Search Term Analysis: Finding Wasted Spend in Google Ads Accounts

The monthly review process we run to cut irrelevant spend and reallocate budget to converting queries.

When we audit a new Google Ads account, the search terms report is one of the first places we look. It is also one of the most consistently neglected parts of account management. In accounts spending $500 or more per month, we typically find 10-20% of budget going to searches with no realistic chance of converting. That is dead spend, and most of it is preventable.

Understanding why it happens requires understanding a distinction that surprises a lot of business owners.

Keywords vs Search Terms: Not the Same Thing

Keywords are what you put into Google Ads. Search terms are what users actually typed before your ad appeared. These are not the same thing, and the gap between them is where wasted spend lives.

When you add "Greek restaurant Adelaide" as a phrase match keyword, Google will show your ad for searches it considers related. That includes "Greek restaurant Adelaide menu", "Greek restaurant Adelaide jobs", "cheap Greek restaurant Adelaide student", and possibly "best Greek restaurant recipe". Some of those are fine. Others are completely off-target.

Broad match makes this problem significantly worse. A broad match keyword like "restaurant Adelaide" can trigger ads for searches as loosely related as "best places to eat in Adelaide Hills" or "cook restaurant quality food at home Adelaide". Without a well-maintained negative keyword list, broad match distributes your budget across a long tail of searches, many of which convert at zero.

The search terms report shows you every query that triggered an ad impression in the selected date range. It is the only place you can see what is actually driving spend.

How to Access the Search Terms Report

In Google Ads, go to Campaigns in the left navigation. Select the campaign you want to review. Then go to Insights and Reports, and click Search Terms. You can also access it from the Keywords section of any campaign by clicking the Search Terms tab at the top.

Set the date range to the last 30 days as a default. For a new campaign, check weekly. For established campaigns, monthly is usually sufficient.

Sort by cost descending. You want to see what is spending the most money first, because that is where the largest savings are.

The 4 Things to Look For

Not all irrelevant search terms are the same. There are four distinct categories to identify.

The first is outright irrelevant queries. These are searches with no connection to your offering. A restaurant running "burger" as a keyword might receive clicks for "burger recipe", "how to make burgers at home", "burger seasoning blend". Someone searching those terms is not going to book a table. Add every one of these to your negative keyword list immediately.

The second is low-intent queries. These are loosely related searches from people in research mode rather than decision mode. For a restaurant, that includes searches like "best suburbs for food in Adelaide" or "what is a traditional Greek meal". The person is curious, not ready to book. These clicks drain budget without converting.

The third is competitor brand terms. If you are using broad or phrase match, your ads may appear when someone searches for a competitor by name. Paying to intercept someone who specifically searched for a competitor is rarely cost-effective. Add competitor brand names to a negative keyword list unless you have a deliberate conquesting strategy with separate campaigns and specific ad copy.

The fourth is high-spend, zero-conversion terms. These are searches that look plausible on the surface but have consumed meaningful budget without a single conversion. A search term that has spent three times your target cost per acquisition without a conversion is a signal worth acting on. Exclude it or reduce the match type to exact match only.

Building a Systematic Negative Keyword List

A one-off cleanup of the search terms report is useful but not enough. The negative keyword list needs to grow continuously.

Organise negatives at two levels. Account-level negative keyword lists apply across all campaigns in the account. These should contain universal exclusions that will never be relevant to your business: "jobs", "careers", "how to", "recipe", "DIY", "free", "cheap" if you are a premium venue, and any unrelated industries.

Campaign-level negatives address intent mismatches specific to that campaign. A campaign targeting catering bookings might exclude "restaurants near me" and "table for two" because those searches suggest individual dining rather than group catering. A dining campaign might exclude "catering quotes" for the same reason in reverse.

Cross-campaign negatives are also worth setting up. If you have a brand campaign and a non-brand campaign, add your brand name as a negative in the non-brand campaign. This prevents both campaigns from competing against each other for branded searches, which inflates CPCs unnecessarily.

Start with a list of 20-30 obvious exclusions before launch. Add to it monthly based on what the search terms report reveals.

How Match Types Affect Search Term Pollution

Match type is the lever that controls how widely Google interprets your keywords.

Exact match is the tightest. Your ad shows only when someone searches your exact keyword or a very close variant, like a plural or a misspelling. "Book Greek restaurant Adelaide" on exact match will not trigger for "Greek restaurant reservations Adelaide". Exact match gives you the cleanest search term data but limits volume.

Phrase match sits in the middle. Your ad shows when a search contains your keyword phrase, with words added before or after. "Book Greek restaurant Adelaide" on phrase match will show for "best place to book Greek restaurant Adelaide" but not for "Greek restaurant book a table". Reasonably controlled, but the search terms report will still show some irrelevant queries that sneak through.

Broad match is the widest. Google decides what your keyword means in context and matches to anything it considers semantically related. In a well-managed account with a strong negative keyword list and good conversion data, broad match can find high-value queries you would not have thought to target. In a new account or a poorly managed one, it is a budget drain.

For new restaurant campaigns, start with exact and phrase match only. Add broad match keywords selectively once the account has conversion data and a mature negative keyword list.

The Monthly Cadence

Thirty minutes per account per month is the minimum for search term review on any account spending $500 or more. For accounts spending $2,000 or more, budget 45-60 minutes.

The monthly process looks like this: filter search terms by cost descending, review the top 50-100 by spend, add irrelevant or low-intent terms to the appropriate negative list, check for any high-spend zero-conversion terms to exclude or investigate, and note any new positive terms worth adding as keywords with tighter match types.

In the first three months of a new campaign, check weekly. The negative list grows fastest in this period, and letting irrelevant spend accumulate for a full month is expensive. After three months, the negative list should be mature enough that monthly reviews catch the stragglers.

What the "Search Terms Driving Most Impressions" View Hides

Google Ads has a summary section in Insights that shows "search terms driving most impressions". This is not the same as the full search terms report. It is a curated subset, and it hides the long tail of low-volume queries that collectively account for a significant share of spend.

The full report is what you want for the monthly audit. The summary view can miss the queries appearing 3-10 times at $4 per click, which adds up to real money across dozens of irrelevant terms over a month.

Sort by cost, not impressions, when reviewing the full report. High-impression, low-cost terms are less urgent than low-impression, high-cost terms. You are looking for budget leaks, not just volume.

Real Examples of Bad Search Terms for a Restaurant

To make this concrete: a Greek restaurant bidding on "Greek food Adelaide" with phrase match received clicks for "Greek food Adelaide recipe", "Greek food history Adelaide festival", and "how to make Greek food at home Adelaide". None of those searchers were booking a table. Each click cost between $2 and $3.50.

The fix was straightforward: add "recipe", "history", "festival", and "how to make" to the account-level negative list. Those exclusions prevented future spend on the same category of searches across all campaigns.

A Vietnamese restaurant bidding on "catering Adelaide" with broad match received clicks for "catering equipment Adelaide", "catering course Adelaide", and "catering supplies Adelaide". None of those searchers wanted food. The fix: add "equipment", "supplies", "course", and "hire" to campaign-level negatives for the catering ad group.

In both cases, identifying and excluding these terms freed budget to go toward searches that were actually converting. The total budget did not change. The return on that budget improved because a higher percentage of clicks came from people with genuine intent.

That is the entire purpose of search term analysis. Not to cut spending, but to make every dollar of spending work harder.