Four query categories drain budget from almost every Google Search campaign running broad match without a negative keywords strategy. The same four. Regardless of vertical, budget size, or how long the account has been running.
The search terms report shows you which four they are. Most accounts haven't opened it this month.
The finding: Without a maintained negative keywords strategy, broad match campaigns spend actively on queries outside your customers' purchase intent. The wasted spend is recoverable. The signal corruption it creates in smart bidding is the deeper cost.
What the search terms audit shows
The pattern below comes from first-look audits on new accounts, anonymised to verticals (ecom/Apparel, DTC/Beauty, SaaS/B2B). The approach is the same each time: pull the search terms report for the most recent 90 days, sort by spend, and categorise every significant query against the campaign's stated purpose.
The same four categories appear in every account that has not run a google ads wasted spend audit recently - or has never run one at all.
Four wasted query types drain most accounts
Google's broad match is the default match type recommendation for standard Search campaigns in 2026. It reaches the widest spread of queries, including many that share semantic meaning with your keyword but carry entirely different intent.
A campaign targeting "accounting software" can trigger queries about accounting degrees, accounting certification, and accounting textbook comparisons. None of those are a buyer. All of them are spend.
| Query category | What it looks like | Why it costs you |
|---|---|---|
| Competitor brand | "[rival name] review", "[competitor] pricing" | Spend on buyers looking for someone else |
| Informational / research | "how does [product] work", "what is [category]" | Clicks from researchers, not buyers |
| Wrong product type | "[category] wholesale", "[keyword] DIY", "[category] hire" | Reaching a different buyer with different needs |
| Wrong intent stage | "[product] return policy", "[brand] refund", "[product] cancel" | Paying for the customer you already lost |
The informational category is typically the largest. Broad match interprets your keyword as a starting point and expands from there - matching on intent signals and semantic relationships that often reach well outside your actual customer. Google search ads best practices for broad match have always included pairing it with a negative keyword list. The expansion in broad match reach over the past two years makes that pairing non-optional rather than recommended.
How wasted queries corrupt smart bidding
This is where a negative keywords strategy stops being a housekeeping task and becomes a structural performance issue.
Smart bidding learns from the conversion events in your account. The algorithm identifies patterns across queries, audiences, devices, and times that lead to conversions, then adjusts bids toward those patterns. When low-intent queries generate low-quality traffic that occasionally converts on micro-events - a page view counted as a conversion, an accidental form fill, a return visit from a different source - the system learns toward those signals.
The result is an account where smart bidding is confidently optimizing in the wrong direction.
A google ads wasted spend audit that only checks CPA and ROAS will miss this entirely. Aggregate performance can appear acceptable while the actual purchase-intent queries represent a fraction of total spend. The deterioration is in the query mix, not the headline numbers - which is why it can run undetected for months.
This is why the smart bidding strategies guide treats negative keywords as a pre-condition rather than an optimisation layer. Data quality in, quality decisions out. Smart bidding is capped by the quality of the signal it receives.
BAVai scans search term patterns every morning across BAVS accounts - not because the negative keyword list changes daily, but because one week of unchecked broad match traffic can introduce new irrelevant query patterns that corrupt conversion signal before the next scheduled review.
Where a negative keywords strategy can go wrong
An account that excludes every non-exact-match phrasing has effectively moved to exact match without the data upside. Real buyers who phrase their query differently than expected get blocked. The algorithm loses volume, compresses delivery, and the campaign starts underperforming against its own history.
Google search ads best practices draw the distinction clearly: a negative keywords strategy should exclude poor-intent queries, not simulate exact match on a broad match campaign. The test for any candidate negative is intent - does this query represent someone who could buy? If yes, it stays in.
Performance Max handles query exclusions differently. Negative keywords on PMax work at the campaign and account level but interact with the asset group matching in ways standard Search campaigns do not. The same intent principles apply; the implementation requires a separate approach.
The Query Hygiene Stack
This is the Query Hygiene Stack - a three-layer negative keywords strategy that builds in sequence and maintains itself without constant manual intervention.
Layer 1: Seed list before launch.
Before any campaign spends, build a negative keyword list from the patterns you already know: competitor brand names, informational stems ("how to", "what is", "DIY", "free", "review", "tutorial", "training"), wrong-category qualifiers ("wholesale", "hire", "certification", "course"), and wrong-intent-stage signals ("refund", "return", "cancel", "complaint"). This blocks the most predictable wasted spend from day one and prevents the early conversion signal from being contaminated before the algorithm has established a baseline.
Layer 2: Campaign-level intent exclusions.
Layer intent-specific exclusions based on the campaign's purpose. A lower-funnel shopping campaign needs different exclusions than a brand awareness campaign. Wrong-intent-stage queries - "return policy", "cancel subscription", "how to cancel" - belong on every lower-funnel campaign without exception. Competitor brand terms belong on lower-funnel campaigns unless competitive conquesting is the explicit, separate goal with its own budget and creative.
Layer 3: Weekly search terms review.
The negative keyword list is never finished. Sort the search terms report by spend, scan for new irrelevant query patterns, update the list. Done consistently, this runs under ten minutes. The compounding effect is significant: a negative keywords strategy that stays current maintains the signal quality that smart bidding depends on to function as designed.
The google ads wasted spend audit that surfaces the most immediate wins is almost always the first honest review of the search terms report. The ad account health check run on every new BAVS account starts here - before campaign structure, before bid settings. What has this account been paying for?
A campaign where smart bidding runs on a clean query mix performs at a structurally different level than one where only the bidding is smart. The negative keyword list is what makes the signal trustworthy. Everything the algorithm learns from that point builds on a foundation that is actually sound.
Of every query in your current search terms report, how many would you have consciously bid on if you had seen them first?