Google is retiring standalone Display campaigns. Migrate now or wait?

Google's migration tool for moving Display campaigns into Demand Gen has been rolling out since June, and the notices are now showing up in enough accounts that the question has become urgent for the wrong reason: it is September, Q4 is six weeks away, and the timing of this migration matters far more than most advertisers realise.

What Google has actually committed to, and when

Standalone Display is being folded into Demand Gen, the campaign type that already covers YouTube, Discover, Gmail and Maps. The Google Display Network itself is not going anywhere; what changes is the campaign type you buy it through. Here is the timeline as it currently stands.

The important structural point: there is no version of this where you keep standalone Display. The only variable you control is whether the migration happens on a date you chose, with a rebuilt and tested campaign, or on a date Google chose, with settings that were carried across automatically.

What the migration tool carries over, and what it does not

The tool is better than a manual rebuild in one specific respect and worse in several others. It ports campaign settings and up to 42 days of performance history, which matters because it shortens the relearning period rather than resetting it to zero. Migrated campaigns appear under the original name with a "#2" appended; the original is marked Removed and stays available for reporting under Google's standard retention.

What deserves manual verification is everything that made your Display campaign safe rather than fast.

The pattern in that middle block is worth naming: what tends not to survive is the negative configuration. Exclusions, brand safety controls and device restrictions are exactly the settings that accumulate over years, are rarely documented, and are invisible when they are missing. A campaign that migrates with its bidding intact and its exclusion list dropped does not look broken in the interface. It looks fine, and it spends into placements you had deliberately blocked.

Migrate now or wait? A decision framework

"It depends" is true and useless. What it depends on is a small number of factors, and they push in opposite directions, so score them rather than debate them. Give each row the points that describe your account, then total.

9 to 12 points: migrate now. You have low seasonal exposure, limited downside if performance wobbles, and the campaigns are not carrying much irreplaceable configuration. Migrating in September gives you a full quarter to learn Demand Gen's reporting and creative requirements before it is compulsory.

5 to 8 points: migrate one campaign now, the rest in January. Pick your smallest or worst-performing Display campaign, move it, and treat it as the pilot. You get the operational learning without exposing peak-season revenue, and January is still ten months clear of auto-migration.

0 to 4 points: audit now, migrate in January. Do not migrate a high-share, well-performing, heavily-excluded Display campaign in October. Do the export and documentation work this month anyway, because in January you will do the migration under time pressure with the same exclusions to reconstruct, and the only difference will be that you did not prepare.

Notice what "wait" does not mean in any of those three outcomes. It never means do nothing. The preparation work is identical either way, and it is the part that determines whether the migration is boring.

Where Demand Gen genuinely differs from Display

Treating this as a rename is the most common mistake, and it produces campaigns that underperform for reasons that look mysterious. Three real differences matter.

1. It is a creative-led campaign type. Display rewarded a responsive ad and a decent feed. Demand Gen sits across YouTube, Discover and Gmail, surfaces built for video and rich visual formats, and it allocates to the creative it can distribute. If your asset library is static banners, you are competing in an auction with the wrong ammunition.
2. Reporting gets better, not worse. Performance can be segmented by channel, and on YouTube by format. That is a real gain over Display's blended reporting, and it is worth using deliberately in the first month after migration to see where your GDN spend is actually going.
3. Lookalike segments are available. These are exclusive to Demand Gen, and for accounts with solid first-party lists they are the strongest argument for migrating sooner rather than later, particularly where signal loss has already degraded your remarketing pools.

Google's own figure is that advertisers adding GDN inventory to Demand Gen campaigns see an average 9.5% ROI increase. Read that the way you would read any platform's number about its own migration: it is directionally interesting, it is measured on Google's terms, and it is an average across a population whose composition is not disclosed. It is not a forecast for your account, and it is not a reason to migrate during your busiest trading weeks.

The pre-migration audit, in order

1. Export placement exclusions, app exclusion lists and content exclusion settings. Screenshot targeting, device and audience layering.
2. Record a 30-day baseline: spend, conversions, CPA or ROAS, impression share, and top placements. You cannot judge post-migration performance without a before.
3. Verify conversion tracking is firing correctly today. If it is not, fix that first, or the migration will get blamed for a measurement problem that predates it.
4. Audit creative against Demand Gen's formats. Identify what video and vertical assets you are missing and whether you can produce them before you migrate.
5. Confirm your budget and bid strategy, and decide in advance what you will do if CPA drifts during relearning. Deciding that under pressure produces panic edits, which extend the relearning period.
6. Migrate one campaign, not all of them. Verify every exclusion and setting on the new campaign against your export before increasing spend.

If Display is a meaningful slice of your budget, this is worth running alongside the other Google Ads migration in flight right now: AI Max on the Search side is auto-upgrading campaigns this month. Two simultaneous forced migrations in one account is a lot of moving parts to attribute performance changes to, which is another argument for staggering them deliberately rather than letting both land in the same fortnight.

How this fits your Q4 plan

The blunt version: do not run a structural campaign migration and your heaviest trading period through the same three weeks. Relearning after a migration takes a couple of weeks of stable data, and the whole point of Q4 planning is that those weeks are the ones you cannot afford to spend on an algorithm finding its feet. That is the same logic that governs a site-side change freeze, applied to the ad account.

The calendar that works for most advertisers: audit and export in September, pilot one campaign in September or early October, freeze structural changes from mid-October through the first week of December, then migrate the rest in January when you have clean data, no seasonal noise, and a full year before auto-migration. If your paid media is managed externally, the useful question to ask this week is simply which of those two windows they are planning for, and whether the exclusion export has already been done.

Frequently asked questions

Is Google shutting down the Display Network?

No. The Google Display Network itself continues, with inventory across more than 2 million sites, videos and apps. What is being retired is the standalone Display campaign type. From January 2027 you will not be able to create new Display campaigns, and GDN inventory is bought through Demand Gen campaigns instead.

When do I have to migrate my Display campaigns to Demand Gen?

The migration tool started rolling out to eligible advertisers in June 2026. Creation of new standalone Display campaigns stops in January 2027, and remaining eligible campaigns are auto-migrated later in 2027, on a date Google has not published. Existing campaigns keep running and stay editable until they are migrated.

Does migrating to Demand Gen reset my campaign learning?

Not entirely. The migration tool carries campaign settings across along with up to 42 days of performance history, which shortens the relearning period compared with building a campaign from scratch. Expect a couple of weeks of unstable performance regardless, which is the main reason not to migrate during peak trading weeks.

What might not transfer correctly during the migration?

The settings most at risk are the negative ones: placement exclusions, app exclusion lists, brand safety and content controls, device targeting and audience layering. These may not transfer cleanly, and a campaign missing its exclusions looks healthy in the interface while spending into inventory you had deliberately blocked. Export them before migrating and verify them afterwards.

Should I migrate before or after the 2026 holiday season?

For most advertisers with meaningful Q4 revenue and a significant Display budget, after. Migrate one small or underperforming campaign now as a pilot, freeze structural changes from mid-October, and move the rest in January. Advertisers with flat seasonality, small Display budgets or already-underperforming Display campaigns are better off migrating now and using Q4 to learn the campaign type.

What is different about Demand Gen compared with Display campaigns?

Demand Gen runs across YouTube, Discover, Gmail and Maps as well as GDN, and it is a creative-led campaign type that expects video and vertical assets rather than static banners alone. It offers reporting segmented by channel and, on YouTube, by format, and it gives access to Lookalike segments, which are not available in Display campaigns.

The takeaway

This migration is not optional and it is not urgent, which is an unusual and useful combination. You have from now until January to choose your own date, and until later in 2027 before Google chooses one for you. Spend September on the part that is genuinely time-sensitive: exporting exclusion lists, recording a performance baseline and checking whether your creative library can actually feed a video-first campaign type. Then pick your window. Migrating a well-performing Display campaign in the middle of Q4 to save a January afternoon is the one version of this that reliably costs money.