How to Filter Junk Keywords in Google Ads (and Stop Wasting Spend on Bad Searches)

Filtering junk keywords in Google Ads comes down to one habit: reading your search terms, deciding which ones don't deserve your money, and blocking them with negative keywords. By the end of this guide you'll have a repeatable routine for doing that, with clear rules for what counts as junk, which match type to use, and how to confirm the blocks worked.

Before you start, you need a Google Ads account with at least a few weeks of Search campaign data and permission to edit keywords. Without some click history, you'd be guessing.

Step 1: Pull Up the Search Terms Report with a Useful Date Range

Open your Google Ads account and go to the search terms view. In the current interface it sits under Insights and reports > Search terms, but Google moves menus around, so if you don't see it there, use the search icon in the left navigation and type "search terms."

Next, set the date range. For a busy account, the last 30 days is a reasonable starting point. For low-volume accounts, go longer, since a term with three clicks in a week tells you almost nothing. Avoid judging terms on one or two days of data, because a single bad day can make a good term look broken.

It helps to be precise about vocabulary here. A search term is the exact phrase a person typed into Google. A keyword is what you bid on. One keyword can trigger many different search terms, and the junk lives in the search terms (the difference between search terms and keywords is worth understanding well). You fix it by adding negative keywords, which tell Google not to show your ads for searches containing certain words or phrases. Deleting the positive keyword is usually the wrong fix, because the keyword itself may be fine.

One limitation to know about: Google doesn't show every query. The report includes only search terms that meet Google's thresholds for query volume and user privacy, so some spend will be attributed to terms you can't see. Check Google Ads Help for the current wording. In practice, it means this report is your best view of the problem, not a complete one.

Step 2: Define What Counts as Junk for Your Account

Junk is relative to what you sell. Suppose you run ads for a paid project management app. The word "free" or "jobs" in a search term is probably wasted spend. For a recruiting agency, "jobs" is the whole business. Same term, opposite verdicts.

Most junk falls into a handful of categories:

• Job seekers: "careers," "hiring," "salary."
• Free or DIY seekers: "free," "template," "how to make your own."
• Wrong product or service: you sell repairs, they want to buy parts.
• Wrong location: cities or regions you don't serve.
• Research-only queries: "what is," "definition," "history of."
• Competitor or brand terms you don't want to pay for.
• Unrelated words sharing a keyword: the same word meaning something else entirely.

Before you review anything, write a short "would I pay for this click?" checklist. Three or four questions is enough. For example: Could this person buy what I sell? Are they in a place I serve? Are they looking to buy, or just to learn? Would I be okay paying my average cost per click for this exact query?

A written checklist keeps decisions consistent from week to week, and from one team member or client account to the next. It also stops you from relitigating the same edge cases every time. If you want more examples of what to look for, this guide on how to find junk keywords in Google Ads goes deeper.

The mistake to avoid is judging a term only by how it looks. A weird-looking query that converted twice is more valuable than a sensible-looking one that burned through budget. Look at cost and conversion data before you decide.

Step 3: Sort and Segment the Data to Surface the Worst Offenders

Start with the biggest leaks. Sort the report by cost, highest first, and scan for terms with meaningful spend and zero conversions. Then sort by clicks to catch high-traffic terms that aren't paying off. Handle these before touching the long tail of one-click oddities, since that's where the money is.

Make sure your columns give you enough to decide. Useful ones include:

• Cost
• Clicks
• Conversions
• Conversion rate
• The match type column (shown as "Match type" or "Added/Excluded," depending on your view), so you can see how the term was matched to your keyword

Then apply filters. A common one is conversions equal to zero plus cost above a threshold you choose. Pick a number that makes sense for your average order value or cost per lead. A threshold that suits a $5,000 service contract will be far too high for a $20 product.

Filter by campaign or ad group as well, and review one chunk at a time. Reviewing 40 terms from a single ad group is far easier than scrolling through 900 across the account, and you'll make better calls because the context is the same.

As you scroll, watch for patterns rather than only individual queries. If "jobs," "salary," and "internship" show up across a dozen different terms, that's a single negative keyword for each word, not a dozen separate blocks. Patterns become your most efficient negatives, and they also catch queries that haven't shown up yet. Accounts with too many irrelevant keywords usually show these patterns quickly.

Keep a running note of repeated words as you go. You'll use that list in Step 5.

Step 4: Review Terms and Sort Them into Keep, Add, or Block

Every term you look at goes into one of three buckets:

1. Block: clearly irrelevant, so add it (or a pattern from it) as a negative keyword.
2. Add: relevant and converting, so add it as a positive keyword to control its bidding and ad copy directly.
3. Leave alone: not enough data to judge yet.

The third bucket matters most. A term with two clicks and no conversions isn't automatically junk. Set your own click threshold before you block on performance alone, for example waiting until a term has had enough clicks that you'd expect a conversion by now. That's a judgment call based on your conversion rate, not a Google rule. Terms that are obviously irrelevant, like a job-seeker query for a software product, don't need to wait.

For ambiguous terms, check intent before you decide. Search the phrase yourself and look at what comes up. If the results are all tutorials and forums, it's likely research traffic. If they're product pages and comparison sites, buyers may be searching it.

Here's an illustrative example for a hypothetical company selling paid invoicing software. These terms and numbers are made up:

• "free invoice template": 42 clicks, $63 cost, 0 conversions. Block. Free-seeker intent, plenty of data.
• "invoicing software jobs": 9 clicks, $14 cost, 0 conversions. Block. Job seeker, no need to wait.
• "invoicing software for freelancers": 18 clicks, $40 cost, 3 conversions. Add as a keyword.
• "best invoicing app small business": 6 clicks, $11 cost, 0 conversions. Leave alone. Good intent, too few clicks.
• "what is an invoice": 25 clicks, $31 cost, 0 conversions. Block. Informational only.
• "invoice maker app": 11 clicks, $22 cost, 1 conversion. Leave alone and keep watching.

Step 5: Choose the Right Negative Keyword Match Type and Level

Negative keywords come in three match types, and they behave differently from positive ones.

• Negative broad: blocks searches containing all of the negative's words, in any order. Searches can contain other words too.
• Negative phrase: blocks searches containing the words in the same order, with other words allowed before or after.
• Negative exact: blocks only that exact search, with no extra words.

The key quirk: negative keywords don't expand to close variants, synonyms, or plurals the way positive keywords can. Blocking "job" doesn't necessarily block "jobs." If you want both, add both. Confirm the current behavior in Google Ads Help, since matching rules get updated.

A workable rule of thumb:

• Use exact to block a single bad query without touching anything near it.
• Use phrase when a specific phrase is bad wherever it appears, like "how to make." There's more detail in this guide to phrase match negative keywords.
• Use broad only for single words you're certain you never want. Multi-word broad negatives can behave unexpectedly, and a careless one can quietly block good traffic.

Then choose the level. You can add negatives at the ad group, campaign, or account level through a shared negative keyword list. Pick the narrowest level that fits. A term that's junk in one campaign but valuable in another belongs at the campaign or ad group level. A word that's junk everywhere, like "free" for a paid product or "jobs," belongs in a shared list you can apply across campaigns. A ready-made negative keywords list for Google Ads can give you a head start on that shared list.

Finally, check for conflicts. A negative that matches your own positive keywords can stop them from serving at all. Before saving a negative, compare it against your active keywords in that campaign. Blocking "template" is fine until you realize you sell templates.

Step 6: Add the Negatives, Faster, Without Leaving the Report

The native route works fine. In the search terms report, select the terms you want to block and use the option to add them as negative keywords. Google then asks you to choose the level and match type. Menu labels change, so verify the current flow in your account. It's reliable, though it can get slow when you're working through hundreds of terms across multiple campaigns, and each selection needs its match type set by hand. If you want the full walkthrough, see this guide on how to add negative keywords in Google Ads.

This is where Keywordme fits. It's a Chrome extension that works inside the Search Terms Report, so you can mark junk terms, apply match types, and add negative or positive keywords with clicks, rather than exporting to a spreadsheet and re-uploading. Keywordme describes this as helping you optimize up to 10x faster. That's the company's own claim, so treat it as marketing, not a benchmark. As of this writing, it's $12 per month per user with a 7-day free trial; check the site for current pricing.

Keywordme acts on the selections you make. You decide which terms are junk using the checklist from Step 2, and the tool applies your choices. It doesn't change your campaigns on its own, and nothing about it guarantees savings. The judgment is still yours.

If you manage several accounts, a few habits pay off:

• Run the same checklist across every client account so decisions stay consistent.
• Keep a shared negative list per vertical (one for home services, one for SaaS, and so on) and apply it at onboarding.
• Still review each new account's search terms, since every business has its own exceptions.

Step 7: Set a Review Schedule and Check That the Filtering Works

Junk search terms keep arriving as people search in new ways, so one pass isn't enough. Match your cadence to spend. As a guideline, review weekly for high-spend or newly launched campaigns, and every two to four weeks for stable ones. That's a suggestion, not a rule, so adjust it as you see how fast new junk appears.

To confirm your negatives are working, recheck the search terms report a couple of weeks after adding them. The blocked terms should stop collecting impressions. Also watch the campaign's cost per conversion and conversion rate over the same period. Improvement there is the real test, though other changes (bids, seasonality, ad copy) can muddy the picture, so don't give the negatives all the credit.

If something looks off, troubleshoot in this order:

• Blocked terms still showing up: check the match type first. A negative exact won't stop a variation, and a missing plural is a common culprit. Then confirm the negative was applied at the right level and to the right campaign.
• Conversions dropped after blocking: you may have over-blocked. Look for a broad or phrase negative that catches good queries, and remove or narrow it. This piece on how many negative keywords you should have can help you judge whether your list has grown too aggressive.

Keep a simple change log: a sheet or doc listing the date, the negative keyword, its match type and level, and why you added it. When results shift, you can trace the cause quickly and reverse a mistake without guessing.

Run It on Your Biggest Campaign First

Start with your highest-spend campaign, since that's where junk costs the most. Work through the seven steps once, then put the next review on your calendar and repeat. The first pass is the longest. After that, you're mostly handling new terms and patterns.

If you want that routine to take less clicking, Keywordme was built for it. Optimize Google Ads Campaigns 10X Faster. Without Leaving Your Account. Keywordme lets you remove junk search terms, build high-intent keyword lists, and apply match types instantly, right inside Google Ads. No spreadsheets, no switching tabs, just quick, seamless optimization. Start your free 7-day trial (then just $12/month) and take your Google Ads game to the next level.