Performance Max now exposes the search terms that triggered your ads, and you can find this natively inside the Google Ads interface or pull it via the API. Treat it as an insights layer for judging conversion quality and adding negatives, not a route back to manual keyword bidding. You’ll find the report under Campaigns in your account, and the real work starts once you export it.
TL;DR:
- The new search terms report in Performance Max shows actual queries that triggered ads, but only for high-volume searches, with low-volume terms grouped for privacy.
- Account-level insights help identify emerging demand patterns across all campaigns, while campaign-specific views clarify performance for individual product lines or brands.
- Exporting data via UI, scripts, or the API allows ongoing analysis, but it requires careful segmentation and validation based on conversion value and margin before making exclusion decisions.
- Negative keywords should be applied carefully by checking landing pages and attribution windows, as the same query can perform differently depending on context and product availability.
- Search-term data is an insights layer, not a performance scorecard, and should be used to improve profitability through feed, asset, and exclusion adjustments rather than purely click volume metrics.
Table of Contents
- How to find the PMax search terms report in Google Ads
- What the report actually shows (and where it falls short)
- Extracting the data: UI exports, scripts, and the API
- A workflow for turning search terms into real decisions
- Before you block a term, run these checks
- Oxedent’s practitioner take on PMax search-term data
- Where search-term visibility actually fits
- How Oxedent can help you act on PMax search-term data
- Sources
- FAQ
How to find the PMax search terms report in Google Ads
You’ll find the report by going to Campaigns > Insights and reports > Search terms and landing pages for Performance Max. This surfaces the queries that triggered your ads alongside the landing page and ad format each one served, which is the context you need before touching anything.
There are two ways to view this data, and each suits a different job:
- Account-level insights show you demand patterns across all your PMax campaigns at once, useful for spotting emerging categories of search intent.
- Campaign-level views narrow the same data to a single campaign, which matters when you’re running separate PMax campaigns for different product lines or brands.
Once you’re inside the report, pull in columns for ad format, landing page, and date range as a minimum. Segmenting by date range matters more than it sounds. PMax’s algorithm shifts spend and query mix as it learns, so a snapshot from launch week tells you very little about steady-state performance. Google’s own guidance on the search terms report frames it as a discovery tool for understanding which queries drove conversions, not a bidding console, and that distinction should shape how you read every column in it.
What the report actually shows (and where it falls short)
The report exposes a defined set of fields, and knowing what’s missing matters as much as knowing what’s there.
- search_term: the actual query text, when Google has enough volume to disclose it.
- search_subcategory: a grouping label for related queries, useful for spotting patterns across near-duplicate searches.
- Clicks, impressions, and CTR: standard engagement metrics at the search-term level.
- Conversions and conversion value: the numbers that actually matter for any optimisation decision.
- search_volume ranges: banded estimates of query frequency rather than exact counts.
Cost data doesn’t sit in the standard UI report. You’ll need the API’s campaign_search_term_view for that, which returns spend alongside performance metrics at campaign level.
The privacy trade-off: low-volume queries get grouped into an “other search terms” bucket rather than shown individually, a deliberate design choice to protect user privacy.
One more gap catches people out: keyword-related segments elsewhere in Google Ads can exclude PMax data entirely, so a search terms pull from a Search campaign report won’t include your Performance Max activity. Historical PMax search-term data only goes back to March 2023, per the customer_search_term_insight documentation, so don’t expect anything earlier however far back you scroll.
Extracting the data: UI exports, scripts, and the API
For a quick manual check, exporting straight from the UI works fine.
- Open the search terms and landing pages report and set your date range.
- Add columns for search term, landing page, ad format, conversions, and conversion value.
- Export to CSV or push directly to Google Sheets for a one-off review.
Scripts and spreadsheet pulls still earn their place when you need a quick, ad hoc audit or you’re working in an account without API access set up. Their limitation is scale: they get unwieldy fast across dozens of campaigns.
For anything recurring, the Google Ads API is the better route. Query campaign_search_term_view when you need granular metrics and cost data together, or customer_search_term_insight when you’re after category-level demand signals instead. Avoid search_term_view, which is ad-group based and explicitly excludes Performance Max data, a mistake that quietly breaks a lot of automated reporting builds.
Pro Tip: If you’re mixing ad-group-level fields with campaign-level PMax resources in the same query, your export will come back incomplete. Keep PMax pulls in their own query using the correct SELECT fields.
Automate the extraction once you’re checking search terms weekly or more often. For a monthly audit, a manual export is still perfectly adequate.
A workflow for turning search terms into real decisions
Raw search-term data is only useful once you run it through a consistent process. Here’s the sequence that holds up across most eCommerce accounts:
- Segment first. Break exported terms down by intent (branded vs generic), product category, landing page, and ad format before you look at performance.
- Validate against profit, not clicks. Cross-reference each segment with conversion value and margin, not conversion count alone. A term with three conversions and a healthy average order value can outperform one with thirty low-margin sales.
- Choose the right action. A poorly converting term doesn’t always mean “add a negative.” It might mean the landing page is wrong, the feed title is misleading, or the product genuinely doesn’t match the query.
- Feed the language back. Recurring high-value phrases belong in product titles, landing page copy, asset headlines, and search themes, not just a negatives list.
Pro Tip: Before you negative anything, check whether the term also appears in your Performance Max campaign structure as a growth signal elsewhere. Killing a query in one campaign can sometimes just push the same traffic somewhere less efficient.
Negative keywords behave differently depending on inventory type. Account and campaign-level negative keyword lists apply to Search and Shopping inventory within PMax, but Display and Video placements need content-suitability exclusions instead, a distinction that trips up a lot of teams applying one blanket exclusion strategy across the whole campaign.
Before you block a term, run these checks
The same query can be a perfectly good performer in one context and a waste of spend in another. A generic search like “waterproof jacket” might convert well through a Shopping listing showing price and reviews upfront, but flop when it lands someone on a vague category page. Context, not the query text alone, decides whether a term deserves a negative.
Run through this before applying any exclusion:
- Check the landing page and ad format the term actually served, not just the query text.
- Confirm your conversion lag and attribution window aren’t hiding delayed conversions.
- Verify stock levels and price competitiveness for the product the query implies.
- Never negative off a single click spike; wait for a pattern across a sensible date range.
Search-term context matters enormously. The same visible query can carry a completely different impact depending on the landing page and ad format behind it, which is exactly why judging a term on click volume alone leads to bad decisions.
Attribution lag catches out more accounts than people expect. A query that looks like a zero-conversion waste this week might convert next week through a longer buying cycle, or convert through a different channel altogether. Rushing to exclude terms before checking attribution settings and product availability risks cutting off traffic that was never actually the problem.
Oxedent’s practitioner take on PMax search-term data
At Oxedent, search-term extraction is a standard part of every Performance Max audit, not an afterthought. You’ll typically see the data pulled straight into feed title reviews, asset copy checks, and exclusion lists in one pass, because the three tend to reveal the same gaps.
An insight only becomes a managed test or structural change once it’s validated against conversion value and margin, not clicks, and once the account has enough budget and history to make the test meaningful. For eCommerce brands wanting that review done properly, Oxedent’s free ads audit covers exactly this ground.
— Biplab
Where search-term visibility actually fits
Search-term reporting is an insights layer, not a scoreboard. It tells you why PMax spent money where it did; it doesn’t replace judging the campaign on conversion value and profitability. The teams who get the most from this data prioritise fixing what hurts margin, not chasing every click that didn’t convert.
How Oxedent can help you act on PMax search-term data
Reading a search terms report is one thing. Turning it into negatives, feed fixes, and asset updates that actually move margin is where most in-house teams run out of hours. Oxedent’s whole operation is built around eCommerce PPC specifically, which means Performance Max optimisation isn’t a bolt-on service squeezed between other channels; it’s the core of what the agency does every day.
An audit starts with a proper search-term extraction across your PMax campaigns, checks your negative-keyword coverage against your Search and Shopping inventory, and reviews feed and asset health for the gaps that quietly cap performance. PPC management with Oxedent runs from £350 per month, with no long-term contract tying you in. If you’d rather see where your account stands first, Oxedent’s free Google and Facebook Ads audit is the practical next step, and it costs nothing to find out what your search terms are actually telling you.
Sources
FAQ
What are search themes in PMax?
Search themes are signals you add to a Performance Max campaign to tell Google’s algorithm the kinds of queries you want to reach, distinct from the search terms report which shows queries that already triggered your ads. Think of themes as steering the campaign forward; search terms as auditing what happened after.
Does PMax have keywords?
Performance Max doesn’t use traditional keyword bidding, but it does expose the search terms that triggered your ads through the search terms report. You can also add negative keywords directly from that report to exclude unwanted queries.
What are good search terms to target?
A good search term is one that converts at a healthy value relative to your margin, not simply one that generates clicks or impressions. Segment terms by intent and product category, then judge them against conversion value before deciding whether to lean into or exclude them.
How do I see how often a term was searched?
The PMax search terms report shows banded search_volume ranges rather than exact search counts, a deliberate design choice tied to privacy protections on low-volume queries. For category-level demand patterns, the customer_search_term_insight API resource groups related searches with volume-range data attached.