The Google Ads search terms report shows you exactly what people typed before clicking your ad. It is one of the most actionable reports in the platform, and ignoring it is one of the most common reasons campaigns waste budget. If you are looking at irrelevant search terms draining your spend, the fix is straightforward: add them as negative keywords. But knowing which queries to exclude, which to promote to actual keywords, and which to leave alone requires a clear decision framework.

This article walks through every column in the report, explains how to sort and prioritize what you see, and gives you a repeatable decision rule you can apply to every query type.

What the Search Terms Report Actually Contains

The search query report in Google Ads is not the same as your keyword list. Your keywords are what you bid on. Search terms are what users actually searched for. These two things often diverge significantly, especially when you use broad match or phrase match keywords.

To access it, go to Campaigns > Insights and reports > Search terms. You can also find it at the ad group level if you want to narrow your analysis.

The columns that matter most

Google Ads shows a lot of columns by default, and even more when you customize the view. Here is what each important one tells you:

• Search term: The exact query the user typed. This is the core of the report.
• Match type: How your keyword matched the search term. This tells you whether a broad, phrase, or exact match keyword triggered the ad.
• Keyword: The specific keyword in your account that caused the ad to show.
• Impressions: How many times your ad was shown for that query.
• Clicks: How many users clicked through.
• Cost: Total spend attributed to that query.
• Conversions: How many tracked conversion actions resulted from clicks on that query.
• Cost per conversion: The average cost to acquire one conversion from that query.
• Conversion rate: The percentage of clicks that resulted in a conversion.
• Avg. CPC: What you paid per click on average for that query.

Some columns you should add manually if they are not visible: Search term match type, Added/Excluded (which shows whether the term already exists as a keyword or negative), and Conv. value if you track revenue.

The Added/Excluded column is underused

The Added/Excluded column tells you whether a search term has already been added to your keyword list or flagged as a negative. When a term shows as “None”, it means your account has taken no action on it. This column alone helps you prioritize: start with high-spend terms that still show “None”.

How to Sort the Report Before You Do Anything Else

The default sort order in the search terms report is usually by impressions or clicks. That is not the right starting point. Sort by cost, descending.

You want to see the search terms that have consumed the most budget at the top. High spend with low or zero conversions is where your losses are concentrated. These are the terms worth acting on first, not the ones with thousands of impressions but a few cents of spend.

Once you have sorted by cost, filter for a meaningful time window. Thirty days is a good default. Ninety days gives you more data on low-volume terms. Avoid going back more than six months unless you are doing a deep historical audit, because older data may not reflect your current campaign settings or audience behavior.

The goal of this analysis is not to review every query in the report. It is to identify the highest-impact decisions first, then work down from there.

The Three-Category Decision Rule

Every search term in your report falls into one of three categories. Once you know which category a query belongs to, the action is clear.

Category 1: Add as a keyword

A search term deserves to become its own keyword when it is performing well and you want more control over how you bid on it.

Specific signals to look for:

• The term has generated conversions at or below your target CPA
• It has a click-through rate significantly above your campaign average
• It is being triggered by a broad or phrase match keyword, meaning you have no direct bid control over it
• The query is highly relevant to a specific product, service, or intent

When you add it as a keyword, use exact match in most cases. This gives you the ability to set a dedicated bid, write a more tailored ad, and track performance independently. Phrase match is reasonable if you expect useful variations of the term.

Example: Your broad match keyword “project management software” is triggering the query “best project management software for construction teams”. If that query has converted twice at a strong CPA, it warrants its own exact match keyword so you can bid more aggressively on it and send users to a relevant landing page.

Category 2: Add as a negative keyword

This is where most advertisers underinvest their time, and where the most recoverable budget usually sits.

Add a search term as a negative when:

• It has meaningful spend but zero conversions
• The query intent is clearly different from what you offer (research, job seeking, competitor-branded, wrong industry)
• The cost per conversion is significantly above your acceptable threshold
• The term contains words that consistently underperform across multiple queries

When adding negatives, think about whether to apply them at the ad group level, campaign level, or to a shared negative keyword list. Ad group level is right when only one ad group is being affected. Campaign level makes sense when the irrelevant term could theoretically trigger ads across multiple ad groups. A shared list is the most efficient choice when you have recurring junk terms across multiple campaigns.

Pay attention to match type for negatives too. Exact match negatives block only that specific query. Broad match negatives block any query containing all those words in any order. Choose based on how broadly you want to exclude.

Example: You are running ads for a B2B accounting platform. The search term “free accounting software for students” has spent $140 with no conversions over 60 days. That should be excluded immediately. Beyond blocking the full phrase, you might also add “free” and “students” as broad match negatives if those words consistently appear in irrelevant queries across the campaign.

Category 3: Leave alone

Not every search term requires action. Some queries are too low in volume to draw conclusions from. A term with one click and no conversion over 30 days is statistically meaningless. Acting on it either way is premature.

Leave a term alone when:

• It has fewer than 10-15 clicks and no strong signal in either direction
• It is borderline relevant and conversion data is insufficient to judge
• It is already handled by an existing keyword or negative

The discipline to leave low-data terms alone is just as important as knowing when to act. Over-managing the account based on insufficient data leads to instability and can disrupt Google’s machine learning.

How to Build a Systematic Review Process

Reading the search terms report once during account setup is not enough. Queries shift as match types evolve, as competitors change bids, and as Google’s smart matching becomes more aggressive. A regular review cadence is essential.

For most accounts, a weekly review covering the past 7-14 days is the right default. High-spend accounts or campaigns in aggressive growth phases may benefit from a twice-weekly or even daily pass. Lower-spend accounts with stable campaigns can stretch to bi-weekly without much risk.

Each review session should follow the same steps:

1. Open the search terms report at the campaign or account level
2. Set the date range to the past 7 or 14 days
3. Sort by cost, descending
4. Filter for terms where Added/Excluded shows “None”
5. Apply the three-category decision rule to each meaningful term
6. Batch-add negatives and flag keywords to create

This workflow takes 20-30 minutes per week for a typical account and consistently returns measurable budget savings when done with discipline.

Common Patterns in Irrelevant Search Terms

After reviewing search terms across many accounts, certain patterns of irrelevant queries appear repeatedly. Recognizing them speeds up the review process.

Informational queries mixed into commercial campaigns

Users searching “how to” or “what is” are often in research mode, not buying mode. If your campaign is targeting bottom-of-funnel buyers, queries starting with these phrases often convert poorly. Adding “how to”, “what is”, “tutorial”, “guide”, and “meaning of” as broad match negatives is a common first step for commercial intent campaigns.

Job-related queries

Software and service advertisers frequently see queries like “[product type] jobs”, “[product type] career”, or “[product type] salary”. These are almost never buyers. Negative out “jobs”, “careers”, “salary”, “hiring”, and “resume” if they appear.

Competitor brand terms

Your broad or phrase match keywords can trigger ads on competitor brand searches. Whether to exclude these or keep them is a strategic decision. If your conversion rate on competitor queries is poor, exclude them. If you are running a deliberate conquest strategy, you may choose to keep them and bid separately.

Geographic mismatches

If you serve specific regions, watch for queries mentioning cities or countries outside your coverage area. These often slip through despite location targeting settings, especially with broad match.

How an AI Agent Applies This Same Logic Every Day

Manual search term reviews work well, but they depend on someone actually doing them consistently. The gap between reviews is where budget leaks. An AI agent that monitors the search terms report continuously and flags actionable queries as they accumulate closes that gap.

Adsroid Copilot applies this kind of analysis as part of its ongoing Google Ads optimization workflow. When it detects search terms that are spending without converting, or identifies high-performing queries that are not yet captured as dedicated keywords, it surfaces those as specific proposed actions inside the Adsroid dashboard.

The decision logic mirrors what an experienced media buyer would do: queries with meaningful spend and no conversions get flagged for negative keyword exclusion; high-performing queries triggered by broad or phrase match keywords get proposed as exact match additions. The difference is that this analysis runs continuously rather than once a week.

Depending on the automation mode you configure, the behavior changes:

• In Manual mode, Copilot surfaces recommendations for you to review at your own pace.
• In Copilot mode, Copilot proposes each specific action and waits for your approval before executing it. You can approve directly from the dashboard, by email, or through AI Chat.
• In Autopilot mode, approved action types execute automatically within the rules and thresholds you have configured, such as your Critical CPA or Critical CPC settings.

This is not about removing judgment from the process. It is about making sure the analysis happens every day and that the decisions that have clear answers get executed without delay.

What to Do After You’ve Cleaned Up the Report

Once you have completed an initial audit of the search terms report and added the most obvious negatives, the report becomes easier to manage on an ongoing basis. You are no longer wading through the same junk queries every week because they have already been excluded.

At that point, the review shifts toward opportunity identification rather than damage control. You are looking for emerging high-intent queries to promote to keywords, for new irrelevant terms that surface as Google expands match coverage, and for patterns that might inform campaign structure decisions.

For example, if you consistently see a cluster of queries around a specific use case or product variation, that might justify creating a dedicated ad group or even a separate campaign with its own budget and landing page. The search terms report is not just a cleanup tool. It is a source of genuine insight about how your audience describes their needs.

Frequently Asked Questions

How do I read the Google Ads search terms report?

Open the report under Campaigns > Insights and reports > Search terms. Sort by cost descending, add the Added/Excluded column to your view, and set a date range of 30 to 90 days. Focus first on terms with meaningful spend and no conversions, then look for high-performing terms not yet captured as dedicated keywords.

What should I do with irrelevant search terms in Google Ads?

Add them as negative keywords. Decide whether to apply the negative at the ad group level, campaign level, or to a shared list based on how broadly the term affects your account. Use exact match negatives for precise exclusions and broad match negatives when a word or phrase consistently signals irrelevance across many queries.

How is the search terms report different from my keyword list?

Your keyword list contains the terms you bid on. The search terms report shows the actual queries users typed before clicking your ad. These differ because broad and phrase match keywords can trigger a wide range of search queries beyond what you explicitly targeted.

How often should I review the search terms report?

Most accounts benefit from a weekly review covering the past 7 to 14 days. High-spend campaigns may warrant more frequent checks. The consistent habit matters more than the exact frequency.

Can I automate search term analysis in Google Ads?

Yes. Tools like Adsroid Copilot monitor the search terms report continuously and propose actions such as adding negative keywords or promoting high-performing queries to keywords. In Copilot mode, each proposed action requires your approval before it is executed. In Autopilot mode, supported actions execute automatically within your configured thresholds.