Yes, Performance Max can and often does capture branded searches, and that inflates the numbers you report to your board. Pull your PMax search terms today and calculate what share of cost and conversion value came from your own brand name. Google Ads gives you the controls to fix this, and specialist agencies run this exact audit for eCommerce clients every month.
TL;DR:
- Branded searches captured by Performance Max inflate reported ROAS and are often from previous demand or organic traffic, not incremental conversions.
- Regularly compare PMax search term data with branded search share and auction insights to identify and quantify actual brand leakage.
- Proper sequence involves building a branded search campaign first, then applying brand exclusions to PMax, to avoid losing top impression share to competitors.
- Ongoing audit and monitoring are essential because PMax continually re-learns and can reintroduce brand leakage if controls are not maintained.
- In certain cases, such as high competition or resellers, paying for branded conversions deliberately may be justified, but it should be managed as a separate, controlled line item.
Table of Contents
- What PMax brand leakage is and why it distorts blended ROAS
- Audit steps: how to detect and quantify brand leakage now
- Controls: brand exclusions, negatives and campaign structure compared
- Operational checklist and monitoring cadence to keep leakage controlled
- When you might accept branded conversions in PMax (exceptions and trade-offs)
- Oxedent practitioner perspective and offer
- Case studies and examples of brand leakage in practice
- What the audit actually tells you
- Get a free audit before you touch your PMax settings
- Primary documentation and practitioner reading
- Sources
- FAQ
What PMax brand leakage is and why it distorts blended ROAS
Brand leakage happens when Performance Max spends your budget on people who were already searching for your brand name, then reports those conversions as if the campaign won them from scratch. You would have picked up most of that revenue anyway through organic search or direct traffic, so the campaign’s reported ROAS looks stronger than its actual contribution.
The mechanism is straightforward. PMax optimises towards your stated goal, whether that is conversion value or ROAS, and branded queries are almost always the cheapest, highest-converting traffic available. Left alone, the algorithm will lean into that inventory because it is the fastest way to hit its target, not because it is generating new demand.
Google’s brand exclusions help, but they have limits worth knowing before you rely on them:
- Exclusions apply across Search, Shopping and YouTube search inventory, but Google’s own documentation notes they will not automatically stop every non-search placement from serving on brand terms.
- The “allow Shopping ads” checkbox lets you block branded text ads while keeping Shopping listings visible, which matters for retailers who still want product visibility on their own brand searches.
- Ongoing search-term checks remain necessary because PMax adapts as it learns, and a control that worked last quarter can drift.
Audit steps: how to detect and quantify brand leakage now
Before you touch a single setting, you need numbers. The audit below takes an afternoon and gives you a defensible figure for how much of your PMax performance is genuinely incremental.
- Export the Performance Max search terms report alongside standard Search campaign search terms, so you can compare brand share across both channel types.
- Pull your branded Search impression share and Auction Insights to see how much of your own brand demand is already covered elsewhere.
- Segment PMax results by ad format (Shopping versus text), since the format breakdown in the search terms report determines which remedy applies.
- Classify every term using a brand token list: your brand name, common misspellings, subsidiary names and reseller phrases such as “official store” or a marketplace SKU.
- Calculate three figures: brand share of cost, brand share of conversion value, and non-brand ROAS once branded spend and revenue are stripped out.
Regex matching against a maintained token list is the fastest way to classify a large export, and practical guidance on classifying search phrases can help you build that list properly rather than guessing at variants.
A Performance Max brand leak audit methodology recommends exporting search terms, classifying them by brand tokens and misspellings, then recalculating ROAS excluding brand entirely. Do this by format: Shopping and text ads behave differently under exclusion, and treating them identically will hide where the real leakage sits.
Pro Tip: Keep the non-brand ROAS figure separate from your headline PMax ROAS in every report you send internally, so nobody mistakes blended performance for incremental performance.
Controls: brand exclusions, negatives and campaign structure compared
Three controls handle brand leakage, and they are not interchangeable. Knowing what each one covers, and where it stops, determines whether your fix actually holds.
Google’s own reporting shows that PMax often captures branded search queries because they are the cheapest, best-converting traffic the algorithm can find, which is exactly why blended ROAS looks strong until you isolate brand.
- Brand exclusions, applied through Google’s brand settings, cover Search, Shopping and YouTube search inventory and are prioritised over any brand inclusions you have set.
- Account-level negative keywords are narrower: they block specific query strings but do not carry the same brand-list logic, so they miss variants and misspellings a proper exclusion list catches.
- A dedicated brand Search campaign, built and verified before you apply any exclusion, should aim for close to full impression share on your own name so you are not handing that slot to a competitor mid-transition.
- Bid ceilings, audience signals and asset optimisation give you finer control over how aggressively PMax chases branded inventory once the bigger levers are in place.
Sequence matters here. Build the branded Search campaign first, confirm it is serving and winning the top position, then apply exclusions to PMax. Skip that order and you risk a gap where nobody, including a competitor bidding on your name, holds that slot. Google’s step-by-step guide to brand exclusions covers the “allow Shopping ads” checkbox in detail, which is the setting most eCommerce accounts need.
Operational checklist and monitoring cadence to keep leakage controlled
Leakage does not stay fixed once you have applied exclusions. PMax keeps learning, and controls drift as new assets, audiences or seasonal campaigns come online. A repeatable cadence catches that before it shows up in a board report.
- Launch the brand campaign, confirm it is serving, then apply brand exclusions to PMax, checking Search, Shopping and YouTube search inventory separately.
- Weekly, export search terms from PMax and Search, scan for anomalies, and review the asset report for anything pulling in branded queries unexpectedly.
- Monthly, review impression share trends and Auction Insights to catch competitors bidding on your name, and check Shopping unit overlap between your brand campaign and PMax.
- Around major seasonal events, run the full audit again, since promotional pushes tend to loosen the controls you set earlier in the year.
- Record every change with a date, the setting adjusted and the expected effect, so stakeholders can trace performance shifts back to a specific action.
Our guide to audit-first Performance Max setup walks through the asset and report checks in more depth if you are building this cadence from scratch.
When you might accept branded conversions in PMax (exceptions and trade-offs)
Excluding brand entirely is not always the right call. Some situations justify paying for branded clicks deliberately, as long as you price and report that spend honestly rather than letting it hide inside a blended figure.
- Competitor bidding on your name: if rivals are bidding on your brand terms, automated coverage can defend the slot while your dedicated campaign catches up.
- Reseller ranking: authorised resellers competing for your branded Shopping placements may justify keeping PMax active there rather than ceding the auction.
- Category ambiguity: when your brand name overlaps with a generic product category, exclusion lists can accidentally block valuable non-brand queries too.
- Shopping unit visibility: retailers sometimes need PMax active on brand searches purely to keep Shopping units visible, even with text ads excluded.
Where you choose to leave brand open, treat it as a controlled line item with its own budget and its own reported ROAS, never folded into the main PMax figure.
Oxedent practitioner perspective and offer
Oxedent runs eCommerce PPC exclusively, managing Google Ads, Facebook Ads, Google Shopping and Performance Max for online retail brands with a focus on profitability and ROAS rather than clicks or impressions. Across client audits, branded leakage inside PMax is one of the most consistent findings. An offered free audit checks brand share, exclusion setup and Shopping overlap, and flags exactly where budget is buying conversions you already owned.
Case studies and examples of brand leakage in practice
A recurring pattern shows up across retail accounts: a PMax campaign reports a strong ROAS, then a search-term export reveals a large share of that value came from the brand’s own name, often people who had already searched the brand elsewhere or clicked through from an email. Once the branded terms are stripped out, the non-brand ROAS is meaningfully lower, and the campaign’s real contribution to new demand looks very different from the headline figure.
The practitioner literature on this cannibalisation pattern documents the same sequence repeatedly: a retailer sets a ROAS or conversion value goal, PMax gravitates towards the cheapest converting traffic available, which is almost always the retailer’s own brand terms, and the account reports a healthy return that is largely recycled demand rather than incremental sales.
The fix in each of these cases follows the same order: build a dedicated brand Search campaign, confirm it is serving and holding the top position, then apply brand exclusions to PMax while testing the “allow Shopping ads” checkbox so product listings stay visible. Retailers who skip the sequencing step, and exclude before the brand campaign is live, tend to see a temporary gap where a competitor’s ad appears on their own brand search, which costs more in lost trust than the exclusion saves in wasted spend.
What the audit actually tells you
The conventional advice on this topic stops at “add brand exclusions and move on,” which misses the harder part: knowing whether the branded traffic PMax was capturing was ever incremental in the first place. Most accounts never run that calculation, so they either leave leakage unchecked or exclude brand entirely and lose visibility they actually needed, such as Shopping units on their own name.
The audit matters more than the fix. Once you know your real non-brand ROAS, the decision between exclusion, a controlled branded line item, or a mixed approach becomes obvious rather than a guess. Readers should prioritise the measurement step first: pull the search terms, classify the tokens, run the maths. Only then decide which lever to pull, because the same PMax account can need different controls for Shopping and text inventory. Treat brand exclusions as a starting point, not a fix you set once and forget, since PMax keeps re-learning and will find its way back to branded queries if nobody checks.
— Biplab
Get a free audit before you touch your PMax settings
This agency runs PMax audits, sets up brand exclusion structures and manages Google Shopping campaigns for eCommerce brands that want their ROAS figures to mean something.
Our guide to campaign budget allocation across Performance Max covers how we split budget once brand leakage is under control. Start with the free Google/Facebook Ads audit to see exactly where your account stands.
| Service | What it covers | Where to look |
|---|---|---|
| Free Google/Facebook Ads Audit | Brand share, exclusion setup, Shopping overlap review | Oxedent |
| PPC Management | Ongoing PMax and Search campaign management from £350 per month | Oxedent |
| Google Shopping Management | Feed optimisation and Shopping campaign structure | Oxedent |
Primary documentation and practitioner reading
For the underlying detail, see Google’s own pages on brand settings and brand exclusions, alongside our own breakdown of blended ROAS and how branded traffic skews it.
Sources
- About brand settings for Search and Performance Max
- The Performance Max Brand Leak Audit: Measuring the Spend You Would Have Won Anyway – PPC Hero
- Why Your Performance Max Campaign Is Cannibalizing Your Branded Search — SUPERMETRIC
FAQ
Is PMax a Google product?
Yes, Performance Max is a Google Ads campaign type that serves ads across Search, Shopping, Display, YouTube, Discover and Gmail from a single campaign, automating placement and bidding decisions.
Where does PMax show up?
PMax inventory spans Google Search results, Shopping listings, YouTube, Gmail, Discover and the Display Network, which is part of why isolating brand leakage requires checking multiple report types rather than one.
How to check PMax?
Export the Performance Max search terms report and segment it by ad format, then cross-reference against your branded Search impression share to see how much of the reported performance is genuinely new demand.
What is the fastest way to reduce PMax brand leakage?
Build a dedicated brand Search campaign, confirm it holds top position, then apply brand exclusions to PMax, using the “allow Shopping ads” checkbox if you still want product visibility on brand terms.
Does excluding brand always improve PMax performance?
Not automatically. Exclusion removes leaked spend from your reported ROAS, but if a competitor is bidding on your name or you rely on PMax for Shopping visibility, a controlled branded line item may serve you better than a full exclusion.