Google Ads and DV360 Updates: A Practical Action Plan

If you run ecommerce media or maintain advertising integrations, two Google changes demand different responses. Smart Bidding Exploration creates a performance decision inside Google Ads. Display & Video 360 API v5 creates a migration decision for your engineering stack.

Do not bundle them into a general “Google update” project. Your media team needs a controlled rollout with financial guardrails. Your developers need a staged beta evaluation that protects production workflows.

Key takeaways

• Smart Bidding Exploration now supports Search, Performance Max, and Shopping campaigns, but it does not create a new targeting method.
• The feature requires Target ROAS bidding and is intended for campaigns that are not limited by budget.
• Performance Max exploration applies to additional search opportunities within the campaign’s existing eligible search universe. Shopping can explore across Search and Shopping traffic when a Merchant Center feed is available.
• More bidding flexibility does not guarantee more conversions or a better actual ROAS. Evaluate the change against your own commercial limits.
• Display & Video 360 API v5 is a public beta, so teams should expect possible breaking changes before a stable release.
• Customer Match management is being removed from the DV360 API in v5. Those workflows must move to Google’s Data Manager API.

Use four gates before enabling Smart Bidding Exploration

Smart Bidding Exploration gives Google’s automated bidding system more freedom to pursue eligible searches that a standard Target ROAS strategy might not prioritize. That can be useful when a profitable campaign has room to grow. It is not a cure for an underfunded campaign, weak conversion tracking, or an unrealistic efficiency target.

Run these checks in order. A failure at any gate is a reason to stop and resolve the underlying configuration before you consider a rollout.

1. Bidding strategy: Confirm that the campaign uses Target ROAS. The feature is not available under another bidding objective.
2. Budget condition: Confirm that the campaign is not limited by budget. If the campaign already lacks enough budget to enter the auctions its current strategy wants, adding exploration will not isolate the value of broader bidding discretion.
3. Bid-management structure: Check whether a portfolio bid strategy controls the campaign. Portfolio strategies are compatible with Smart Bidding Exploration only for Search campaigns, not for Performance Max or Shopping.
4. Management platform: Check whether Search Ads 360 sits in the workflow. Smart Bidding Exploration is not available in Search Ads 360 for Shopping or retail Performance Max campaigns.

Know what the setting actually expands

For Performance Max, exploration affects search-related bidding opportunities. It looks for additional searches among terms for which the campaign is already eligible. It does not give you a new keyword-targeting layer or an unrestricted path into unrelated queries.

For Shopping, exploration can operate across Search and Shopping traffic when the campaign has a Merchant Center product feed. That makes feed quality and conversion value especially important: the bidding system can only optimize against the products, eligibility, and conversion signals available to it.

This distinction prevents a common planning error. If you need to change what the campaign is eligible to advertise, work on targeting inputs, product data, or campaign structure. If you want Target ROAS bidding to consider more opportunities inside the eligible universe, Smart Bidding Exploration is the relevant lever.

Measure exploration as a controlled commercial rollout

The setting’s purpose is to uncover additional conversion potential, but neither higher conversion volume nor improved ROAS is guaranteed. Treating activation as proof of incremental value exposes your budget to an assumption the platform itself does not make.

A useful rollout keeps the decision reversible and the result interpretable:

1. Select one clean candidate. Start with an eligible campaign whose products, conversion actions, economics, and normal performance range your team already understands.
2. Save the baseline. Record the budget, Target ROAS setting, spend, conversion volume, conversion value, and actual ROAS before activation. Include the change date in your account log.
3. Set business guardrails. Define the maximum spend increase you can tolerate and the minimum actual ROAS or conversion value the campaign must produce. These thresholds should come from margin and acquisition economics, not from a generic platform benchmark.
4. Isolate the change. Avoid changing the budget, Target ROAS, product structure, or primary conversion actions at the same time. Stacked changes make it difficult to determine what caused the result.
5. Respect conversion lag. Do not make the keep-or-stop decision before the campaign’s normal conversion and reporting delays have passed. Use the delay observed in your own account rather than inventing a universal test period.
6. Review outcomes, not feature status. Ask whether spend increased, whether conversion volume and value followed, whether actual ROAS remained inside your guardrail, and whether the campaign stayed free of budget constraints throughout the evaluation.
7. Expand only after the first decision is clear. A positive result in one campaign justifies another controlled rollout. It does not prove that every Search, Shopping, or Performance Max campaign will benefit.

Be particularly careful with a campaign that appears to gain conversions while actual ROAS falls below the level your business can support. Extra volume is not inherently valuable. Conversely, judging only by ROAS can hide useful growth if the campaign produces materially more conversion value while staying above your economic floor. The guardrail must settle that trade-off before spend moves.

Treat DV360 API v5 as a migration project, not a version bump

Display & Video 360 API v5 adds useful controls, but its public-beta status changes how you should adopt it. A production cutover without integration mapping could disrupt audience, inventory, or exclusion workflows if the interface changes before the stable release.

Customer Match deserves attention first because this is not simply a changed resource shape. The capability is moving out of the DV360 API. Any job that assumes it can create, update, or otherwise manage those audiences through v5 has a missing dependency until the Data Manager API path is implemented.

Use a staged migration sequence:

1. Inventory the integration surface. List the applications, scheduled jobs, third-party connectors, and internal services that call the DV360 API. Mark which ones touch exclusions, inventory, audience targeting, Digital Out of Home, or Customer Match.
2. Separate Customer Match from campaign management. Identify the exact workflows that must move to the Data Manager API. Give that migration its own owner and acceptance criteria instead of hiding it inside the broader v5 upgrade.
3. Update dependencies deliberately. Use the latest client libraries before testing v5, then record the library version alongside your application change.
4. Test away from production. Exercise v5 in a development or non-production path. Compare the resulting targeting, exclusions, audiences, and inventory handling with the behavior your team expects from the existing integration.
5. Write regression checks around business outcomes. A successful API response is not enough. Confirm that the right exclusions remain applied, audience assignments land on the intended ad groups, and inventory settings match the campaign plan.
6. Monitor beta changes. Review the release notes and migration guidance before each promotion toward production. Public beta means an implementation that works now may still require changes before the stable release.

Give each change an owner and a stop condition

These updates cross media, analytics, engineering, and brand-suitability responsibilities. A single change register keeps decisions from falling between those teams.

• Media owner: validates Smart Bidding Exploration eligibility, chooses the candidate campaign, records the baseline, and decides whether the result meets the commercial guardrails.
• Analytics owner: confirms that conversion volume, value, spend, and actual ROAS are assessed consistently before and after the bidding change.
• Engineering owner: maps v4 dependencies, tests v5, and manages the Data Manager API work needed for Customer Match.
• Brand-suitability owner: verifies that advertiser-level YouTube video and channel exclusions still represent the organization’s placement policy.
• Release owner: blocks a production v5 rollout when a critical workflow lacks regression coverage or still depends on Customer Match management through the DV360 API.

Define the stopping rules before activation. Pause a broader bidding rollout when incremental spend fails your value or ROAS guardrail. Hold the API migration when exclusions, audience assignments, or inventory behavior do not match expected results. Do not move the goalposts mid-test simply to make the change look successful.

If you own media, begin by checking one campaign against the four eligibility gates. If you own the platform integration, locate every Customer Match job before you write v5 code. Those first actions expose the most consequential constraints while the cost of changing course is still low.

References

• Search Engine Land — Google Ads expands Smart Bidding Exploration to Performance Max and Shopping campaigns
• Search Engine Land — Google launches Display & Video 360 API v5 in public beta