Blog Playbooks

Performance Max Search Terms: How to Actually See Them (2026)

You can see Performance Max search terms now - Google shipped a native report in 2025. Here is exactly where it lives, what it still hides, and a triage framework for what to do with it: brand cannibalization, junk placements, channel waste, and negatives up to the 10,000 cap.

Yes, you can see Performance Max search terms - natively, in Google Ads, without a script or a workaround. As of mid-2026 the report lives inside the campaign: open Campaigns, click into your Performance Max campaign, and under Insights & reports open Search terms. That is the whole trick. Google rolled out full search terms reporting for PMax during 2025, and any article still telling you “Performance Max hides your search terms - here’s a hack to see them” was written for a product that no longer exists.

The interesting part is what comes after. Seeing the terms turned out to be the easy half of the problem. The report shows you real queries with real metrics, but it does not tell you which terms are your own brand being resold to you, which placements are junk, which channel is eating the budget, or what to add as a negative. This guide covers both halves: exactly where the visibility now lives and what it still hides, then a triage framework for turning a few thousand rows into decisions - including the negative-keyword mechanics that changed dramatically in 2025.

Where the Performance Max search terms report lives

The click path, as of mid-2026:

  1. In Google Ads, open Campaigns in the left navigation.
  2. Click the name of your Performance Max campaign to enter it.
  3. In the campaign’s page menu, open Insights & reports.
  4. Click Search terms.

You get a table of actual search queries the campaign matched, with performance metrics per term, filterable and sortable like the Search-campaign report you already know. You can select terms and act on them from the report. The same data is exposed through the Google Ads API, which is how third-party tools read it at scale.

Two dated notes, because the timeline is exactly why so much advice on this topic is stale:

  • Before 2025, this report did not exist. What you had was “search terms insights” on the Insights page - queries bucketed into aggregated categories (“search themes” like “running shoes for women”), with no per-query rows and no way to act on individual terms. An entire cottage industry of scripts, spreadsheet exports, and “PMax transparency” blog posts existed to route around that. Google shipped native per-term reporting for PMax during 2025 and, in the same wave of changes, raised the PMax negative-keyword cap from 100 to 10,000 (announced March 2025) and rolled out channel-level performance reporting.
  • This is now table stakes across the industry. Microsoft Advertising brought equivalent search-term transparency to its own Performance Max product in May 2026. Visibility into automated campaign types stopped being a differentiator and became the baseline expectation.

If you read a guide that claims PMax search terms are unviewable, check its date. Anything written before 2025 is describing a different product.

What the report shows - and what it still hides

The native report is genuinely useful. It is also not the whole picture, and knowing where its edges are matters more than celebrating that it exists. As of mid-2026:

What you get. Real queries, per campaign, with impressions, clicks, conversions and cost-level detail - enough to rank terms by money spent and judge them individually. This is the same class of report Search campaigns have had for years, applied to PMax.

What stays hidden:

  • Low-volume queries are aggregated away. Like the Search search-terms report since Google’s 2020 privacy changes, queries below volume thresholds are not shown individually. In practice a meaningful share of your PMax spend can sit behind rows you will never see. How large that share is varies wildly by account - we call it the black-box share, and measuring it for your own account (visible term cost divided by total campaign cost, subtracted from one) is more honest than quoting anyone’s average.
  • Search terms only cover search-shaped traffic. Performance Max serves across Search, Shopping, YouTube, Display, Discover, Gmail and Maps. A YouTube impression has no search query behind it. So the search terms report explains the search-flavored slice of your spend - which is often a minority of impressions - and says nothing about the rest.
  • No term-by-channel split. You can see what someone searched, and separately (via the channel-level reporting Google rolled out in 2025) roughly how spend distributes across channels - but not which channel a given term served on, or a clean term-level cost breakdown by surface.
  • Placement reporting is impressions-only. The PMax placement report - where your ads appeared across websites and apps - carries impressions but no clicks, cost, or conversions, by API design. You can see that a mobile game showed your ad 40,000 times; you cannot see what it cost you. Any tool or article that puts a dollar figure on PMax placement waste is estimating, whether it admits it or not.

None of this makes the report useless. It makes it a starting point that needs interpretation - which brings us to the actual problem.

Seeing the terms was never the hard part

Here is the uncomfortable thing about 2026: PMax search-term visibility is solved, commoditized, free, and native. And most accounts we see have never acted on a single row of it.

That is not because advertisers are lazy. It is because a report is not a decision. Open the search terms report on a real account and you get a few thousand rows. Which ones are fine? Which are your own brand name - traffic that would likely have found you anyway, now billed through PMax? Which are junk that deserves a negative? Which are one-off noise not worth touching? The report answers none of that. It hands you evidence and walks away.

This is why “how to see PMax search terms” is the wrong question to stop at. The valuable question is “what should I do about them” - verdicts, not visibility. Everything below is a framework for getting from the raw report to three specific verdicts, manually, in under an hour. (If you want the same triage run automatically, that is what our Performance Max suite does - covered honestly at the end.)

The triage framework: three questions to ask the report

1. Is PMax buying your own brand?

The single most expensive pattern hiding in PMax search terms is your own name. Performance Max optimizes for conversions, and the cheapest conversions in any account are people who already searched for you. Left unguided, PMax will happily spend on brand queries, harvest those near-certain conversions, and report them as wins - taking credit for demand you already owned, and sometimes cannibalizing a brand Search campaign that would have caught the same click for less.

How to check manually:

  1. Open the PMax search terms report (path above).
  2. Filter or search for your brand name, product names, common misspellings, and your domain.
  3. Total the spend and conversions on those rows, and compare them against the campaign’s overall numbers. Ask: what share of this campaign’s reported conversions came from people who already knew our name?
  4. If you run a brand Search campaign, compare the cost per conversion on PMax brand terms against it. PMax brand traffic that costs more than your brand Search campaign is pure leakage.

There is no universal “acceptable” brand share - a retailer with a generic name and a company with a famous one live in different worlds. But if a large slice of your PMax conversions trace back to brand queries, the campaign’s real (incremental) performance is worse than its reported performance, and every budget decision built on the reported number inherits the error.

What to do about it: Google Ads has offered brand exclusions for Performance Max since 2023 - brand lists you attach in campaign settings to stop PMax serving on brand searches - and since 2025 you also have the negative-keyword capacity to block brand variants directly (details in the negatives section below). Whether you should exclude brand is a judgment call: some advertisers deliberately let PMax defend the brand SERP. The point of triage is that it becomes your call, made on numbers, instead of a default you never noticed. This deserves its own deep-dive, and we have one: does PMax cannibalize your Search campaigns?

2. Which terms and placements are junk?

Second question: of the spend that is not brand, how much went to traffic with no realistic path to a conversion? This splits into two surfaces with very different evidence quality.

Junk search terms - real cost, real verdicts. Sort the search terms report by cost and read the top 50 rows with zero conversions. You are looking for the classic patterns: job-seeker intent (“careers”, “salary”), free-seekers, DIY intent when you sell a service, other companies’ product names you do not sell, and plain gibberish matches. Because these rows carry actual cost data, the verdict is clean: enough clicks, enough cost, zero conversions - that is wasted spend, the same standard we apply everywhere (our wasted ad spend guide publishes the exact thresholds we use). These terms are your negative-keyword shortlist.

Junk placements - impressions-only, so keep your claims honest. Run the Performance Max placement report (in the UI, look under your reporting section for the predefined PMax placement report; as of mid-2026 it still reports impressions only). You cannot attach a cost figure to any row, but patterns still tell you plenty about where Google is expanding your delivery:

  • Mobile game apps - endless rewarded-ad inventory, famously accidental-tap-prone.
  • The anonymous placeholder - impressions Google attributes to an unnamed property, which you cannot evaluate at all.
  • Kids’ content - nursery rhymes and cartoon channels showing B2B ads is a real thing that really happens.
  • Made-for-advertising-style domains - coupon/quiz/horoscope/wallpaper sites and parked or numeric domains that exist to sell ad space, not to have readers.

High impression volume on these patterns is a hygiene signal, not a dollar figure - treat it as “Google is filling my budget’s edges with low-grade inventory” and respond with placement exclusions, tighter audience signals, and better creative coverage. Be suspicious of anyone who tells you exactly how many dollars your placements wasted; the data to know that does not exist.

3. Where is the money actually going by channel?

Third question: how does spend split across Search, Shopping, YouTube, Display, Discover, Gmail and Maps - and does any channel’s cost per result stand apart from the rest? Google’s channel-level performance reporting (rolled out in 2025) finally lets you see the distribution instead of guessing at it.

What you are looking for is an outlier: a channel absorbing a serious share of spend at a cost per conversion several times the campaign’s Search-flavored traffic. That pattern usually means PMax has found cheap inventory that produces impressions and clicks but not customers, and is filling budget with it.

Honesty about the levers: seeing channel waste is easier than acting on it. PMax does not give you a simple “turn this channel off” switch the way manual campaigns do; as of mid-2026 your practical levers are indirect - creative coverage (campaigns without video get video made for them, and served heavily), audience signals, brand and negative exclusions, and in some cases restructuring what you ask PMax to do versus what you keep in dedicated campaigns. Google has been expanding channel-level controls over time, so check what your account currently offers before assuming you have none.

Acting on it: Performance Max negative keywords in 2026

Triage produces a list. Negatives are how you act on it, and this is where 2025 changed everything.

The capacity problem is gone. When campaign-level negative keywords first reached PMax, the cap was 100 per campaign - a toy limit that forced brutal rationing. In March 2025, Google announced the cap was rising to 10,000 negative keywords per PMax campaign. That is not an incremental tweak; it moved PMax negatives from “emergency brake” to “real list management”, the same discipline you run on Search.

Three places to apply negatives, as of mid-2026:

  1. Campaign-level negatives. Inside the PMax campaign, open the campaign’s settings / negative keyword section (or add directly from the search terms report by selecting terms). Use these for campaign-specific junk: terms that are wrong for this campaign but fine elsewhere.
  2. Account-level negatives. In your account settings, the account-level negative keyword list applies across campaigns, PMax included. Use it for the never-ever terms: job seekers, adult content adjacency, competitors you refuse to serve on.
  3. Shared negative keyword lists. As of August 7, 2025, Google completed the rollout of shared negative lists for Performance Max - the same shared-library lists you attach to Search campaigns can now be attached to PMax. This is the structural win: maintain one master junk list, apply it everywhere, and a term you learned was garbage in Search protects your PMax spend on day one.

And one distinction worth keeping straight: brand exclusions and negative keywords are different tools. Brand exclusions block serving on searches for a brand (yours or a competitor’s, in all their variant forms, managed as brand lists); negative keywords block the specific terms you list. For brand traffic, exclusion lists are usually the cleaner instrument; negatives are your scalpel for everything else.

A sustainable cadence beats a heroic cleanup: fifteen minutes weekly on the search terms report, sorted by cost, zero-conversion rows reviewed, obvious junk negated, patterns promoted to the shared list. Accounts that do this quietly compound the benefit; accounts that did one big purge in 2025 and stopped are back where they started.

How Keyword Ninja automates the triage

Everything above is doable by hand, and we just showed you how. What our Performance Max suite sells is not visibility - Google gives you that for free now - it is the triage, run continuously, with verdicts attached. Concretely:

  • A waste meter with receipts. One recoverable-spend number per account, built from zero-conversion term cost, brand overlap, and spend held back by limited assets - and every component is labeled estimated when it is estimated. Where Google exposes real term cost we use it; where only category-level data exists we say so.
  • A black-box meter. The share of your PMax spend that no visible search term accounts for - the honest measure of how much Google still is not showing you, computed from your own numbers instead of an industry average.
  • Brand cannibalization, labeled - never silently filtered. We split PMax spend and conversions into brand, non-brand, and hidden, so you can see what share of “PMax performance” was your own name. The verdict stays yours: one click adds your brand terms as PMax negatives if you decide to shield them, but we relabel and show - we never quietly exclude traffic on your behalf.
  • Junk-placement flags with impressions-only honesty. Pattern detection for mobile-game apps, the anonymous placeholder, kids’ content, and MFA-style domains - flagged by pattern and impression volume, excludable account-wide in one action, and never dressed up with invented dollar figures, because the placement data contains none.
  • Channel outliers and campaign grades. A channel-mix view that flags when a channel’s cost per result is a multiple of your Search cost, and an A-F grade per campaign with the components shown - so “this campaign is a C” comes with the arithmetic, not vibes.
  • One-click negatives up to the cap. Flagged terms become negatives from the same screen - individually, or all at once from the recover-this-spend view - using the post-2025 capacity properly.

What it cannot do, stated plainly: it cannot reveal queries Google hides (nothing can - we measure the size of that black box instead of pretending to open it); it cannot put a real dollar figure on placement waste (the data is impressions-only for everyone); and its estimates are marked as estimates, because a savings number you cannot audit is marketing, not analysis. PMax triage is part of the same wasted-spend engine that covers your Search campaigns (how that works), and if you are weighing the whole tool category, our honest comparison of Google Ads management software includes the competitors.

FAQ

Can I see Performance Max search terms now?

Yes. As of mid-2026, Google Ads has a native search terms report for Performance Max: open the campaign, then Insights & reports → Search terms. It rolled out during 2025. Guides claiming PMax search terms are hidden predate the rollout.

Why is my PMax search terms report empty or tiny?

Three usual causes. First, low volume: new campaigns and small accounts may not have enough query traffic to clear Google’s privacy thresholds, below which terms are aggregated rather than shown. Second, the report only covers search-shaped traffic - if your PMax spend skews to YouTube/Display/Discover, there are simply few queries to show. Third, check your date range; the per-term report only has history to show from after the 2025 rollout reached your account.

How many negative keywords does Performance Max allow?

10,000 per campaign, since Google raised the cap from 100 in March 2025. On top of that you can apply account-level negative keywords and - since August 7, 2025 - shared negative keyword lists to PMax campaigns.

Does Performance Max cannibalize brand searches?

It can, and in accounts with meaningful brand demand it often does by default: PMax optimizes toward the cheapest conversions available, and people already searching your name are exactly that. Check by filtering the PMax search terms report for your brand and comparing cost per conversion against your brand Search campaign. We cover measurement and fixes in depth in our PMax brand cannibalization guide.

Can I see which channel a search term served on?

No. As of mid-2026 the search terms report and the channel-level performance report are separate views - you can see what people searched, and how spend distributes across channels, but not a per-term channel split.


The fastest way to know whether any of this is costing you money is to look at your own numbers, not anyone’s benchmarks. Our free Google Ads audit connects read-only, covers Performance Max campaigns alongside Search, and returns the specific terms, placements, and brand-overlap figures for your account - and the Performance Max suite keeps the triage running after the audit. The report Google gave everyone in 2025 is the starting line. The verdicts are the race.

Find your wasted spend in minutes

Connect Google Ads and Keyword Ninja shows you exactly what to cut.

Start free