Performance Max search terms, what you can (and can't) see
What query visibility Performance Max actually gives you, including search terms insights, search themes and negative keyword options, the gap vs Standard Shopping, and workarounds.
Performance Max shows you search terms insights, aggregated, categorised query data, rather than the full row-by-row search terms report Standard Shopping gives. You can now apply negative keywords, but you're still steering with less information than Standard Shopping provides. This guide covers exactly what PMax exposes, where the gaps are, the practical workarounds, and why your feed becomes the main steering wheel when you can't see every query.
What Performance Max actually shows you
Standard Shopping hands you a complete search terms report: every query that triggered a click (and most that triggered impressions), row by row, with its own metrics, raw material you can read, mine for negatives, and steal title ideas from.
Performance Max gives you search terms insights instead, and the difference is the point of this article. Under the campaign's Insights, PMax shows:
- Search categories: queries clustered into themes ("mens waterproof jackets", "hiking gear"), with impressions, clicks, conversions and conversion value per cluster.
- Top search terms within categories: the higher-volume individual queries inside each cluster.
- Trend data: which categories are growing or shrinking for your account.
What it withholds: the full row-level long tail. Low-volume queries, which in Shopping are collectively where much of the spend and much of the waste lives, get bucketed into aggregated rows or omitted below privacy thresholds. You can also pull a fuller (though still thresholded) picture per channel through the campaign's reporting and Report Editor, but nothing reaches the completeness Standard Shopping gives you by default.
On the control side, the position has genuinely improved: PMax now supports campaign-level negative keywords, account-level negatives apply, and brand exclusions fence branded queries in or out. So the old summary, "you can't see and you can't exclude", is out of date. The current truth is subtler: you can exclude, but you're aiming with partial sight.
The visibility gap vs Standard Shopping
| Standard Shopping | Performance Max | |
|---|---|---|
| Full row-level search terms | Yes | No, aggregated insights + top terms |
| Long-tail query visibility | Yes | Bucketed / thresholded |
| Negative keywords | Yes, campaign and list level | Yes, campaign level (plus account level) |
| Negative keyword lists | Yes | Limited, check current availability in your account |
| See which channel a query came from | All Search/Shopping | Partially, channel reporting is separate from query data |
| Query-level bid control | Via structure and priorities | No |
Why the gap exists: PMax is a cross-channel campaign, covering Search and Shopping, but also YouTube, Display, Gmail, Discover and Maps, and much of that inventory has no query at all. Google's design trades your visibility for its optimisation freedom. Whether that trade pays depends on the account, which is the Standard Shopping vs PMax decision proper. This article's concern is narrower: given you run PMax, how do you steer with what it shows?
Working the visibility you do have
1. Make search terms insights a scheduled habit, not a curiosity. Monthly at least. You're looking for two things: categories converting well that your feed under-serves (growth signal), and categories spending without converting (exclusion signal). The aggregation actually helps at this altitude: themes are what you act on anyway.
2. Build negatives top-down. Without the full long tail, precision negatives are hard, so work from the visible categories: irrelevant themes ("repair", "hire", "jobs", "free"), clashing intents (parts vs whole products), and anything your Standard Shopping search terms history already proved wasteful. Which leads to the next point.
3. Use a Standard Shopping campaign as your query telescope. The most effective workaround going: run Standard Shopping alongside or before PMax on the same products, and mine its complete search terms report for negatives and title insight that you then apply to PMax. Many merchants keep a Standard campaign running partly for this intelligence value alone.
4. Fence off branded traffic deliberately. Brand exclusions (or their absence) decide whether PMax's numbers are flattered by shoppers who searched your name and would have bought anyway. Deciding this explicitly makes every other number in the campaign more honest.
5. Read the category trends against your feed. A rising search category you barely appear in is a merchandising signal wearing an advertising costume. Often the fix is product data, not campaign settings. Which brings us to the real steering wheel.
Deep dive When you can't see queries, the feed is the targeting
Here's the reframe that makes PMax manageable: in Standard Shopping, query-level visibility let you steer after the match, seeing the query, judging it, excluding it, restructuring around it. PMax removes most of that post-hoc steering. What remains is steering before the match: controlling which queries your products are eligible to meet in the first place. And in Shopping, that control lives almost entirely in the product feed.
Google matches Shopping ads to queries by reading your product data, and there are no keywords to bid on in either campaign type. The title carries the most weight, then description, product type, Google product category, brand, GTIN and attributes. That was always true; PMax's opacity just raises the stakes, because feed-driven matching is now the only precise targeting you have:
- Titles decide the query space. A title of "Ghost 16" can match anything ghost-related; "Brooks Ghost 16 Men's Running Shoes Size 10.5 Grey" matches high-intent running-shoe queries and little else. In PMax you won't see most of the junk queries a vague title attracts, you'll just see mediocre category-level ROAS and not know why. Precise titles prevent, invisibly, what negatives would have cured visibly. See titles and query matching.
- Complete attributes narrow the guesswork. Missing GTINs, wrong categories and absent colour/size/gender attributes force Google to infer what the product is, and inferred matching is exactly the kind you can no longer audit through a search terms report.
- Listing groups are your remaining structural lever. You can't pick queries, but you can pick which products run and how they're grouped, and product data (product types, custom labels) defines the groups. See PMax listing groups and feed signals.
The blunt way to say it: in PMax, feed optimisation is query targeting. Merchants who ran thin feeds on Standard Shopping could at least see the damage in the search terms report and prune it. In PMax the same damage is silent: it shows up only as spend drifting to poorly-matched auctions inside an aggregate number. This is why feed enrichment work, the completeness and precision that /tools' Optimise module exists for, applied non-destructively in a supplemental layer over your source data, moves from "good practice" to "your primary steering input" the moment a catalogue enters PMax. A feed health score is, in effect, a proxy for how much of PMax's matching you're controlling versus leaving to inference.
The honest position
Performance Max's query visibility has improved from "nothing" to "aggregated insights plus real negatives", and it may improve further. Check what your account currently offers rather than trusting year-old blog posts, including this corner of anyone's docs. But the structural trade is permanent: PMax buys cross-channel reach with your query-level sight. Run the campaign accordingly: insights reviewed on schedule, negatives built from themes and from Standard Shopping intelligence, brand traffic fenced deliberately, and a feed precise enough to do the targeting your search terms report no longer can.
Frequently asked questions
Can you see search terms in Performance Max?
Partially. PMax provides "search terms insights", queries grouped into aggregated categories with impression, click and conversion data, rather than the complete row-level search terms report Standard Shopping offers. You see the big themes and top queries; the long tail is summarised or bucketed under low-volume "other".
Can I add negative keywords to Performance Max?
Yes. Campaign-level negative keywords are now available for PMax, and account-level negative keywords apply too. Brand exclusions handle branded traffic separately. The catch is precision, since you're often excluding queries you can only partially see, so build negatives from your insights data and from Standard Shopping learnings.
Why does Google hide search terms in Performance Max?
Google's stated reasons are privacy thresholds and the campaign's cross-channel design. PMax serves across Search, Shopping, YouTube, Display, Gmail and Maps, and much of its inventory isn't query-driven at all. Whatever the rationale, the practical effect is the same, less query-level steering, more reliance on inputs like the feed.
How do I control what queries my PMax Shopping ads show for?
Indirectly, through your product feed. Titles and product data largely decide which queries products match, listing groups decide which products run, negatives prune what you can see, and brand exclusions fence off branded traffic. In PMax, feed optimisation effectively is query targeting.