Your Google Ads Search Terms Report Is Where the Wasted Money Lives

Most people who manage Google Ads accounts know about the search terms report. Very few have a system for what to do with it.

That gap is expensive.

Accounts can be hitting their targets, but still bleed budget on searches that will never convert, tended by ad managers who open the search terms report every now and then, catch something really expensive or really irrelevant, add a few negative keywords, and close the tab. That is not a system.

What are search terms?

When someone performs a search on Google, their search is matched to your ad account's target keywords and, if your ad rank is high enough, Google will show them one of your ads. A search term is what the person actually typed into the search bar, which is distinct from the keyword you bid on. Your keyword is the targeting instruction you gave Google, and the search term is what triggered Google to serve your ad, based on its relevance to your keywords.

What are Match types?

In the early days of Google Ads (then called Google Adwords), you would tell Google that you wanted to serve ads for a keyword, and you would serve ads for that keyword. You would enter every variation and misspelling of that keyword you could think of, because Google would only serve your ad when someone's search term matched your target keyword.

In 2002, Google let you target Exact matches to your keyword, as well as when your exact keyword was used in a Phrase (i.e. Phrase match keywords). 2006 brought us the infamous Broad match keywords, which majorly expanded the types of searches that could be matched to your keywords. Google realized that a little control would increase the number of people using Broad match keywords, so they included a Broad match modified variant in 2010, which let advertisers add a "+" symbol to certain words in their Broad match keywords to indicate which words were most important to match closely.

In 2014, Google finally started expanding to "close variants" and plural versions of keywords, which saved advertisers from needing to add dozens of keyword variants, but it also began the slippery slope of increasing the tolerance advertisers had to less relevant keyword matching on Exact or Phrase match keywords.

The match types, as entered into Google Ads, look like this:

• Broad match
• Broad match +modified (deprecated in 2021)
• "Phrase match"
• [Exact match]

Modern keyword match types are much more liberal with their definition of "relevant". This expansion of matching coverage was implemented to account for voice search and other changes to how people actually search, but also to increase Google's quarterly numbers (as revealed in their anti-trust hearings).

Thus, Broad, Phrase, and even Exact match keywords are designed to expand the types of search terms to which your keywords can be matched. The idea is that they match, not based on semantic relevance, but on relevance of intent. Google serves your ad to searches it believes are related to your keyword's intent. Sometimes it is right. Often it is not. The search terms report is where you get to judge which is which.

The report is available for every Search campaign, which includes Performance Max campaigns (a multi-purpose campaign type that includes matching to search terms based on audience signals and website content, rather than explicitly chosen keywords). To see the search terms report, go to Insights and reports, then Search terms, set your date range to the last 7, 14, or 30 days (or more, in some cases) depending on your account volume, sort by cost (descending), and start at the top.

What are "other search terms"?

Before you get too far into the report, you should know that you can't see every search term.

Google began obscuring search term data in 2020, citing user privacy as the reason (but really, they just want to hide data so we can't block more searches and they make more money). What this means in practice is that a significant portion of the search terms that triggered your ads simply appear as an aggregated totals line at the bottom of the report, labeled "Other search terms". In most accounts, roughly 30% of search terms are hidden. In accounts with lower volume or narrowly themed campaigns, I've seen that number reach 80%. In some cases, the hidden terms include all of the converting searches.

So while it can seem frustrating and overwhelming that you're making optimization decisions based on incomplete data, that's not a reason to skip the review. I just don't want you to be surprised if you go to check which search term converted and you can't find it. Or if you see 15 conversions in the keywords report, but only 11 (or 6) in the search terms report.

What are negative keywords?

Negative keywords are the tool Google gives you to block their algorithm from matching your keywords to irrelevant or low-quality search terms.

An example would be if you were targeting "microsoft copilot governance", and you see "microsoft 365" and "microsoft copilot tutorial youtube" in the search terms report.

You could add negative keywords to prevent Google from serving to these low-quality and irrelevant terms, preventing the ad platform from wasting your budget on traffic for these searches.

However, just as target keywords have match types, negative keywords have match types as well. As broad as keyword match types are, those match types were designed to increase revenue (and also to genuinely help advertisers serve ads to more relevant searches without needing a billion keywords per ad group). Negative match types are designed with the same end goal, blocking the least number of search terms possible.

So, as you could imagine, Exact match negative keywords block the exact text you add in square brackets, phrase match negatives block searches that include the text you enter into quotation marks, and broad match negatives block searches that contain all of the words in your negative keyword (without adding symbols).

How important are negative keywords?

Adding negative keywords is an essential part of search terms management, which is an essential part of Google Ads management.

Aside from preventing your keywords from being matched to irrelevant or low-quality search terms, you are reducing the total number of impressions for which your ads can serve, while still serving them for relevant searches. This increases your click-through rate (CTR), which sends strong signals to Google that your ads are relevant to the searches to which you're being matched.

When Google sees that serving your ads makes them more money, they reward you with a higher Quality Score (we'll cover that in another article) and that generally leads to lower cost-per-click prices (CPCs). Lower CPCs means you get more clicks for the same total budget, and because those clicks are more relevant to your target keywords, and because you're smart and chose to send the ads to a relevant landing page, you should see more conversions for a lower cost per action (CPA).

So when you see all of the ways adding negative keywords can improve your account, it is astounding to learn that a lot of people don't add many negatives, or even skip search terms management completely.

I have reviewed accounts spending six figures a month where the previous ad manager and the PPC Director above them (at a recognized agency) had added a combined three negative keywords over six months (hundreds of thousands of search terms, 3 negative keywords). All three were exact match negatives.

This is not a fluke, either. I have seen it from several big-name agencies with hundreds of clients and from several freelancers. The search terms report gets opened, something bad enough catches the eye, and then the tab gets closed. The routine review never gets built.

Can I use AI to review my search terms?

Yes! And no! This is a common trend with most early adopters in the advertising world. They upload all of their search terms into Claude, ChatGPT, or Gemini, and ask it to tell them what negative keywords to add. Some more savvy ad managers will ask AI to categorize the search terms, then grab the irrelevant or low-quality search terms from the more obvious category lists.

The problem with using AI to manage your search terms is that AI doesn't often understand the intent behind the search terms, nor does it usually have the context needed to determine what is and isn't relevant or high-quality for your account.

What usually happens is the ad manager will add some negative keywords, usually in exact match, though sometimes in other match types if they think to specify, and they don't see all of the search terms that were ignored by AI, they don't add the negatives to the proper account level, and they don't usually see all the metrics tied to the search terms and understand the effect their work is having on the account.

I am not anti-AI. I use Claude Code every day.

Most ad managers are using a blank chat with no context, so the AI relies on the prompt you give it and the information it can find quickly online.

I have a massive context library using the file and folder system used by the top AI experts in the world. My context library contains a ton of information about Google Ads, including my principles, transcripts of my training and coaching calls with clients, and great content from other ad managers and coaches I respect.

However, even with all of this context as a reference for my AI to draw on, it still fails to beat the effectiveness of simply pulling up the search terms report in Google Ads and manually looking through with sorting and filters. Looking manually, I miss fewer opportunities to add negatives, can see which search terms are spending too much without converting, and can see which are driving a lot of impressions without driving clicks, which lowers the CTR and skews performance.

Use AI if you want, but use it as a broad sweep of large date ranges to supplement your manual reviews. If you're curious how far AI can actually go in replacing manual PPC work, Issue 47 of the Agency Overhaul Newsletter goes deep on that question: How Well Can AI Replace Me?

The two-threshold rule for search terms management

If you review search terms a lot, you'll run across situations where you aren't sure whether to add a negative keyword or not. A relevant term with a lot of clicks but no conversions often gets negated, or a search term with really high CPCs gets negated to reduce the CPA.

Both decisions can hurt you, if made too quickly. (As I always say, "never make big decisions on small data")

While there are always nuances, there are some general rules you can follow to make decisions more easily with less fear of blocking searches that could convert if given enough time.

You can add a negative keyword to block a search term when that term has hit 100 clicks with no conversions, or when it has spent more than twice your target cost per acquisition without a conversion. Use whichever threshold you hit first.

At 100 clicks with no conversions, you're falling below a 1% conversion rate, which is low for a search campaign. At twice your target CPA, you've spent what the conversion is worth to you twice over with no return. Waiting past either of those thresholds is usually born out of the "what if" sentiment, which will only waste more of your budget.

After all, the more waste you block, the more quality terms you can show for and your account continues to benefit from blocking that waste and sending clean signals to Google about what you do and don't want.

Before negating something I'm "iffy" about, I like to look at the past 90 days for that term, not just the current window. A term that spent $120 this month with no conversion might have converted twice last month on half the spend. Again, don't make big decisions on small data. (Note: This tends to be much more important for relevant terms that contain buying intent, or for keywords you want to pause. You can block irrelevant terms without checking wider date ranges.)

How to add negatives so they actually block waste

This is where most people, including experienced ad managers, make more work for themselves and leave a lot of future waste unblocked.

We've already explained how negative match types work, but haven't explained how to use them. In the example above, the advertiser was targeting "microsoft copilot governance", and saw the search terms "microsoft 365" and "microsoft copilot tutorial youtube".

If you were to select those search terms in the search terms report and choose to add them as negative keywords, they would be entered as exact match negatives by default, so they'd look like this:

[microsoft 365]
[microsoft copilot tutorial youtube]

However, if you were to just click "Save" and add them, it would add them as exact match negatives at the ad group level.

That means you would still show for searches like "microsoft copilot tutorial you tube" and "youtube tutorial for copilot", or even for the same search terms you're trying to block if they were matched to a different keyword in a different ad group.

So not only do you need to select which match types to use when adding negative keywords, but you also need to select the level to which you're adding them.

The possible levels are:

• Ad group
• Campaign
• Negative keyword list

You can also add negative keywords that apply to the entire account, but that's accessed elsewhere.

Ad group-level negatives block searches matched to keywords in the ad group to which you added the negative. Searches you blocked in that ad group can still trigger your ads in other ad groups in the same campaign. This is useful if you have similar, but different keywords in multiple ad groups. When Google gets confused and matches a search term to the wrong ad group, the person who was searching will see the wrong ad. Adding a negative keyword to the offending ad group can help Google match to the desired keyword in the other ad group, serving the ad you want the searcher to see.

Campaign-level negatives block searches matched to any keywords in the campaign to which they were added, but allow them to trigger ads in other campaigns in your account. As with ad group-level negatives, you can use campaign-level negatives to direct traffic to the appropriate ads. The difference here is when your campaigns, usually segmented by business goal, geography, or other higher-level distinction, match to the wrong traffic, you often skew the data Google has been collecting at the campaign-level about how much your conversions should cost, which affects the algorithm's ability to accurately adjust bids if you use automated bidding strategies (a topic for another post).

A good example would be if you have searches for your company's name (i.e. branded searches) matching to your keywords in your non-branded campaigns. Branded searches tend to convert at a much lower cost per conversion, and you can usually get much cheaper clicks, as well. This would send signals to Google that it can get cheaper conversions, so it lowers bids, reducing your ranking for the non-branded searches you were targeting initially and reducing your chances of converting.

Negative keyword lists are my favorite. You can use them the same way you would use campaign-level negatives, but you can form lists of negative keywords by theme. You can then apply those lists to as many campaigns as you want.

A few of the lists I create for every account are:

• Brand (apply to all non-branded campaigns)
• General Negatives (all irrelevant searches that don't fit other list categories)
• Competitors (block searches for competitors, which tend to be much more expensive and convert far less often; if you want to run a campaign targeting searches for competitors, like "[competitor name] alternatives", just don't apply this list to that campaign)
• Job/Career Negatives (can be combined into general, but if you want to run recruiting ads at any point, you'll want this in a separate list)

Sometimes, if I'm having a lot of trouble controlling costs, I'll also add a separate negative keyword list for "Non-converters", which are search terms that meet one or both thresholds from my two-threshold rule described above. You can also create a "CTR killers" list and include some of the search terms that have 100 impressions and no clicks.

Adding negative keywords to lists gives you a lot of control over what you negate, allowing you to test how your campaigns (or your entire account) would perform if you blocked search terms that you're not completely sure about blocking, or to block terms that you may want to serve ads to in the future.

Once you've understood which level to add a negative keyword to, you'll want to make sure that you're adding them with the proper match type.

As mentioned above, exact match only blocks the exact search included in the brackets, so even if you add it to a list or to the account-level, you're only blocking one search term out of dozens, hundreds, or even thousands of variations of the same search. You're essentially signing up for a never-ending game of whack-a-mole where you could be wasting hundreds, or even thousands of dollars on searches that you already know are irrelevant or low-quality.

To really block waste, and to save yourself a lot of time, you'll want to use phrase match or broad match negative keywords, adding them to higher levels.

In our "microsoft copilot governance" example, we already looked at what can happen if you use the default exact match negatives.

Choosing the right match type takes some nuance, but the idea is that you should block the smallest irrelevant n-gram (i.e. word or set of words) with the match type that will block the most future irrelevant searches while still allowing your target keywords to match to relevant and high-quality search terms.

With our example, adding the phrase match negative "365" to a "General Negatives" list would potentially be safe and would block a lot of variations of irrelevant searches, if that list was applied to all campaigns. Alternatively, if you're not sure, you can start with the exact match negative [microsoft 365], applying it to the same negative keyword list, or just the offending campaign. If you see a lot of search terms including the n-gram "365", and none of them convert, you could then add that phrase match negative to safely block all searches that include "365".

For the "microsoft copilot tutorial youtube" example, you could add "youtube" as a phrase match negative, or you could add "tutorial". If you were to add the entire irrelevant part of the search, or "tutorial youtube" as a phrase match negative, you would still serve for search terms like "microsoft copilot tutorial" and "youtube microsoft copilot setup", wasting budget if you don't actually have tutorials that serve as lead magnets.

When adding a single word as a negative keyword, it doesn't actually make a difference whether you choose to add it as a phrase match or a broad match negative. Both will block any search containing that word.

The nuance comes when you want to block searches that contain two or more words.

Adding "youtube tutorial" as a phrase match negative will block "microsoft copilot tutorial youtube", but would still match to "youtube microsoft copilot tutorial". If both words are irrelevant when placed together, but become relevant when added to a sentence in a different order, phrase match is the way to go.

If you want to block any search that includes both words, but are ok with either of the words appearing without the other, you'd want to use broad match negatives. Adding youtube tutorial as a broad match negative would negate any search that contains both words, but would not block searches that contain only one of the words like "microsoft copilot setup youtube" or "microsoft copilot tutorial".

Understanding the mechanics of negative keyword match types and levels is the key to blocking the majority of wasted ad spend before it happens, which is the foundation of optimizing a Google Ads account.

Adding search terms as keywords

The other benefit to the search terms report is the ability to spot search terms that were matched to your keywords, but which differ slightly.

In the past, it was recommended that you add these top-converting search terms as exact match keywords. While this is still worth testing, Google tends to favor matching these kinds of searches to broad match keywords, rather than giving you the conversion lift you would expect by adding a top-converting search term as an exact match keyword. Doing so tends to increase cost per click prices and actually hurt performance more than if you had just let Google match to the keyword that initially triggered it.

That said, it's not a bad idea to test it, just don't expect the "peel and stick" method that Perry Marshall coined to perform the way it used to.

How often should you manage search terms?

How often you review the search terms report depends on how much data is flowing through your account. You can generally set a review schedule based on your ad spend, as accounts with low CPCs will have lots of clicks and give you plenty of data. Even though accounts with high CPCs will have less data at the same level of ad spend, the damage that can be done by a few bad clicks is significantly higher than the same number of clicks at low CPCs.

There aren't any hard and fast rules for how often to review the search terms report, but you can start with a basic set of ad spend thresholds and adjust your frequency as you get used to how much data you're seeing when you review.

Regardless of spend, you should always check a new account or campaign more frequently, sometimes even daily, to make sure you're not seeing unexpected sources of waste and setting the algorithm up to fail.

With that preface, and knowing that there are always exceptions to the rules, accounts spending $1,000 per month or less don't usually have enough data to justify more than a monthly search terms review, but if it's the only account you're managing, you can get away with a quick review every week to stay on top of it.

For accounts between $1,000 and $5,000 a month, biweekly review of the last 14 days is usually enough.

For accounts spending $5,000 to $15,000 a month, review weekly. When an account is spending $3k-$4k per week, you've got plenty of data to go over.

Once you get past $7,500-$10,000 per week, you start to need review multiple times per week.

Depending on how often your account is changing (i.e. adding new campaigns, adding new keywords, etc), or how heavily weighted your account is toward broad match keywords or Performance Max, you could set a rule to review the search terms report every time your account spends $1,000, or $2,500, or $5,000. It's all up to you and how robust your negative keyword profile is. You'll just want to take into account the current accuracy of your search terms, how often you expand keyword coverage, and how painful it is to waste money on an irrelevant click.

So once you know the frequency of your search terms review, set a recurring calendar event and make sure you stick with it! This is Google Ads management 101, but that doesn't mean you should quit when you become advanced.

All of the best Google Ads veterans know that success comes from consistent application of the fundamentals.

Watch for partial conversions

A quick note before wrapping up.

If your account uses data-driven attribution, which all accounts do by default now, conversion credit is distributed across all the interactions that contributed to a conversion. Your Conversions column will sometimes show decimals like 0.7 conversions or 1.3 conversions.

While it's cool to have more accurate attribution based on how your ads actually influenced a conversion, it screws up the math in your performance tables.

If you see a relevant search term with a $2,000 CPA, your instinct might be to add it as an exact match negative at the ad group level. However, if you see that it has 0.1 conversions, then the math Google did to get your cost per conversion was "cost / conversions", meaning it divided your total search term cost by 0.1, or multiplied it by 10. The search term only spent $200, which may still be within range of your target CPA, and it influenced a conversion, so it is helping the account, but the math is skewing the totals line.

Just keep that in mind when reviewing search terms. Performance may not be as bad as it looks, and negating those search terms could hurt the mechanics of the account as the different pieces work together to influence the pre-contact customer journey.

The bottom line

The search terms report is one of the most consistently neglected parts of Google Ads management. Google hides a significant portion of what you need to see, the industry has largely decided not to look closely at what it can see, and when negatives do get added, they often get added in the way that blocks the least amount of future waste possible.

None of this is complicated, but it is work. AI is helpful, but it has heavily affected the mechanism in our brain that weighs cost against benefit for the actions we consider taking. If given the choice between looking through thousands or tens of thousands of search terms and asking AI to do it, guess which one your brain is going to choose?

It is a lot of work, but I promise, it's worth it while Google still lets us do it.

To quickly summarize this post: understand what you're looking at and what Google is not letting you look at, know what negative keywords and their match types are and when to add them to different levels of your account, set a review cadence based on what your account spends, use consistent thresholds so your decisions are based on data so you don't become a search terms packrat, and check for partial attribution before you negate something that looks expensive but is actually contributing.

Do all of that consistently, and you will outperform most of the Google Ads industry, including a lot of people charging premium thousands (or even tens of thousands) per month to manage these accounts.

If you want help working through your own search terms report and building this into your account management process, that's exactly what coaching calls are for. If you want a full picture of where your account is leaking before you start managing it yourself, start with an account audit.

If you run a Google Ads agency, there are 52 issues of agency-focused strategy in the Agency Overhaul Newsletter archive — covering client retention, account management processes, hiring, and more. Issue 21, Ad Account Audits Done Right, is a good place to start if this post landed for you.