Performance Max has always operated on a take-it-or-leave-it basis when it comes to where your ads actually appear. You hand Google your assets, your budget, and your conversion goals, and the algorithm decides the rest. For many advertisers, that has meant accepting that some portion of their spend lands on placements they would never have chosen manually. A test currently underway - labelled Partners in Alfa by Google - suggests that may be about to change, at least partially.
What the Test Actually Involves
According to reporting from Search Engine Roundtable, Google is testing a feature that lets selected advertisers choose whether their PMax campaigns run on the search partner network, the Google Display Network, or both. This is currently in alfa, meaning it is available to a small group of advertisers before any broader rollout.
The search partner network includes third-party sites that have agreements to show Google search ads - properties that sit outside core Google search domains but still carry search inventory. The Display Network, meanwhile, spans millions of websites, apps, and YouTube placements. In standard PMax, both are bundled in by default. The prospect of being able to opt out of either is a meaningful departure from how the campaign type has worked since launch.
It is worth being clear about what this is not. It is not full placement reporting. It is not the ability to exclude specific sites or build exclusion lists at scale. It is a binary inclusion choice at the network level. That is a modest step, but given where PMax started, it still represents a genuine move towards advertiser control.
Why Placement Control Has Always Mattered in PMax
One of the most persistent criticisms of Performance Max, since its rollout replaced Smart Shopping and Local campaigns, has been the opacity of spend distribution. Advertisers could see that conversions were being attributed to PMax, but breaking down exactly which channels or placements were driving that performance required workarounds - scripts, third-party tools, or reading between the lines of asset group data.
The concern with partner network and Display placements is not abstract. Display inventory in particular varies enormously in quality. A campaign optimising towards conversion signals can end up spending meaningfully on low-quality sites if the algorithm determines those placements are generating the signals it is chasing - even if the underlying conversion quality is poor. For advertisers running PMax alongside brand campaigns, there is also the question of how partner network traffic interacts with branded search and inflates attribution.
The ability to exclude these networks entirely would give advertisers a cleaner read on what PMax is actually doing. If you strip out Display and partner traffic and your conversion volume holds steady, you learn something valuable. If it drops sharply, you learn something equally valuable - and can make an informed decision about whether that volume is worth the trade-offs.
The Bigger Picture: A Gradual Shift in Google's Approach
This test does not exist in isolation. Over the past 18 months, Google has incrementally added reporting and control features to Performance Max - asset group level insights, search term category data, and channel-level spend breakdowns for some accounts (as recent coverage of PMax spend visibility confirms), among other additions. Each addition has been incremental and often limited to specific account tiers or alfa programmes before broader availability.
The pattern suggests Google is responding, at least partially, to sustained pressure from advertisers and agencies who have argued that handing over complete control to the algorithm is not a viable strategy for serious campaign management. Transparency and control are not the same thing, but transparency is usually a precondition for meaningful control - and Google appears to be moving in that direction, deliberately and slowly.
For UK advertisers in particular, where budgets in sectors like financial services, legal, and retail can be substantial, this kind of granularity matters. Compliance requirements around where certain ads can appear add another layer - the Display Network includes a wide range of content environments, and not all of them are appropriate for regulated industries. Partner network exclusion would be a useful tool for any advertiser who needs to demonstrate placement governance.
How to Position Your Campaigns Before This Rolls Out Broadly
If you are not currently in the alfa, the immediate action is to tighten your measurement foundation so that when these controls become available, you can actually use them meaningfully. That means having clean conversion tracking - ideally server-side - and a clear view of which conversions in your PMax reporting represent genuine business outcomes versus noise. If your attribution is messy now, the ability to toggle off Display placements will not tell you much.
It is also worth auditing your current PMax campaigns for what you already know about placement quality. Google's placement reports under the Display tab - accessible via the campaign's Placements section - give a partial view of where Display spend is going. Review this now. If you are seeing a high volume of spend on app inventory or low-quality content sites, document it. That baseline becomes useful evidence when you have the option to exclude and want to quantify the impact.
For campaigns where brand safety is a direct concern, speak to your Google account team about alfa access if you have a managed account relationship. Alfa programmes typically prioritise accounts with sufficient spend data, so smaller accounts may need to wait for a wider rollout - but flagging interest early is worth doing.
What This Does Not Solve
Selective network inclusion is useful, but it does not address the deeper structural challenge with Performance Max: the algorithm still controls creative selection, audience targeting, bid decisions, and channel prioritisation within whatever network you leave switched on. You can exclude Display entirely and still have limited visibility into whether your Search, Shopping, or YouTube placements are performing at the asset group level in a way that is meaningful for decision-making.
The transparency conversation around PMax has multiple dimensions. Placement-level control is one. Creative performance reporting is another. Channel-level budget capping is a third. Each of these has been a point of friction since PMax launched, and this alfa addresses only the first - and only partially. Advertisers who treat partner network selection as a complete solution to PMax opacity will be disappointed. Those who treat it as one useful tool among several, and who pair it with rigorous conversion tracking and structured creative testing, will get more from it.
The broader direction of travel is positive. Google is gradually acknowledging that full automation without visibility is not a sustainable offer for professional advertisers. But the gap between where PMax transparency is now and where it needs to be for proper campaign management remains significant. This test is a step. It is not a destination.