Google tests a Channels slider in Performance Max

Google is testing a Channels slider in Performance Max that lets an advertiser adjust how much the system prioritises conversions from each channel, and Search Engine Roundtable reported it on 26 August 2026 after it was spotted in-product and shared on LinkedIn by Heidi Sturrock. The channels named in the control are Google Search, Google Partners, Discover, Maps, YouTube, the Display Network and Gmail. The control is labelled Alpha, which is a stage earlier than beta, so it is not generally available and there is no published rollout timetable.

That single paragraph is close to the whole of what is confirmed. Everything below separates what the report states from what an advertiser might reasonably do with it, because this is the control Performance Max advertisers have been asking for since the campaign type launched, and a wish granted in Alpha is still a wish rather than a feature.

What the control actually does

The slider runs from lower prioritisation on the left, through an Auto setting, to higher prioritisation on the right. The mechanism Google states in the product is specific and it is worth quoting the shape of it rather than the vibe: a positive adjustment relaxes that channel's CPA, while a negative adjustment tightens it.

That wording matters more than the word prioritise. This is not a budget split and it is not a channel on and off switch. It is a bid-side instruction expressed through cost per acquisition tolerance. Pushing YouTube positive tells the system you will accept a higher cost per conversion from YouTube before it backs off. Pulling the Display Network negative tells the system to demand a cheaper conversion there before it keeps spending. The system still allocates; you are moving the price it is allowed to pay per channel, not carving the budget into slices.

Google's own in-product text alongside the setting says that Performance Max automatically optimises for total conversions across all channels, and that adjusting these settings will directly impact your performance. Read plainly, that is Google telling you that the default is the optimum for total conversions, and that a change is a deliberate departure from it.

The table below is everything Search Engine Roundtable's report establishes about the test, kept separate from interpretation so it is clear where the evidence stops.

Why this is the request advertisers have been making for years

Performance Max has always traded control for reach. One campaign spends across Search, YouTube, Display, Discover, Gmail and Maps, and the advertiser sees a single blended result. The complaint has been consistent since launch: when a Performance Max campaign underperforms, the advertiser can often see which surface is absorbing the money but cannot do much about it beyond restructuring assets, tightening audience signals or splitting the campaign apart.

A channel-level lever changes that. Sturrock described the upside in the case where the system has fixated on spending heavily in a channel that is not producing, and said the setting can help steer the system to back off faster than it would work out on its own. That is a real problem with a plausible fix, which is exactly why it needs handling carefully rather than enthusiastically.

The caveat that matters more than the feature

Sturrock attached her own warning to the feature, and it is the part of the story worth carrying into an account. Performance Max is designed to work across the entire customer journey. A user might discover a brand through a YouTube ad at the top of the funnel and convert days later through Search at the bottom of it. Her advice was to watch for attribution blind spots and to watch for shifts in performance over a longer window if the setting is used.

Relevant Audience's reasoning extends that in one direction. If your evidence for a weak channel is a last-click-shaped view of conversions, tightening that channel does not remove waste, it removes demand generation that the converting channel depended on. The Search conversions that look efficient today may be efficient partly because YouTube and Discover spent last week creating the demand that the branded query captured. Cut the top of the funnel using bottom-of-funnel evidence and the bottom of the funnel gets thinner a few weeks later, at which point the slider looks innocent because the damage arrives with a lag.

This is not an argument against the control. It is an argument that the control amplifies whatever attribution model you already trust, which makes the quality of that model the actual constraint.

What to look at before touching the slider

None of the following is a method the source endorses; it is what the stated mechanism implies about the evidence you would need. Three things are worth establishing first.

Look at channel-level performance in your own account rather than at general advice about which Performance Max surfaces are wasteful. Performance Max channel reporting has become more transparent over successive releases, and the question is whether your own data shows a channel consistently absorbing spend without producing conversions, or simply producing conversions that your model credits elsewhere.

Look at how your conversions are attributed. If the account is on last click, a slider that reallocates on CPA tolerance is being fed a model that structurally undervalues everything upstream. Knowing whether you are on data-driven attribution or last click changes what a weak-looking channel actually means, which is a broader point about how Google Ads accounts are measured before they are optimised.

Look at where a weak-looking channel sits in the path. A channel that appears early in converting paths and rarely takes last-click credit is the classic case for leaving alone, and it is the case where a negative adjustment is most likely to cost more than it saves. This is most obviously true of video, where YouTube advertising commonly does introduction work that a branded search closes days later.

What the source does not say

The honest list of unknowns here is longer than the list of knowns, and pretending otherwise would be the mistake. Search Engine Roundtable's report gives no rollout timetable and no date for general availability. It names no markets, so there is nothing to say about whether Thai accounts, or any specific country, are in the test. It states no eligibility criteria beyond contacting a Google rep for access if the control is not visible. It gives no guidance on how large an adjustment is safe, no numeric scale for the slider, and no indication of how the setting interacts with target CPA or target ROAS bid strategies at the campaign level.

It also gives no indication of whether the feature survives Alpha at all. Features at this stage are frequently changed or dropped, and Andrew Goodman's public comment on the post, that he would await the rollout, is a fair description of the current state. Anyone writing a process document around this today is documenting something that may not exist in three months.

What this means for Thai marketers

Search Engine Roundtable's report does not mention Thailand, and no market list has been published, so there is no basis for saying whether accounts in Thailand can see this control. If it is not in your account, the only stated route is asking a Google rep.

What follows is Relevant Audience's reasoning rather than anything the source states. The caveat lands harder in a Thai account than in a large US one for a structural reason: conversion volumes per campaign are often smaller here, and a smaller conversion count makes channel-level judgements noisier. A channel that looks like a failure across 40 conversions in a month may be within normal variance rather than genuinely broken, and a negative adjustment made on that evidence is a guess dressed as an optimisation. The same smaller volumes make data-driven attribution less stable, which is precisely the model you would want to be confident in before using a CPA-tolerance lever.

There is a second local wrinkle worth naming. Where a large share of conversions completes off-platform, through LINE, a phone call or a walk-in, the in-platform channel report is already an incomplete picture of what each surface contributed. Adjusting channel priority on that picture compounds a measurement gap rather than fixing it. The sequence that makes sense is to fix the measurement first and reach for the slider second, and in most accounts the first job outlasts the Alpha.

The practical answer for now is to read this as a signal about where Performance Max is heading rather than as a task. Google is at least willing to test giving channel-level control back to advertisers, after several years of not doing so. That direction is worth knowing. The control itself is not something to plan around until it leaves Alpha.

The original report, including the in-product screenshot, was published by Search Engine Roundtable.

FAQ

Can I use the Performance Max Channels slider right now?

Probably not, because it is labelled Alpha and is not generally available. Search Engine Roundtable reported on 26 August 2026 that the feature is rolling out and that advertisers should contact their Google rep for access if they do not see it, and no general availability date has been published.

What does a positive adjustment on a channel actually do?

Google's in-product text states that a positive adjustment relaxes that channel's CPA, while a negative adjustment tightens it. In practice that means you are changing the cost per conversion the system is willing to accept from that channel, not splitting the budget between channels and not switching a channel off.

Should I turn YouTube down if it shows no conversions?

Not on last-click evidence alone. Heidi Sturrock's caveat is that Performance Max is designed to work across the entire customer journey and a user may discover a brand through YouTube and convert days later through Search, so a channel with few last-click conversions may still be generating the demand that Search closes. Relevant Audience's view is that the attribution model behind the number should be checked before any negative adjustment is made.

Is the Channels slider available in Thailand?

The source does not say. Search Engine Roundtable's report names no markets and no eligibility criteria beyond contacting a Google rep, so there is no published basis for saying which countries are included in the Alpha.

Will this feature become a permanent part of Performance Max?

There is no indication either way. The control is labelled Alpha, which is a stage before beta, and the report gives no rollout timetable and no statement from Google about whether the test will graduate, be changed or be dropped.