Google Display to Demand Gen Migration: A Practical Runbook

If your standalone Google Display campaigns depend on Manual CPC, portfolio bidding, HTML5 ads or tightly constrained audience targeting, the move to Demand Gen is not a simple campaign rename. Some of those controls will not survive the migration, and pretending they will can leave you diagnosing a strategy change as if it were a temporary performance dip.

Your safest approach is to separate migration from expansion. First reproduce the campaign as closely as Demand Gen allows. Confirm that bidding, creative, audiences, conversion signals and spending behave as expected. Only then test Discover, Maps, additional formats or other Demand Gen capabilities.

The deadline matters, but the unspecified date matters more

The ability to create new standalone Display campaigns will be disabled permanently in January 2027. Google expects automatic migrations later in 2027, but it has not specified when those migrations will begin.

That gap is a planning risk. Waiting for automatic migration means you may lose control over whether the change lands near a product launch, seasonal promotion, reporting deadline or peak sales period. Migrating deliberately gives you a chance to resolve incompatibilities and choose a period when your team can watch the account closely.

The migration tool began appearing through a phased rollout in June after the May announcement, so availability can differ between accounts. When the tool is available, it can update a live campaign while preserving the previous 42 days of performance history. That continuity is intended to reduce the new campaign’s learning period, not eliminate performance volatility.

You also do not have to adopt every Demand Gen network during the move. Demand Gen supports individual network selection, and the migration tool initially creates a Google Display Network-only Demand Gen campaign. Other networks can be added afterward. Use that separation: migration should answer whether the replacement works; expansion should answer whether new inventory adds value.

Audit the features that will not transfer cleanly

Build a compatibility sheet before touching the migration control. For every Display campaign, record its current bid strategy, targeting logic, creative formats, data dependencies, measurement studies and business objective. Then mark every feature that requires a replacement.

The supported bidding options are Target CPA, Target ROAS and Maximize Clicks. That does not make them interchangeable. Maximize Clicks is aligned with traffic acquisition. Target CPA requires a meaningful conversion action and a cost-per-acquisition objective. Target ROAS makes sense only when the conversion values being supplied reflect the value you actually want the system to pursue.

Do not choose the nearest-looking setting just to clear the migration screen. Write down what the old strategy controlled and what the replacement will optimize. A campaign moving from Manual CPC to automated conversion bidding, for example, is undergoing a buying-strategy change as well as a platform change. Its before-and-after results need to be interpreted accordingly.

Creative needs the same discipline. Business Data feeds, uploaded HTML5 ads and third-party ads are not currently supported. HTML5 and third-party ad support is planned for late 2026, while no precise availability date is stated for Business Data feeds. If one of those formats is essential to the campaign, the missing replacement is a migration blocker, not a minor setup detail.

Use a controlled six-step migration runbook

A controlled migration gives you a clean baseline and limits the number of explanations for any change in performance. Use the same runbook for each campaign so that results are comparable across the account.

1. Capture the baseline. Export at least the same 42-day performance window the migration tool can retain. Record the campaign budget, bidding method and target, conversion actions, audiences, creative formats and any account-level process that depends on the campaign. Save the date and time of the export.
2. Map every unsupported dependency. Use the compatibility table above. Assign a replacement, owner and decision date to each unsupported bid strategy, adjustment, feed, creative format or lift study. Do not begin while an essential dependency remains unresolved.
3. Choose the cutover window. Performance may fluctuate after migration, so avoid placing the first move immediately before a peak trading period. Pick a time when someone can inspect delivery and spending during the rest of that day and over the following reporting periods.
4. Account for same-day spending. The newly migrated campaign will not respect the amount the old campaign already spent earlier that day. That creates a risk of more combined same-day spend than you expected. Watch total account spend at the cutover rather than assuming the new campaign knows how much daily budget has already been consumed.
5. Verify the migrated campaign before expanding it. Confirm that the new Demand Gen campaign is restricted to Google Display Network inventory, as the migration tool is designed to configure it. Check the budget, bid strategy, conversion inputs, audience setup and live creative against your migration sheet.
6. Hold expansion until the replacement is interpretable. Let the Display-only campaign establish a usable post-migration pattern before adding Discover, Maps or materially different creative. When you do expand, change one major dimension at a time and record the date.

The fourth step deserves special attention. If a Display campaign has spent a large portion of its daily budget before migration, the new campaign can still begin without credit for that earlier spending. You do not need to predict the exact effect. You do need an operator watching the combined spend and authorized to respond if delivery departs from the day’s plan.

During the first post-migration review, compare more than totals. Look at spend, delivery, conversion volume, conversion value and the metric tied to the chosen bid strategy. Annotate the migration so nobody later mistakes the cutover for an unexplained market movement. If you change the bid target, network mix and creative simultaneously, you will have no reliable way to tell which change produced the result.

Rebuild audience and measurement assumptions for Demand Gen

The deepest change is not in the interface. Demand Gen uses audiences and conversion signals as inputs to a more automated buying system, so controls that once acted as boundaries may now act as guidance.

Treat Lookalike Audiences as signals, not fences

As of March 2026, Lookalike Audiences changed from restrictive targeting to audience suggestions. Your first-party seed still informs the model, but Google can reach high-intent users outside the similarity thresholds associated with the seed.

If you previously relied on a Lookalike segment to define who could receive an ad, revise that assumption before migration. The segment now supplies direction rather than a hard perimeter. Judge it by the outcomes it helps generate, and use the campaign’s available network controls to govern where ads can run.

Create required Lookalike segments well before the campaign needs them. A new segment can take up to 96 hours to populate. Building it on the day of migration can leave you troubleshooting a campaign whose intended audience signal is not yet ready.

Decide whether view-through conversions belong in optimization

Demand Gen can optimize bids toward users who view a YouTube or Discover ad, do not click it, and convert later. This view-through-conversion optimization can provide an additional signal when click-led conversion volume is too limited for conversion-focused bidding.

That capability is not automatically an improvement for every account. Decide whether a later conversion following an ad view represents an outcome you want the bidding system to pursue. Make the decision explicitly, document it and avoid comparing a view-through-optimized campaign with a click-focused predecessor as though their optimization inputs were identical.

Billing also changes in a specific configuration. Google has announced that Demand Gen campaigns running on Discover with view-through-conversion optimization will move from CPC to CPM billing. In other words, cost is tied to impressions rather than clicks. This does not affect a Demand Gen campaign serving only on the Display network, but it becomes relevant as soon as you add Discover under that optimization setup.

Before enabling that combination, update the reporting brief. Stakeholders should know that a change in click volume or CPC no longer describes the entire buying model. Mark the billing transition date and evaluate delivery, cost and conversions within the new setup rather than forcing a direct CPC-era comparison.

Add new reach as a test, not a migration default

Demand Gen can extend reach into Discover and Maps, offers formats unavailable in standalone Display, and includes additional generative image tools. It also provides more ways to use YouTube creative. An Affiliate Partnership boost for organic YouTube affiliate videos is in pilot, so it should be treated as an emerging option rather than a dependency in your migration plan.

Test those capabilities after the Display replacement is stable. State one hypothesis for each expansion: the audience you expect to reach, the conversion signal the campaign will optimize and the result that would justify continued spend. Add one network or major creative approach at a time. Otherwise, Demand Gen’s wider inventory can obscure whether the migration itself succeeded.

Key takeaways

• New standalone Display campaign creation ends in January 2027; automatic migration is expected later in 2027, but its start date remains unspecified.
• The migration tool can retain 42 days of history and initially creates a Display Network-only Demand Gen campaign.
• Manual CPC, Viewable Impressions, Pay for Conversions, bid adjustments, seasonality adjustments and portfolio bidding require alternatives.
• Business Data feeds, uploaded HTML5 ads, third-party ads, Brand Lift studies and Search Lift studies have support gaps that may block some migrations.
• Same-day spend from the old campaign is not credited against the newly migrated campaign’s budget, so the cutover needs active spending oversight.
• Lookalike Audiences are suggestions rather than restrictive targeting, and a new segment can take up to 96 hours to populate.
• Discover inventory using view-through-conversion optimization is moving from CPC to CPM billing; Display-only delivery is not affected by that billing change.

Start by opening your highest-spend standalone Display campaign and marking every compatibility-table row as supported, replaceable or blocked. If a blocked item has no approved alternative, solve that first. If the campaign is clear, choose a monitored migration window, preserve the Display-only network setup and put any Discover, Maps or creative expansion into a separate test plan.

References

• Search Engine Land — Moving from Google Display to Demand Gen: What you’ll lose and gain