Google Removes Legacy Ad Format Requirements; What Can Advertisers Control Now?

Google recently removed several legacy ad format requirements from its policies; those cover text, responsive, image, and form ads. If you've been around long enough to remember managing standard text ads, you might wonder what this means for how you run campaigns today.

The short answer: in practice, not much has changed. This update just clears out the policy documentation to reflect how Google Ads actually works now. The older rules governed formats that have largely been replaced by Responsive Search Ads, Performance Max, and other automated campaign types. Removing them is housekeeping, not a strategic pivot.

The more interesting question, and the one worth spending more time on, is what creative control looks like in the current environment, and how advertisers can still meaningfully influence outcomes when Google is making more of the decisions.

What Control Actually Looks Like Today

Advertisers don't have the same direct control they once had. Point-blank, period.

With RSAs, you're providing inputs, headlines, descriptions, and variations, and Google's system determines which combinations to serve based on predicted performance. With Performance Max, the automation goes further: asset groups, audience signals, and conversion data all feed a system that decides which shows to show to whom and when. Then there’s AI Max, which is about to replace Dynamic Search Ads.

We’re getting less control as we go, but advertisers do have influence. The levers are different, but they're real.

For Responsive Search Ads, pinning is the most direct tool available. You can lock specific headlines or descriptions to fixed positions, which guarantees those messages appear regardless of what the algorithm would otherwise choose. Use it selectively, since pinning too aggressively limits Google's ability to test and optimize combinations, but for brand-critical messaging, legal copy, or high-performing CTAs, it's a legitimate guardrail.

Beyond pinning, the quality and variety of your assets shape what gets served. Provide distinct, non-repetitive variations. If your headlines are variations of the same idea, Google doesn't have much to work with, and performance will reflect that. Group your copy around coherent messaging themes rather than mixing angles within a single ad, which gives the system clearer signals about what you're trying to communicate.

For Performance Max and Display, your primary levers are asset quality and structure. Don't rely on auto-generated images. Upload your own assets with intentional brand consistency, logos, color palette, formats, so that whatever combination gets assembled still looks like your brand. When possible, opt out of auto-generated images and regularly monitor what's running to remove underperforming or off-brand creative.

Asset group structure also matters in PMax. Segmenting by audience, funnel stage, or messaging theme gives Google more relevant material to work with in each context and helps prevent your awareness-stage creative from appearing in bottom-funnel placements.

Landing page alignment is worth naming here, too. A strong message match between your ads and the pages they point to influences which creative combinations Google favors. It's not a direct control mechanism, but it shapes the optimization signal in ways that benefit you.

How to Test Creative in an Automated Environment

Testing in Google today requires a different approach than it did when you could run discrete ad variations with explicit control over what showed and when.

The most effective method right now is testing messaging themes rather than individual elements. Within RSAs, this means building separate ads around distinct value propositions, say, Cost Savings vs. Efficiency vs. AI Messaging, and letting performance data tell you which angle resonates. This approach surfaces strategic insight rather than just tactical wins.

CTA testing is another high-signal variable. "Watch the Demo" versus "See How It Works" might seem like a minor copy difference, but the performance gap between a direct-action and a curiosity-driven prompt can be meaningful, especially across B2B audiences where buying intent varies significantly within the same campaign.

The discipline that makes creative testing work in this environment is restraint. Don't test too many variables simultaneously. Keep asset groups focused on a single theme. Rotate in new variations on a regular cadence and monitor closely enough to identify what's driving engagement, and retire underperformers before they drag down overall quality scores.

Work With the System, Not Against It

The broader shift here is that Google is increasingly in control of creative delivery, and that's unlikely to reverse. Trying to manually override the system at every turn usually results in worse performance, not better control.

The more useful posture is treating the system as something to guide with strong inputs rather than something to circumvent. Use clear, focused messaging within asset groups, high-quality owned assets, thoughtful segmentation, and clean conversion data feeding the algorithm. These inputs shape what the automation produces more reliably than any single toggle setting.

What this update signals, beyond the policy cleanup, is that Google is continuing to reduce legacy complexity and consolidate around AI-driven campaign types. Advertisers who adapt their workflow to influence inputs rather than chase direct output control will be better positioned as that trend continues.

The rules changed, but the goal hasn’t: put the right message in front of the right person at the right moment, even if the path to that goal now runs through the algorithm.