5 common Performance Max mistakes, and how to fix them

Performance Max is a goal-based, automated campaign type in Google Ads that optimizes across Search, Shopping, YouTube, Display, and Maps. It is now the most widely used campaign type across managed and audited Google Ads accounts. Most account owners treat it as a black box they should not question. The logic sounds reasonable: Google built it, Google has the data, so Google should know best where to spend the budget.

In practice, Performance Max is not that smart. Left unmanaged, it frequently wastes ad spend, and the waste is rarely visible without digging into the account. This is not unique to Performance Max. Most automated advertising systems reward whatever converts fastest, not whatever converts most profitably.

During an audit of an account spending €886,000 a year on Google Ads, Performance Max accounted for 72% of the total spend. Of that budget, 80% went to products with a ROAS under 2.0. The result was up to 19% in missed revenue, without spending a single extra euro.

Performance Max campaigns can still be a high-leverage tool for e-commerce brands. But the campaign type requires active oversight, not blind trust in the algorithm. Below are five mistakes found repeatedly across PPC audits, and the specific fix for each.

How Google Performance Max actually works

Performance Max through Google is a goal-based campaign type. An advertiser sets a spend level, an objective, and a group of creative assets, and Google's automation decides where and when to deliver ads across its full inventory. Every one of these controls interacts with the others once the campaign is live. Unlike a traditional Search campaign, there is no manual control over individual placements. The system relies on audience signals, historical conversion data, and real-time bidding to reach a target audience wherever it is most likely to convert.

In theory this is very useful. A single campaign can reach users across Google search, YouTube, Display, Discover, Gmail, and Maps, instead of an advertiser having to manage multiple Google campaigns separately. It also removes a layer of manual work that used to sit between Search campaigns and Shopping campaigns, which used to run as entirely separate line items.

The trade-off is visibility. Google provides a basic summary of performance, but the platform does not make it easy to understand exactly which channel, which asset group, or which product is actually driving results, or where ads appear most often within that channel. Getting this detail usually means several clicks through separate report tabs rather than one clear view. That is the gap teams need to close manually, and it is where most of the value leaks out of an account.

Key takeaways

Unbalanced budgeting is the most damaging and most common issue. On the €886k account above, 80% of PMax spend went to low-performing products with a ROAS under 2.0. Reallocating Shopping spend toward profitable SKUs, increasing their share from 25.6% to 45%, lifted overall ROAS from 3.43 to 4.26. That is up to 19% more revenue at a constant spend level.

Brand cannibalization is close behind. Without exclusions in place, Performance Max chases easy conversions on branded search terms the advertiser would have captured organically anyway, which eats into real incremental growth.

Flawed acquisition tracking sits underneath both. The "New Customer Acquisition" toggle demands near-perfect tracking pixels and continuously updated audience lists. Most accounts do not have this infrastructure in place, and the option often underperforms a dedicated campaign as a result.

Asset optimization waste rounds out the list. Leaving asset auto-generation switched on pushes budget toward YouTube, Display, and Search, at the expense of Google Shopping, which remains the primary revenue driver for most e-commerce advertisers.

Mistake 1: spend dispersed across the wrong channel

Performance Max campaigns serve ads across Google Shopping, Search, YouTube, Display, and Maps from a single campaign. That is the whole pitch: one campaign type across the full Google ecosystem, instead of managing multiple Google campaigns separately.

For e-commerce brands, Google Shopping is the channel that actually drives growth. Shopping generates 3x the sales volume of Search, with roughly 10% higher ROAS. Yet on most audited accounts, only 40% to 60% of PMax spend goes to Shopping. The remainder gets scattered across lower-converting channels that dilute overall campaign performance.

Target: allocate at least 85% of your PMax budget to Google Shopping.

How to check it natively: open your primary Performance Max campaign, go to Insights, then Channel Performance, and review the Search Network row under the sub-line "Ads using product data."

Why this is hard to see: standard Google Ads reports hide granular channel placement data by design. This is precisely the kind of channel dilution that Clarmix PMax Insights was built to expose, since it surfaces breakdown metrics you cannot find natively inside the Google Ads dashboard.

This matters more for e-commerce than for lead generation accounts. A search ad on Google search can still convert well for a service business with a long consideration cycle. For an online retailer, a user scrolling YouTube or browsing Display inventory is rarely in a buying mindset, and PMax campaigns will still spend there unless an advertiser actively limits it.

Mistake 2: poor Shopping budget distribution across products

Google does not automatically shift budget toward your highest-margin products. In practice, the algorithm often over-invests in low-profit SKUs while under-investing in bestsellers, because it optimizes for conversion signals rather than actual profitability.

A SKU-level analysis, based on spend versus actual ROAS, sorts every product into one of six performance buckets:

Profitable products combine strong spend with high ROAS. High potential products have low spend but high ROAS, meaning there is room to scale them. Expensive products carry high spend with low ROAS. Low potential products have both low spend and low ROAS. Zero conversion products have spend above zero but no conversions at all. Zombies have zero spend and, in most cases, zero visibility.

Analyzing a real €16,590 Shopping spend breakdown makes the pattern concrete. Only 24% of spend went to Profitable products and 11% to High potential products, meaning just 35% of the budget was working efficiently. The remaining 65% was wasted: 29% went to Expensive products, 20% to Low potential products, and 16% to products with zero conversions. On top of that, 2,388 zombie SKUs received zero impressions, leaving significant revenue opportunity untapped.

Managing this manually across thousands of SKUs is not realistic for most teams. Clarmix Labelizer automatically labels SKUs in real time, based on live spend and ROAS data, so advertisers can cut costly products, exclude zero-conversion items to protect margin, and automatically scale the SKUs that are already proven to convert.

The right option depends on catalog size. A retailer with a few hundred SKUs can review this bucket breakdown manually on a monthly basis and still catch most of the waste. A retailer with several thousand SKUs, or a product feed that changes weekly, cannot keep up manually, and the zombie bucket alone can represent a meaningful share of missed revenue. A messy or outdated feed makes the problem worse, since Performance Max relies on that same feed data to decide where to spend in the first place. Automating the labelling step is less about convenience and more about making the data usable at all.

Mistake 3: brand keywords absorbing non-brand targeting

Performance Max is built to prioritize low-hanging fruit. If a company name is active inside the campaign with no exclusions, the algorithm will capitalize on high-intent branded searches to artificially inflate its own reported ROAS.

That budget targets users who would have converted anyway, through organic search or direct navigation. It looks like performance, but it is cannibalizing incremental revenue rather than generating it.

Set up brand exclusions: go to the Performance Max campaign settings, scroll to Brand Exclusion, and apply the exclusion list.

Apply negative keywords: open the campaign, select Audiences, Keywords, and Content, then Keywords from the left menu, and add company terms to the negative keyword list.

Both steps matter. The exclusion controls and negative keyword lists work differently under the hood, and skipping one leaves a gap the algorithm will find.

Mistake 4: misconfigured New Customer Acquisition settings

To address advertiser complaints about retargeting existing customers, Google introduced New Customer Acquisition settings, generally offered as three options: increased bids for first-time buyers, bidding restricted to first-time buyers only, or reporting only.

These options sound like exactly what most advertisers want. In practice, New Customer Acquisition requires near-perfect pixel tracking and a continuously updated audience list of existing customers. Most accounts do not have this infrastructure at the level Google's automation needs. Across repeated tests, enabling these native controls has consistently produced worse results than running dedicated acquisition campaigns with clean tracking.

Recommendation: disable native New Customer Acquisition settings inside Performance Max. Handle new customer acquisition through a dedicated campaign with reliable tracking and active audience management instead.

Directly below the acquisition control sits Customer Retention optimization. This should generally be disabled too, since it lets Performance Max spend on retargeting users who are already loyal customers, rather than reaching untapped audience segments.

Mistake 5: automatically created video and image assets left switched on

Performance Max includes an option that lets Google auto-generate video and image assets, rewrite ad text, and dynamically alter final destination URLs, all without advertiser review.

Auto-generated video assets often resemble low-quality slideshows that do not match a company's creative standards. The bigger cost is what happens to budget: enabling asset optimization triggers Performance Max to shift spend toward YouTube, Display, and Search, drawing budget away from Google Shopping, which is where most e-commerce conversions actually happen.

Recommendation: turn off Automatically Created Assets and Component Optimization. Keep manual control over ad creative, and use dedicated Demand Gen campaigns when video or image placements are genuinely part of the strategy.

This does not mean rejecting AI-generated creative outright. It means choosing when to use it deliberately, inside a campaign built for image and video content, rather than letting Performance Max make that choice silently inside asset groups meant for Shopping and Search. Every asset group carries its own set of creative assets, headlines, and descriptions, and reviewing them individually still saves time compared to troubleshooting a budget leak after the fact.

Actionable Performance Max strategy checklist

Five controls are worth verifying directly inside the Google Ads interface, ideally on a recurring basis rather than once:

Google Shopping share should sit at 85% or higher of the total Performance Max budget. This is easiest to audit through Clarmix PMax Insights, since native reporting hides the breakdown.

Bestseller allocation should reach at least 45% of Shopping budget directed to Profitable SKUs, which can be automated using Clarmix Labelizer instead of manual SKU review.

Exclusion lists should be active in both the campaign controls and the negative keyword list, not just one of the two.

New Customer Acquisition and Customer Retention toggles should both be turned off unless the account has the tracking infrastructure to support them.

Asset auto-generation and URL expansion should be turned off across every asset group, so creative and destination decisions stay with the advertiser.

None of these five fixes require a platform migration or a new tracking setup. Each one is a control that already exists inside the account. The general pattern behind all five is the same: Performance Max defaults to the option that is easiest for the algorithm to test, not the option that protects margin. Reviewing these controls on a recurring basis, rather than once at campaign launch, is what separates accounts that treat PMax as a genuine growth channel from accounts quietly funding it out of their own profit.

For advertisers managing this across several accounts or a large product catalog, the value of a dedicated PMax audit tool comes down to time. Manually pulling channel share, SKU-level ROAS, and asset configuration inside the native Google Ads interface is possible, but it does not scale past a handful of accounts. That is the specific gap Clarmix was built to close for Google Shopping and Performance Max management.

Where teams get this information wrong

Most of the confusion around Performance Max comes down to where teams go to look for detail. The Google Ads help center offers basic guidance on each setting, but it rarely explains how one setting interacts with another, or why a good result on paper does not always translate into good margin.

A few rules of thumb help align expectations before diving into account-level changes. First, do not judge a single setting in isolation. Asset auto-generation, exclusion lists, and Shopping share all interact with each other, so changing one without checking the others rarely produces a clean result. As a general rule, a fix that looks good on one metric should always be checked against the other four before it is treated as finished. Second, understand that Google's own reporting page is built to show a general summary, not the level of detail an e-commerce marketer actually needs to make a good decision. Third, continue reviewing the account after a fix is applied. A setting that looked correct at launch can drift over weeks as the algorithm re-learns, so a single audit rarely holds for more than a month.

Choosing the right level of oversight depends on team size and catalog complexity. A lean team managing one account can work through the checklist above inside the native Google Ads interface, adjusting each setting as fresh data comes in. A larger operation managing several client accounts benefits from a tool built specifically for this, since it turns scattered account data into a single page that shows exactly where spend is going and why, rather than requiring a manual click through several report tabs.

None of this replaces good campaign strategy. It ensures the strategy already in place is not being quietly undermined by controls most teams never revisit after initial setup. Understanding how Performance Max campaigns work at the configuration level is what makes the difference between an account that improves after a fix and one that reverts within a few weeks.

Where to find reliable information on Performance Max

Google's own help center and support pages remain the most reliable source for how a specific setting is meant to function , since third-party advertising blogs sometimes describe outdated versions of the interface, and general PPC advice rarely accounts for how Performance Max behaves differently from a standard Search campaign. For anything related to spend allocation, exclusion lists, or asset configuration, checking the official documentation first, then verifying against real account data, avoids acting on advice that no longer applies.

Case studies and audit breakdowns, like the account example used throughout this guide, are useful for understanding what good and bad allocation actually looks like in practice. General benchmarks help set expectations, but the only detail that matters for a specific account is that account's own channel share, SKU-level ROAS, and asset performance, checked directly rather than assumed from an industry average. The real value of an audit is not the benchmark itself, it is knowing exactly where a given account sits against it, and whether the current settings still optimize for that position or need adjusting.

Frequently asked questions about Performance Max results

Is Performance Max worth using for e-commerce?

Yes, when Performance Max is actively used the right way. Performance Max through Google can be a genuinely strong campaign type for e-commerce advertisers, since it gives access to Search, Shopping, YouTube, Display, and Maps from a single campaign. The channel mix and budget allocation still need regular human oversight, or the algorithm will drift toward low-quality placements and low-margin products.

Why does Performance Max spend so much outside Google Shopping?

By default, Performance Max is free to allocate budget across every channel in its inventory, including Search and Display. Without a clear Shopping share target and without turning off automatically created assets, the campaign type will often push a meaningful portion of spend into channels that convert less efficiently for e-commerce than Shopping does.

Do brand exclusions actually reduce reported ROAS?

Reported ROAS can drop after these exclusions are applied, because branded conversions that would have happened anyway are removed from the mix. That is the point: the metric becomes a more honest read on incremental performance, rather than one inflated by traffic the company already owned.

How often should Performance Max campaigns be reviewed?

Weekly review of channel share, SKU-level ROAS, and asset settings is a reasonable baseline for active e-commerce accounts, with a deeper audit monthly. Accounts with large or fast-moving catalogs benefit from tools like Clarmix that surface this data continuously, rather than relying on manual checks inside the native Google Ads dashboard.

What high-quality results actually look like

A well-run Performance Max account does not need every control maximised at once. It needs a Shopping share close to 85%, a bestseller allocation near 45%, exclusion lists active, acquisition toggles disabled unless tracking supports them, and asset auto-generation switched off. Choose the fixes that apply to your account's current data rather than applying all five blindly, since the exact numbers on any given website or benchmark study will not match every catalog.

Google will not surface this level of detail on its own results pages inside the Ads interface. Checking channel performance, SKU-level buckets, and asset settings requires either a manual pull each week or a tool that already does it. Either option is good. What does not work is trusting that Performance Max campaigns will self-correct, since nothing in how the algorithm operates gives it a reason to prioritise margin over volume on its own.

Fixing these five points will not save time on its own in the first week. It takes a review cycle or two before the account settles into the adjusted allocation. What it does allow is a account that spends closer to where it should, instead of quietly funding the channels and products working against its own return.

‍