dynamic search ads ai max migration: a step-by-step for indian accounts

dsa campaigns move to ai max in february 2027 whether you act or not. here is what google's upgrade tool changes, how each dsa control maps across, how to test it first, and the checklist we would run on an indian lead-gen account.

our guide to google search ads in 2026 gives the short migration checklist, and the pillar on ai max for search campaigns in india covers which settings to keep, change or switch off. this post goes one level deeper on the dynamic search ads ai max move: what the upgrade does to each dsa control, how to test it, and what to watch week by week.

• the automatic upgrade starts in february 2027, per google's dynamic search ads help page, after a june 2026 delay (google ads developer blog).
• dynamic ad groups become standard ad groups, and dsa ads become responsive search ads (google ads help).
• "page title contains" and "page content contains" rules become read-only. you can remove them, not edit them.
• you can test first. google documents two ways to run an ai max experiment against a dsa campaign.

dynamic search ads ai max: what has google announced, and when?

google announced on 15 april 2026 that dynamic search ads would move into ai max. in june 2026 it delayed the automatic dsa upgrade to february 2027 and restored new dsa creation for now. new dsa creation is expected to close around january 2027. until then, upgrading is voluntary and on your timeline.

for indian advertisers, december and january, after the festive peak, are the clean window to migrate on your own terms.

dsa upgrade to ai max: what does the tool change, and what carries over?

the dsa upgrade to ai max converts dynamic ad groups to standard ad groups, turns dsa ads into responsive search ads and switches on all three ai max features. google says settings and data come across and legacy url controls are kept. older rule types go read-only, and unsupported targets trigger a warning in the upgrade dialog.

google's setup help says the upgrade starts from a callout in the dsa settings panel, and the old ads keep serving while the new ones are policy-reviewed. the campaign and its history stay put.

what we could and couldn't confirm in google's own pages, as of 11 november 2026:

if your dsa leaned on page-content rules or negative targets to keep traffic off thin pages, rebuild that protection before you click upgrade.

how do dsa controls map to ai max features?

every dsa control has an ai max equivalent, but the defaults differ. dynamic targeting becomes search term matching, dynamic headlines become text customisation, and google's page choice becomes final url expansion. page targets become url inclusions per ad group, negative targets become url exclusions per campaign, and brand and location get new controls.

levels per google's setup help, api docs and brand settings help. locations of interest read geographic intent in the query (google).

ai max search term matching vs dsa matching

ai max search term matching reads keywords, ads and pages together, where dsa read only the page. a keywordless dsa migrates into keywordless ad groups, so add five to ten keywords per offer from your best dsa search terms. reporting separates ai max expanded matches from landing page matches (ai max reporting help).

ai max final url expansion vs dsa landing pages

ai max final url expansion picks the page it thinks fits, as dsa did, but now across keyword traffic too. one catch, reported by intercept: when expansion picks a different url, pinned assets are not used. if you pin a rera number or fee disclaimer, keep that page in an inclusion. our landing page design guide covers pages google can read clearly.

google ads url exclusions: how do you rebuild page targeting before upgrading?

google ads url exclusions in ai max sit at campaign level and stop final url expansion using a page. url inclusions sit at ad group level and say which pages each ad group may use. before upgrading, list every page that earned a lead, make each offer's pages an inclusion, and exclude the rest.

beyond the parent guide's blog, careers, login and policy pages, indian sites usually need more exclusions, because cms templates leave thin pages indexed.

• pull 90 days of the dsa landing page report, sorted by cost, and exclude every url that spent without converting.
• find pages in a language you don't sell in. with language targeting gone, a hindi blog post can pull hindi queries into an english funnel.
• map each converting page to one offer, with no overlap between ad groups.

google's setup help notes that url inclusions and exclusions are ignored if ai max is switched off.

ai max experiment: how do you test the upgrade before committing?

run an ai max experiment on the dsa campaign before upgrading it. google documents two routes: a custom experiment, where you select "upgrade campaign" in the draft's dsa settings, or the experiment template. split mixed campaigns first, because in a mixed campaign only the standard ad groups get ai max in the test.

the routes, per google's dsa help page as summarised by search engine roundtable:

since august 2026, ai max experiments can run with brand and location controls on, as reported by search engine land. without them, cheap brand searches can make the treatment look like a win.

small indian lead-gen budgets rarely give a confident answer in two weeks, so we plan four. one real estate developer we work with averages 18 days time-to-lead, so a short test can judge traffic before its leads have arrived.

google's quoted 7% conversion uplift compares all three ai max features with search term matching alone, not with dsa, as reported by search engine journal.

what does the step-by-step migration checklist look like?

migrate in four stages: audit, prepare, test, then switch. export dsa search terms and pages, fix conversion goals, rebuild inclusions, exclusions and brand controls, then run an experiment for four weeks. upgrade only when the treatment wins on cost per qualified lead, and review search terms weekly for the first month.

stage one: audit and export

• export 90 days of search terms, landing pages and dynamic ad targets. legacy targets go read-only after the upgrade.
• tag each search term as buyer, researcher, job seeker, wrong city or competitor.
• turn the best buyer terms into keywords for the new standard ad groups.

stage two: conversion data and controls

• make a qualified lead the primary conversion, imported from the crm; see our google ads conversion tracking guide and the cross-platform conversion tracking guide.
• use conversion-based bidding, with a value per offer if bidding on value. google says manual cpc limits ai max.
• add brand exclusions for your own brand on non-brand campaigns, and for competitors.
• set locations of interest where an offer is city-specific, and review text customisation for fee or regulatory claims.

stage three: test

• split mixed campaigns so dsa ad groups sit alone, then launch the experiment with brand and location controls on, and leave the control arm untouched.

stage four: switch and watch

google advises waiting two weeks after enabling ai max before adding negatives (ai max reporting help). we follow that for marginal queries but block job seekers and wrong cities immediately.

migrate now, wait, or keep keyword-only? a decision table

migrate now if the dsa campaign has steady conversions, a clean site and crm-fed goals. wait until december or january if you're mid-season or tracking is broken. keep keyword-only, with ai max features off after the upgrade, if your site is thin, your claims are regulated, or one wrong landing page costs real money.

campaign types by indian business are in google ads for business in india. if performance max runs alongside, see our performance max guide, as both chase broad queries.

illustrative example: migrating an education account

in this illustrative example, a pune coaching institute spends ₹1.2 lakh a month on a whole-site dsa campaign. migration moves it to three course ad groups with keywords, url inclusions per course, exclusions for the blog and old batches, and a four-week experiment judged on counselling sessions booked.

illustrative example, made-up numbers, not a client. the 90-day audit finds about ₹3.6 lakh spent: roughly ₹1 lakh on blog posts about exam dates, ₹40,000 on careers and old batch pages, and the rest on course pages, which produced nearly every enquiry.

if the treatment books sessions at, say, ₹1,900 against ₹2,300 for the control, the institute upgrades. if not, it keeps the new keywords and exclusions and retests in january.

what carries over from real accounts is tracking discipline. for lodha developers, 45 flats were sold with sales 100% attributed to digital, an attribution you can only report when every lead is traced to its source. our marut drones campaigns ran on meta and google across 13 states with state-specific creative and forms, producing 23,420 leads from ₹25 lakh.

mistakes we see in dsa to ai max migrations

the common mistakes are upgrading without an experiment, carrying a whole-site page feed across, forgetting that page-content rules go read-only, judging on form fills, and panic-adding negatives in week one. each either lets final url expansion send traffic to the wrong page or hides whether the upgrade worked.

• testing a mixed campaign. only its standard ad groups get ai max in the experiment.
• planning to edit "page content contains" rules later. they are read-only after the upgrade.
• leaving the blog in the page feed. final url expansion will use it.
• measuring on form fills. ai max finds more of whatever you count.
• waiting for february. an automatic upgrade gives you no baseline and no say.

who wrote this, how and why

who: puneetraj chavan, founder of rapchik media, with 8+ years running performance marketing and automation for 30+ brands across india and the us.

how: sources were checked on 11 november 2026. google's pages were read through search summaries where they could not be opened directly, and trade press is attributed. platform-reported figures were not independently audited. the education account is an illustrative example.

why: most dsa migration guides stop at "click upgrade", and indian accounts with thin pages, mixed languages and form-fill tracking need the steps in between.

• google: dsa upgrading to ai max · ads developer blog: dsa automigration delayed · ai max launch · google ads api: ai max
• google ads help: about dynamic search ads · set up ai max · ai max faq · ai max reporting · brand settings
• trade press: ppc land · search engine land: dsa delay · search engine land: ai max testing tools · search engine roundtable · search engine journal

want your dsa campaigns migrated with an experiment instead of a february surprise? see our performance marketing services or talk to us.

found this useful? share it

written by puneetraj chavan

founder of rapchik media. 8+ years scaling 30+ brands across india and the us, from agri-drones to a us job portal. strategy and execution, same brain.

more about me →

want this on your brand?talk to puneetraj on whatsapp reply within 24h →