Why Your Google Ads Search Terms Report Hides Real Queries

The Google Ads search terms report does not provide a complete record of the searches triggering your ads. Privacy thresholds hide some low-volume queries, AI Max expands matching beyond explicitly targeted keywords, and Performance Max (PMax) reporting covers only part of its multichannel activity. Advertisers need to combine query-level reporting, search terms insights, and n-gram analysis to identify wasted spend and make informed negative keyword decisions despite incomplete visibility.

Why Do Search Terms Insights and the Search Terms Report Disagree?

Search terms insights and the Google Ads search terms report measure search activity differently. Search terms insights groups queries into themes and subthemes using a rolling 56-day labeling window, while the search terms report provides individual query rows. Google also processes conversions differently between the two views, so their totals and reported performance may not match.

The search terms report is where advertisers go to see which searches triggered their ads. It shows individual search terms alongside impressions, clicks, cost, and conversions, making it a primary tool for finding irrelevant traffic and building negative keyword lists. Search terms insights takes a broader view, grouping search activity into categories and subcategories to show the types of demand driving campaign performance.

The problem starts when advertisers use both reports to evaluate the same traffic. A search category can appear in insights even after its associated queries have been excluded, and conversion totals may not match what appears in the search terms report. Without understanding how Google builds each view, it's easy to mistake a reporting difference for a targeting problem or make unnecessary changes to a campaign.

The 56-Day Labeling Window Creates a Lag

Search terms insights is designed to show patterns in search demand, not provide a current list of queries triggering ads. Google generates its category and subcategory labels using search activity from the previous 56 days.

Say a campaign has been attracting searches for free accounting software. The advertiser adds "free" as a campaign-level negative keyword, but a week later, search terms insights still shows a category associated with free accounting tools.

The immediate assumption might be that the negative isn't working. But the category label can still reflect searches from before the exclusion was applied. Google's documentation confirms it: the labels for these categories and subcategories (it also calls them themes and subthemes) may contain terms that are no longer serving or are excluded. Adding more negatives based on that label alone risks blocking valuable traffic without fixing an actual targeting problem.

To verify the exclusion, open the campaign's search terms report and set the date range to begin the day after the negative was added. Filter for searches containing "free" and check whether those queries are still generating impressions or cost. Then check the negative keyword list and change history to confirm the exclusion was applied to the correct campaign.

If the category remains visible but the excluded searches no longer appear in the report, don't assume the negative failed. Privacy restrictions mean the report cannot prove that every unwanted query has stopped, but the category label alone is not a reason to add more exclusions.

Conversion Totals Can Differ Even When the Date Ranges Match

Google explicitly warns that conversions are processed differently in search terms insights and the search terms report. This means that two people can pull conversion data from the same account, select the same dates, and report different results.

This becomes a problem when search terms insights is used to justify budget changes while the search terms report is used to evaluate the underlying queries. A category may appear to generate more conversions than the visible query rows account for. Some of that difference comes from search activity that does not appear as individual rows, and conversion-processing differences can add to it. Google describes the differences between the two views as slight and ties them to conversion lag, so a large gap is a reason to recheck your date range and filters first.

The wrong response is to assume the missing conversions belong to a particular hidden query or that one report is necessarily inaccurate.

For performance reviews, establish which reporting surface supplies the metric before comparing results. Use campaign-level reporting as the common reference for spend and conversion performance, and account for conversion lag when evaluating recent dates. Do not combine category-level conversions and individual search-term conversions as though they were interchangeable.

Which Report Should You Use?

Use search terms insights to decide where to investigate. Use the search terms report to decide what to exclude.

The two reporting surfaces answer different operational questions.

For example, if search terms insights shows a growing category around informational searches, investigate whether that traffic fits the campaign's acquisition objective. Then move to the search terms report, filter the relevant campaigns and date range, and review the visible queries for recurring modifiers associated with poor intent.

Build negative keyword candidates from those actual query rows, checking match type and potential impact on valuable searches before applying exclusions.

For larger accounts, this should be a repeatable workflow: identify categories worth investigating, examine the available query evidence, validate exclusion candidates, and monitor traffic after implementation.

Neither reporting surface provides a complete record of search behavior.

How Does Privacy Thresholding Hide Real Queries From Your Report?

Google Ads omits some low-volume search terms from the search terms report to protect user privacy. Long-tail queries, unusual phrasing, and searches that occur infrequently are more likely to be excluded from individual reporting. Those missing queries can still generate impressions, clicks, and cost, leaving advertisers with an incomplete picture of the traffic their campaigns attract.

Google limits individual search-term reporting based on privacy thresholds and does not publish the minimum. Ginny Marvin, Google's Ads Liaison, said in July 2025 that search terms are only reported when they meet sufficient search volume across all Google searches. As a result, advertisers cannot reliably predict which low-volume searches will be included in the report.

In practice, the reporting gap is most likely to affect longer, less common searches. A broad query such as "accounting software" may generate enough activity to appear consistently, while a highly specific search describing an unusual use case may never appear as an individual row.

This creates a problem for negative keyword mining. Broad, frequently searched terms are generally easier to inspect, but irrelevant traffic often arrives through specific combinations of words. A single low-volume query may not justify an exclusion. Hundreds of different low-volume queries expressing the same unwanted intent can collectively consume meaningful budget without providing enough individual query visibility to identify the pattern.

Broad match and automated targeting can make the investigation harder because the campaign can match against variations that advertisers never explicitly added as keywords. The search terms report shows the eligible queries Google makes available, not every search that contributed to campaign delivery. Search terms insights does account for them. Google says its search categories take all search terms into account, including those not exposed in the search terms report for privacy reasons.

How to Estimate the Missing Search-Term Volume

The reporting gap can be estimated by comparing campaign-level performance with the totals from individually reported search terms over the same period.

In Google Ads, open the relevant Search campaign, select a completed reporting period, and record its total impressions, clicks, and cost. Then open the search terms report for the same campaign and dates and export the individual query rows.

Before calculating the difference, check the export. Google Ads reports can include aggregate or summary rows that will inflate the totals if they're added to the individual queries. Remove those rows, confirm that no filters or export limits have excluded data, and make sure the campaign, date range, and network scope match the campaign-level figures. If the account uses Search partners, check whether that traffic is included consistently in both datasets.

For larger exports, use a pivot table in Excel or Google Sheets to sum impressions, clicks, and cost by campaign. Compare those totals against the corresponding campaign-level metrics. If the numbers don't reconcile, check the reporting scope and export completeness before treating the difference as missing query data.

The remaining difference estimates how much campaign activity cannot be accounted for through individually reported search terms. It does not establish how much of that activity Google withheld specifically for privacy reasons.

Consider a hypothetical Search campaign with the following results over 30 days:

In this hypothetical example, $12,000 of campaign spend is not represented by the exported search-term rows, or 15% of total spend.

That 15% is a reporting-coverage gap, not proof that every missing dollar was suppressed specifically by privacy thresholding. Differences in reporting scope, network coverage, and other reporting limitations must be checked before attributing the entire gap to privacy restrictions.

The advertiser cannot inspect the individual queries behind that portion of campaign spending. That limits how confidently the team can assess query relevance, identify waste, or claim that its negative keyword coverage is comprehensive.

Run this comparison consistently by campaign and reporting period. A growing gap deserves investigation, particularly when broad match or automated targeting is expanding traffic into new areas. A stable gap is still a visibility limit. It tells you nothing about the quality of the hidden traffic.

Why Does Performance Max Still Have Search-Term Reporting Blind Spots?

Performance Max now provides individual search-term reporting alongside broader search terms insights. Advertisers can inspect reported queries rather than relying exclusively on grouped categories, but privacy restrictions still limit query visibility. Because Performance Max also serves ads beyond Search, its search-term data cannot explain every impression, click, or dollar spent across the campaign.

Google Ads provides two ways to investigate search behavior in Performance Max.

The first is search terms insights, available through the campaign's Insights section. This view groups search activity into categories and subcategories, helping advertisers identify the types of searches associated with campaign performance.

For example, a PMax campaign promoting accounting software might show categories related to small business accounting, bookkeeping, and tax preparation. Those categories show where Google is finding demand, not whether every search inside one is relevant or profitable.

The second is the Performance Max search terms report, available through the campaign's reporting interface. Google provides individual reported search terms, along with supporting information such as landing pages and ad formats.

This gives advertisers a more useful starting point for identifying irrelevant searches than category-level insights alone. However, the same privacy restrictions that limit Search campaign reporting also apply to individual search-term visibility.

Neither view provides a complete account of everything Performance Max is doing. Search-term reporting covers search activity, while Performance Max can also serve ads across YouTube, Display, Discover, Gmail, and Maps. Search queries cannot explain performance from placements that do not originate from a conventional text search.

How PMax Reporting Compares With Search Campaign Reporting

Search and Performance Max both report individual search terms and category insights, but Performance Max also serves on inventory that search terms cannot explain.

What This Means for PMax Negative Keyword Coverage

Individual search-term reporting makes negative keyword mining more practical in PMax, but it does not eliminate the underlying visibility problem. Start with the PMax search terms report and look for recurring irrelevant queries. Review the associated landing pages and available performance data before adding exclusions.

A query that appears irrelevant in isolation may still reflect useful intent when paired with the product or service being advertised.

Next, review search terms insights for broader categories that deserve investigation. If a category suggests traffic outside the campaign's target market, inspect the individual queries Google makes available. Avoid excluding entire categories based solely on their labels.

Google Ads supports negative keywords for Performance Max, including campaign-level exclusions. Advertisers can use those controls to block known irrelevant search intent, but exclusions cannot recover queries omitted from reporting or explain performance across non-search inventory. Account-level negative keywords also apply to Performance Max campaigns. Both apply to Search and Shopping inventory only, so ads shown on Display and video placements need a different control: excluded content keywords.

After adding negatives, monitor search traffic, conversion volume, and campaign performance. An exclusion can reduce irrelevant traffic, but an overly broad negative can also remove valuable searches that Google's automated targeting previously captured.

For PMax campaigns with substantial spend, the practical objective is to improve coverage of identifiable irrelevant intent while documenting the portion of search behavior that remains unobservable.

How Does AI Max Expand Search Matching Beyond Your Keywords?

AI Max can match searches beyond the keywords advertisers explicitly target, including exact match keywords. That makes keyword-level performance less reliable as a measure of the search intent a campaign is actually capturing. An analysis by Smarter Ecommerce (smec) found that approximately 29% of exact-match-attributed impressions in its studied EMEA ecommerce campaigns with AI Max enabled involved AI Max expansion by July 2026.

An exact match keyword can appear to be performing well in Google Ads while AI Max brings in traffic from searches the advertiser never explicitly targeted. Strong keyword-level conversion rates or return on ad spend (ROAS) can therefore mask differences in intent between the original keyword and the expanded searches.

Consider a software company bidding on the exact match keyword [enterprise accounting software]. The advertiser wants demo requests from finance teams evaluating enterprise platforms. With AI Max enabled, Google might also match searches such as "accounting software for small businesses" or "how to automate bookkeeping." Those searches are related to the original keyword, but they may attract smaller companies, students, or people looking for information rather than enterprise software buyers.

If the expanded searches generate inexpensive demo requests, the keyword's overall cost per conversion could improve even as the percentage of qualified leads declines. Without separating expanded matches from the original targeting, the advertiser could increase budget based on what appears to be stronger performance while acquiring fewer viable sales opportunities.

The scale of AI Max expansion makes this worth investigating. An analysis by Mike Ryan at Smarter Ecommerce (smec) examined 383 million impressions across EMEA ecommerce Search campaigns between January 2025 and July 2026. By the end of the study period, approximately 29% of exact-match-attributed impressions in the studied AI Max-enabled campaigns involved AI Max expansion, up from nearly zero at the beginning. Ryan's LinkedIn post, published July 24, 2026, describes the result as over 25% of exact match keyword impressions; the 29% figure, the EMEA ecommerce scope, and the date range come from PPC Land's coverage of the chart and comments. The dataset covers ecommerce Search campaigns, so the 29% is not a benchmark for lead generation accounts like the one in this example.

Expanded matching is not inherently a problem. It becomes a problem when advertisers evaluate keyword performance without examining the additional search intent Google is capturing.

To investigate, open the Search campaign's search terms report in Google Ads and use the AI Max reporting controls, including the Source column and AI Max match-type reporting, to identify searches attributed to expansion rather than conventional keyword matching. Review those queries alongside cost, conversions, and conversion value to see whether expanded traffic is attracting the same type of buyer as the original keywords. Google recommends waiting at least two weeks after enabling AI Max before making changes such as adding negative keywords. Filtering the match type to AI Max does not count the Other search terms row, so it can understate AI Max traffic. The Keywords report adds two total rows at the bottom, AI Max expanded matches and AI Max landing page matches, which give aggregate totals for each source.

For the accounting software example, the next step is to compare demo requests from expanded searches against qualified opportunities in the customer relationship management (CRM) system. If broader research queries are generating leads that sales consistently rejects, a low cost per conversion is not evidence of efficient acquisition. Before reducing bids or adding exclusions, check whether the poor-quality leads are concentrated in specific search patterns or spread across the expanded traffic.

Even with those reporting controls, privacy restrictions prevent some low-volume queries from appearing individually. Advertisers can evaluate the expanded searches Google discloses, but cannot assume the visible rows represent every query AI Max matched. That limits how confidently keyword-level performance can be used to justify additional budget without checking downstream conversion quality.

How Should You Re-Scope Negative Keyword Mining Given All This?

Negative keyword mining should focus on recurring patterns of irrelevant intent rather than treating the search terms report as a complete record of user queries. N-gram analysis can expose patterns across visible search terms, while search terms insights help identify areas worth investigating. Advertisers should also measure reporting coverage and document the limits of query visibility before making exclusions or compliance claims.

Use N-Gram Analysis to Find Patterns Across Visible Queries

N-gram analysis groups recurring words and phrases across reported search terms, making it easier to identify patterns of wasted spend that individual query reviews can miss. It works with incomplete reporting, although it cannot recover queries Google has withheld.

Start by exporting the search terms report from Google Ads with search term, campaign, impressions, clicks, cost, and conversions. For lead generation accounts, include available conversion value or qualified-lead data where appropriate.

Break each query into one-, two-, and three-word combinations, then aggregate performance across every visible query containing each combination. This can be done in a spreadsheet, SQL, or a reporting script.

Consider a hypothetical B2B software company selling a paid enterprise accounting platform. The company wants demo requests from businesses evaluating paid software, but its search terms report shows several queries from people looking for free alternatives.

Looking at these searches individually, the wasted spend might not stand out. But an n-gram analysis groups queries containing the same word or phrase. In this case, grouping all searches containing "free" reveals $300 in spend across three different queries, with no conversions.

The pattern is more useful than any individual search term. The company sells paid enterprise software, yet Google is repeatedly matching its ads to people searching for free products. Adding "free" as a negative keyword could prevent similar searches from consuming additional budget, including variations that haven't appeared in the report.

Before applying that negative, however, the advertiser should check for valuable searches containing the same word. A query like "enterprise accounting software free trial" could represent a qualified buyer evaluating a paid platform. Blocking "free" broadly would exclude that traffic too.

The goal of n-gram analysis is to identify recurring patterns of irrelevant intent, then choose exclusions narrow enough to reduce wasted spend without blocking potential customers.

However, the analysis still covers only the search terms Google reports. N-grams make better use of available evidence; they do not bypass privacy thresholding.

For larger accounts, run the analysis separately by campaign, product line, or business objective. Combining unrelated campaigns can produce misleading recommendations when the same phrase represents valuable traffic for one product and irrelevant traffic for another.

Measure How Much Search Activity You Cannot Inspect

The difference between campaign-level metrics and the totals of individually reported search terms provides a practical measure of reporting coverage. The resulting gap identifies activity that cannot be evaluated at the individual query level, although it should not automatically be attributed entirely to privacy thresholding.

Once the reporting coverage gap has been calculated, track it consistently by campaign and reporting period. A growing gap deserves investigation, especially after changes to broad match targeting, AI Max, or campaign structure. Before drawing conclusions, verify that the reporting scope and export method have remained consistent.

Adthena, a Revvim partner and paid search intelligence vendor, reported in October 2024 that across the accounts it reviewed, an average of 51% of spend sat under "other" search terms. Treat that as one vendor's sample rather than a benchmark, since the share varies by account. That figure provides context for why reporting coverage deserves attention, but it should not be treated as a universal benchmark or attributed entirely to privacy thresholding. Individual accounts need to calculate their own coverage gaps.

Keep Search and Performance Max reporting separate. Performance Max includes inventory beyond conventional search activity, so comparing total PMax spend against individually reported search terms would overstate the query-reporting gap.

Tracking reporting coverage tells you how much confidence to place in query-level analysis. A larger gap means less visibility into the searches driving performance, not necessarily more wasted spend. Negative keyword decisions should still be based on the available query evidence and downstream conversion quality.

Document Reporting Limitations for Compliance Reviews

Search-term reporting should be described as a partial view of search activity, not a complete transcript of user queries. Compliance documentation should distinguish reported search terms, omitted low-volume queries, and category-level insights without implying that all underlying searches can be reconstructed.

For regulated advertisers, this becomes relevant when a legal, compliance, or audit team asks whether campaigns were exposed to prohibited or restricted search intent.

Google's documentation on ad group and asset group prioritization describes the search term shown in reporting for advanced search experiences such as Lens, AI Mode, AI Overviews, and autocomplete as the "best approximation of the user's intent." Separately, Google acknowledges that some low-activity queries are omitted from individual search-term reporting to protect user privacy.

A compliance note could use the following language:

Google Ads search-term reporting does not provide a complete record of every user query associated with ad delivery. Google describes the search term shown in reporting for searches on Lens, AI Mode, AI Overviews, or autocomplete as the "best approximation of the user's intent" and excludes some low-activity queries under its privacy standards. Performance Max provides individual search-term reporting alongside aggregated search category insights, but these reports do not cover every interaction across its advertising inventory. Negative keyword reviews and exclusions are therefore based on available reporting data and cannot establish that every underlying query has been individually inspected.

Document what the team reviewed, which exclusions were applied, and what the reporting system cannot verify. Keep dated exports, negative keyword change records, and approval notes so the review can be reproduced.

Negative Keyword Mining Checklist

Work through these seven checks for each campaign and reporting period to turn visible queries into defensible exclusions.

• Export individual search terms using a consistent campaign scope and reporting period.
• Build one-, two-, and three-word n-gram summaries to identify recurring irrelevant intent.
• Review search terms insights for additional categories that warrant investigation.
• Validate negative keyword candidates against valuable searches before applying exclusions.
• Separate Search and Performance Max reviews, including their available negative keyword controls.
• Calculate the share of campaign activity missing from individual search-term reporting.
• Record exclusions, reporting limitations, and subsequent performance changes.

Frequently Asked Questions

Does the Google Ads Search Terms Report Show Actual User Queries?

The Google Ads search terms report shows individual reported searches, but it is not a complete transcript of every user query. Google omits some low-activity searches for privacy reasons and may show an approximation of user intent for searches from Lens, AI Mode, AI Overviews, or autocomplete. Advertisers cannot assume every query that triggered an ad appears in the report.

Why Are PMax Search Terms Only Shown as Categories?

Performance Max search terms are no longer limited to categories. Google Ads provides individual search-term reporting for PMax alongside category-level insights that group related searches. However, privacy restrictions still prevent some queries from appearing individually, and search-term reporting cannot account for all activity across PMax's non-search advertising inventory.

How Do I Mine Negative Keywords When Query Data Is Approximated?

Export the available search terms and use n-gram analysis to identify recurring words and phrases associated with irrelevant traffic. Cross-check those patterns against search terms insights, then validate exclusions against valuable queries before applying negative keywords. Monitor performance afterward and account for the search activity missing from individual query reporting.

Ready to see if you’re overpaying for branded search?

Request a demo to start the risk-free, 30-day AdAi trial to quantify your trapped capital and reclaim budget to fund your growth campaigns.