Search Query Report Analysis

9 min read

Learn

Quick Reading

Estimated 9 mins

Prereq

Intermediate

Basic ML concepts helpful

Interactive

Static Playbook

Static guide & reference tables

The Gap Between Keywords and Queries

A keyword is what you bid on; a search query is what the user actually typed. Because of broad match, phrase match expansion, and Google's close-variant matching (which now includes synonyms and related searches even on 'exact' match), the two can diverge significantly. The Search Query Report (technically the Search Terms report) is the raw ledger of what triggered your ads, and it's the primary tool for finding both negative keyword candidates and new keyword opportunities.

Reading the Report Correctly

The report surfaces query, matched keyword, match type, impressions, clicks, cost, and conversions. The analysis pattern is consistent: sort by cost descending, and for every query above a spend threshold with zero conversions, decide whether it's irrelevant (negate it) or relevant-but-underperforming (needs a landing page or ad copy fix, not a negative). Queries with strong conversion performance but no dedicated keyword are 'harvest' candidates — promote them to their own keyword with tighter match type and bid control.

sqr_negative_candidates.sql

sql

-- Identify negative keyword candidates from Google Ads data exported
-- to BigQuery via the Google Ads data transfer service
SELECT
  search_term,
  matched_keyword,
  match_type,
  SUM(cost_micros) / 1e6 AS cost,
  SUM(conversions) AS conversions,
  SUM(clicks) AS clicks
FROM `project.google_ads.search_term_view`
WHERE segments_date BETWEEN '2026-09-01' AND '2026-09-30'
GROUP BY search_term, matched_keyword, match_type
HAVING cost > 25 AND conversions = 0
ORDER BY cost DESC
LIMIT 100

Broad Match Needs the Tightest SQR Discipline

Broad match keywords rely entirely on Smart Bidding signals and query-level relevance rather than your explicit targeting. They can deliver excellent incremental reach, but only if you're disciplined about reviewing the search query report weekly — broad match without active negative management is how accounts bleed budget on irrelevant queries fastest.

N-Gram Analysis for Patterns

Rather than reviewing queries one at a time, break them into single words and word pairs (n-grams) and aggregate cost/conversions across all queries containing that term. This surfaces systemic issues a line-by-line review misses — e.g. discovering that every query containing the word 'free' has zero conversions across hundreds of distinct search terms, which is a single negative keyword insight hiding inside thousands of rows.

Match Type Interaction Effects

A search query report row shows which keyword and match type actually served, but doesn't show what *would* have happened under a different match type setup. When analyzing, always check for keyword cannibalization — multiple match types or ad groups competing for the same query internally, which shows up as Google's auction choosing one over another based on Ad Rank, not your intended hierarchy. Reviewing the SQR alongside the Auction Insights report helps separate 'losing to a competitor' from 'losing to yourself.'

Low-Volume Noise Looks Like Signal

A query with 1 click and 0 conversions is not evidence of anything — it's statistical noise. Set a minimum spend or click threshold (commonly 2x your average CPA, or a fixed $20-50) before treating a zero-conversion query as a negation candidate, or you'll negate your way into irrelevance on pure variance.

What's Next?

Search query analysis feeds directly into negative keyword strategy and keyword expansion. From here, building a systematic, recurring negative keyword workflow — rather than ad-hoc reviews — is the natural extension.

I build these systems professionally.

Whether it's a RAG pipeline, analytics migration, or AI workflow — let's talk.

Work With Me →