One of the most persistent frustrations with Performance Max has been the opacity of spend allocation. You put budget in, conversions come out - or they don't - and working out which channels were responsible for either outcome required a lot of inference. That has started to change.
Google has updated the channel performance section within the PMax insights report to show spend over time, broken down per channel. It sounds like a minor update. It isn't.
What the Update Actually Gives You
Previously, the channel performance report gave you a snapshot of where budget was being allocated across channels like Search, Shopping, YouTube, Display, Discover, and Gmail. Useful, but static. You could see what proportion of spend had gone where, but you couldn't easily see how that allocation had shifted over a given period.
The new spend-over-time view changes that. You can now track channel-level spend trends across a date range, which means spikes become visible. If YouTube suddenly consumed a disproportionate share of your budget during a particular week, you'll see it. If Shopping spend dropped off around a key promotional period, that's now something you can identify rather than guess at.
This matters because PMax campaigns are managed by Google's bidding systems, not by manual channel selection. Spend decisions are made algorithmically. Having a time-series view of those decisions is the first step toward understanding - and eventually influencing - them more effectively.
Why Spend Spikes Are the Metric Worth Watching
Budget spikes in PMax are not unusual. The algorithm responds to signals - seasonality, competitor activity, audience behaviour, inventory availability - and can shift spend meaningfully from one channel to another in a short window. Without time-series data, those shifts were largely invisible unless you were correlating cost-per-conversion data manually.
Now you can see the spike directly. More importantly, you can correlate it. Did a YouTube spend spike coincide with a drop in conversion rate? Did a surge in Display spend happen to overlap with a period of poor return on ad spend? These questions were difficult to answer before. They're answerable now.
For UK advertisers running PMax across retail, finance, or services categories - where margins are tight and efficiency expectations are high - this kind of correlation analysis is not a nice-to-have. It's how you build a case for adjusting campaign structure, tightening asset groups, or making the argument internally for a budget reallocation.
The Broader Transparency Problem This Addresses
PMax has always asked advertisers to trust the system. Supply Google with good creative assets, solid audience signals, and a sensible target ROAS or target CPA, and the algorithm will optimise toward your goals. That proposition has always been complicated by the fact that 'trusting the system' is difficult when you can't see what the system is doing.
Google has been incrementally improving PMax reporting over the past two years. Search term insights, asset group performance, channel breakdowns - each update has added a layer of visibility that didn't previously exist. The spend-over-time addition continues that trajectory. It doesn't give advertisers manual channel controls, but it does give them data they can act on.
The practical effect is that advertisers can have more informed conversations with the algorithm, via campaign structure and signal inputs, rather than operating entirely blind. That's a meaningful shift in the working relationship between human strategists and automated bidding systems.
How to Use This Data in Practice
The most immediate use case is anomaly detection. Pull the channel spend-over-time report and look for periods where one channel's share deviated significantly from its average. Then overlay that with your conversion data, impression share, and any external factors you know about - a competitor promotion, a seasonal event, a site issue. Patterns tend to emerge quickly.
A second use case is informing asset strategy. If you consistently see Display taking a large share of spend without contributing meaningfully to conversions, that's a signal worth acting on. You can't tell PMax to avoid Display, but you can restructure asset groups, tighten audience signals, or adjust your bidding targets to steer the algorithm toward more productive placements over time.
A third application is client reporting and internal stakeholder communication. One of the recurring challenges with PMax is explaining its behaviour to people who aren't close to the platform. A chart showing spend by channel over time is considerably easier to interpret than a single-period allocation table. It gives non-specialists something concrete to engage with.
What It Still Doesn't Tell You
It's important to be clear about what this update doesn't provide. Knowing that spend went up on YouTube in a given week doesn't tell you why. The algorithm doesn't expose its reasoning. You can observe the output - the channel-level spend pattern - but the decision logic remains inside the system.
You also still can't control channel allocation directly. PMax does not allow advertisers to set a budget cap per channel or exclude channels from consideration - with the exception of brand exclusions and content suitability settings. So the reporting is genuinely useful for diagnosis and strategy, but it doesn't translate into direct operational controls.
That gap between visibility and control remains the central tension in managing PMax campaigns. This update moves the dial on visibility. Control, for now, still works through indirect means - campaign structure, asset quality, audience lists, and bidding targets. The skill in managing PMax is learning to use those levers well, and better data makes that considerably more achievable.
A Small Update With Real Diagnostic Value
Feature updates to reporting tools rarely generate much attention. They don't carry the headline appeal of a new campaign type or a bidding strategy change. But for practitioners managing PMax at scale, spend-over-time channel data is genuinely useful in a way that many higher-profile announcements are not.
If you're running PMax and haven't explored the channel performance insights section recently, this is a good reason to go back to it. The data is there. Using it well is where the advantage is.