion rate, cost per acquisition or return on ad spend, conversion value, search terms, major landing pages and the share of traffic that DSA currently contributes. For campaigns with a large catalogue, note which categories or page groups produce profitable demand and which ones mainly generate low-value visits. This creates a reference point for the period after AI Max is enabled. Without a baseline, an increase in conversions can look positive even when average order value falls, poor-quality leads rise or more budget is shifted to searches the business would not have chosen manually. The purpose of the baseline is not to preserve every DSA pattern, but to make sure changes in volume can be judged against commercial quality.

Google reported when AI Max was introduced in 2025 that advertisers activating it typically saw 14% more conversions or conversion value at a similar CPA or ROAS, while campaigns relying mostly on exact and phrase match keywords saw a larger typical uplift of 27%. Those figures came from Google’s own internal data and should be treated as a directional benchmark, not a forecast for an individual account. A retailer with thousands of strong product pages, a local service business with a small site and a regulated advertiser with strict wording requirements will not respond in the same way. In 2026, the safer decision is to judge AI Max against each campaign’s own economics, conversion quality and search demand rather than assume that an average uplift reported by Google will automatically apply.

Audit Landing Pages, Ad Assets and Controls Before Enabling AI Max

Landing-page quality becomes more important when Final URL expansion is active because Google can select a page that differs from the final URL originally entered in the responsive search ad. The feature is enabled by default when an advertiser opts into AI Max, although it can be turned off. Google states that expanded URLs should be relevant to the query and themed to the ad group, but the advertiser remains responsible for the pages available for serving. Before enabling it, check whether the site contains outdated offers, expired product pages, internal search results, recruitment pages, legal information, account areas or support content that should never receive paid Search traffic. A clean site structure gives automation better material to work with; a disorganised site can turn broader reach into irrelevant clicks.

This is especially important for businesses that previously used DSA exclusions to block unavailable inventory. Google still recommends preventing ads from sending users to pages containing products that are sold out or unavailable. The same principle should be applied more broadly during an AI Max audit. Product status, regional availability, pricing, delivery restrictions and service eligibility should be clear on the page itself. Marketers should also inspect page titles and visible copy because AI Max can use landing-page information to understand relevance and to support text customisation. A page written mainly for internal navigation or SEO fragments may technically be indexable, but that does not make it a good paid-search destination. The page should answer the user’s likely intent and make the conversion route obvious.

URL controls need particular attention during migration because AI Max handles them differently from some legacy DSA rules. Campaign-level URL exclusions can stop Final URL expansion from using specific pages, while URL inclusions can guide eligible pages at ad-group level. If URL inclusions are used while Final URL expansion remains on, Google may still use other relevant URLs that it predicts will perform well. If an advertiser needs traffic to stay only within a defined set of pages, that configuration should be reviewed carefully rather than assuming an inclusion is an absolute boundary. Existing DSA rules based on page title or page content also deserve documentation before upgrade because unsupported legacy rules can become read-only afterwards. For large sites, mapping current DSA targets to the future URL-control approach should be completed before campaign conversion.

Review Creative, Brand and Landing-Page Safeguards

Text customisation is another area where marketers should decide how much flexibility is acceptable. The feature can create additional headlines and descriptions using the domain, landing pages, existing ads, keywords and other assets, with generative AI used as part of the process. Google identifies these generated assets as “Google AI” in reporting, and individual customised assets can be removed if they do not suit the business. Before activation, review the existing responsive search ad copy and landing-page wording for statements that should not be reused automatically. Claims about prices, guarantees, delivery, professional qualifications, medical outcomes, financial results or limited-time offers require particular care because a sentence that is correct on one page may be misleading when reused in another context.

Advertisers that depend on pinned responsive search ad assets should also check Final URL expansion before migrating. Google states that pinned RSA assets are not respected when Final URL expansion chooses a more relevant URL, because the pinned copy may not fit the newly selected page. This matters for brands that use mandatory disclaimers, tightly controlled legal wording or fixed promotional messages. If those pinned assets must always appear, Final URL expansion should be disabled. Where available, text guidelines can add another layer of control by defining restricted terms or messaging rules for AI-created text, but they should support human review rather than replace it. The practical standard is simple: if a sentence would require approval before a copywriter published it, automatically generated variations should be reviewed with the same care.

Brand controls can reduce another common concern around expanded search matching. AI Max supports brand exclusions at campaign level and brand inclusions at campaign or ad-group level. These controls are useful when an advertiser needs to avoid competitor-brand traffic, separate branded and non-branded demand, or restrict an ad group to searches associated with a defined brand set. Negative keywords remain relevant as well, especially for clearly unsuitable searches, but they should be based on evidence rather than added aggressively after a few early clicks. The objective is to give AI Max enough room to find incremental demand without letting it ignore commercial boundaries. Before launch, marketers should therefore review brand lists, negative keyword lists, URL exclusions and any location-specific rules together, because each one shapes a different part of the traffic AI Max can reach.

Measure the Transition Without Losing Useful DSA Benchmarks

AI Max reporting in 2026 gives marketers more detail than a simple campaign total. The Search terms report can show searches and landing pages associated with AI Max, along with the headlines that were used. The Keywords report includes aggregate rows for AI Max expanded matches and AI Max landing-page matches, helping distinguish traffic added through broader matching from traffic found through pages and assets outside the advertiser’s keyword list. Asset reports identify text created through customisation, while the Landing pages report shows advertiser-provided URLs and pages selected through Final URL expansion. These views should become part of routine Search campaign review because they answer three practical questions: what did the person search, what message did Google show, and where did the click land?

Google recommends allowing at least two weeks after AI Max is enabled before making changes such as adding negative keywords, giving the system time to learn and optimise. That guidance is useful, but it should not be interpreted as a reason to ignore obvious problems. Policy issues, broken tracking, irrelevant high-spend queries or unsuitable landing pages still require prompt action. For normal optimisation, however, reacting to a handful of searches on the first day can cut off new demand before enough conversion data exists to judge it. A better review rhythm is to monitor basic health immediately, then assess search-term quality, landing-page performance, conversion value and cost efficiency over a meaningful period. The exact period should reflect the campaign’s traffic volume and sales cycle rather than a fixed number of days alone.

Historical DSA reporting should remain part of the comparison during 2026. Google’s Search terms reporting provides views for AI Max traffic as well as Dynamic Search Ads, which makes it possible to compare the older model with the new one while both data sets are still relevant. Marketers should keep a dated record of major settings before the upgrade: DSA targets, exclusions, page feeds, bidding strategy, conversion goals, budget, core search terms and top landing pages. After migration, compare not only cost and conversions but also how the traffic mix changes. If AI Max finds more conversions but shifts users towards lower-margin categories, weaker leads or pages with poor retention, the campaign may need tighter URL, brand or query controls even if headline performance metrics look acceptable.

Build a 2026 Migration Process Before Automatic Upgrades Begin

A sensible migration order starts with campaigns where measurement is reliable and the website has enough high-quality landing pages to support broader automation. Large product or service catalogues with accurate availability, clear category structures and stable conversion tracking are natural candidates for earlier testing. Campaigns with strict legal copy, narrow geographic eligibility, unusual tracking requirements or a small number of approved landing pages may need more preparation. This does not mean those campaigns cannot use AI Max; it means the cost of an unsuitable query, message or page can be higher. Prioritising by readiness gives the marketing team time to learn how the controls behave before the same approach is applied to the most sensitive parts of the account.

Before pressing the upgrade button, document any DSA rules that cannot be recreated in the same form, especially legacy conditions based on page title or page content. Export or record the pages those rules currently include, the exclusions that protect budget, and the search terms that repeatedly convert. Then check tracking templates and URL parameters on the new responsive search ads and expanded landing pages so analytics continues to identify paid traffic correctly. After the upgrade, confirm that the new RSAs are eligible, that the expected pages can serve, and that legacy DSAs have moved into their intended view-only state once the transition completes. This practical record is valuable because it separates migration effects from unrelated changes in seasonality, budget or website content.

Timing should guide the 2026 plan. Current Google Ads help documentation says campaigns using Dynamic Search Ads are scheduled to begin automatic upgrades to AI Max from February 2027, so advertisers still have time to test and prepare before that process starts. A separate change begins earlier: Google states that campaigns using text customisation, formerly called automatically created assets, will automatically upgrade to AI Max starting in September 2026. Marketers should therefore audit Search campaigns now rather than wait for DSA automation in 2027. The priority is not to preserve an older setup for as long as possible, nor to enable every AI option immediately. It is to enter the transition with accurate conversion signals, suitable pages, reviewed creative controls and reporting that makes new automation accountable to business results.