Every Google Ads account has a report that shows, without anaesthetic, what a shop is really paying for: the search terms report. Read regularly, it delivers two things at once: a list of queries to exclude before they eat the budget, and a vocabulary your customers use to describe your products. This guide shows how to read it sensibly and how to turn it into a process rather than a one-off effort.
Concepts
How a search term differs from a keyword
Confusion between these two concepts costs shops real money, so let us start by setting things straight:
Definitions
A keyword is the phrase you want to appear for: you set it in a Search campaign. A search term is the customer’s actual query that triggered the ad.
With today’s broad match, a single keyword collects dozens or hundreds of different search terms. Shopping campaigns have no keywords at all: their role is played by titles and descriptions from the feed, yet search terms still occur. So it is the search terms report that needs monitoring, not the keyword list.
For a shop, the search terms report serves two functions at once. The defensive function: you find queries that accumulate cost without sales and exclude them. The offensive function: you find the words customers really use to describe products and carry them over into feed titles and new keywords. Most accounts use only the first function at best, and only occasionally.
Anatomy of the problem
Three types of query where shops overpay
Irrelevant search terms fall into recurring patterns. This is an extract from the report of an example bicycle shop, with cost and conversions as reported by Google Ads (illustrative data, 30 days):
Search terms report · Shopping and Search campaigns · 30 daysbicycle shop · illustrative data
The three crossed-out search terms add up to €170.75 a month without a single conversion. Each represents a different pattern: a different purchase intent (used), an informational query (repair) and a different service (hire).
These patterns are worth knowing by name, as they recur in every industry:
Different intent
“used”, “bargain”, “free”, “rental”, “jobs”: the customer is looking for something, but not your product in your sales model. Exclude with general phrases, not individual search terms.
Informational queries
“how to fix”, “what is the difference”, “reviews forum”: advice-seeking traffic that rarely ends in a purchase from an ad. This is a topic for content (and organic rankings), not for paid clicks.
Someone else’s territory
Brands and models you do not sell, and services you do not provide. The click ends in an exit, and you pay for someone else’s brand recognition. Exclude without hesitation.
The search terms report should be read with costs and conversions side by side, over a window longer than a single day. Without that pairing it is just a list of phrases; with it, it becomes a price list for decisions.
Process
How to build a negative keyword list monthly
Negative keywords work best as a dull, repeatable process. A proven routine looks like this:
Collect search terms with costs and sales
Pull together search terms from the last 30 days with clicks, cost and conversions as reported by Google Ads. Sort by cost in descending order: the top of the list is your money.
Apply a margin-based threshold
A search term with no conversions that has spent more than the margin from one average order is a candidate for exclusion. Exclude clearly irrelevant terms (different intent, other brands) straight away, regardless of cost.
Exclude at list level
Use shared negative keyword lists for the account instead of individual entries per campaign: new campaigns inherit the clean-up, and you maintain one list instead of ten.
Spot search terms for product titles
Move converting search terms whose words are missing from product titles into the feed. This is the other half of the report’s value, covered in more detail in the guide to Google feed fields.
Record decisions and return in a month
A short note on “what was excluded and why” prevents decisions from being reversed when someone else takes over the account. A monthly routine is enough; for new campaigns, every two weeks.
Methodological note: judge “zero conversions” at search term level over a longer window, not a single day, and remember the role of assists: a generic term is often the first touchpoint, with the order arriving later through another channel. Margin-based thresholds limit the risk but do not replace common sense when data series are short. The attribution mechanism is described in the guide to conversions and attribution.
Real risk
Excluding based only on a few days of “no conversions in the dashboard” also cuts terms that sell. Measurement misses some transactions (consent, blockers): check the scale by comparing the total number of shop orders with purchase transactions in GA4 for the same period (GA4 coverage). The lower that coverage, the longer the window and the more caution generic terms with assists need before exclusion.
The other side of the report
Search terms as a customer glossary for titles and offers
Converting search terms are free language research. Customers name products differently from manufacturers: “front bike child seat up to 15 kg” instead of a model code, “waterproof panniers” instead of a series name. If these words are missing from feed titles, Shopping campaigns match less well than they could.
The practice is simple: once a month, take the search terms with conversions, compare them with the words in the titles of the products that convert in Shopping campaigns, and add the missing terms (type, use, specification) following your title structure. The same review suggests new keywords for Search campaigns and phrases worth building content around instead of paying for informational traffic.
Key takeaway
Every search term ends up in one of three places: on the negative list (cost without value), in titles and keywords (customer language that sells), or left alone (working as it should). A report reviewed without these three buckets is a report reviewed for nothing.
DataOrganizer
How DataOrganizer speeds up search term work
The whole process described above needs one thing: search terms with costs, conversions and product titles in one place. DataOrganizer pulls the full search terms report from Google Ads every day (with cost, clicks and conversions per term) and places it next to the Shopping product report, product titles from the shop catalogue and total shop sales. The monthly review turns into a single question:
Eight search terms spent a total of €460 with no conversions in Google Ads in 30 days. The three largest:
| Search term | 30-day cost | Clicks | Pattern |
|---|---|---|---|
| used bike bargain | €85.50 | 228 | different intent |
| how to fix a derailleur | €46.75 | 141 | informational query |
| bike rental | €38.50 | 96 | different service |
I have prepared the full list to paste in as negative keywords, plus, separately, five converting search terms whose words are missing from product titles. At current spend, this list saves around €450 per month, counting only terms without a single conversion.
Key takeaway
Everyone has access to the search terms report; the time saving comes from combining it with other data. A search term × cost × conversions × product titles view, refreshed daily, sits next to total shop sales. With it, an exclusion decision takes seconds and rests on a 30-day window instead of a single day.
FAQ
Frequently asked questions about search terms
How does a search term differ from a keyword?
You set a keyword in the campaign; a search term is the actual customer query that triggered the ad. With broad match, one keyword collects hundreds of search terms, and Shopping campaigns have no keywords at all: feed titles play that role. The search terms report is what you need to control.
Where can I find the search terms report in Google Ads?
In Search campaigns: in the campaign statistics, as the search terms report. In Shopping and Performance Max campaigns, a view of search terms or search term categories is available, with less detail. The report shows queries with clicks and cost; the value emerges once it is read alongside conversions and margin.
How often should I review search terms and add negative keywords?
A monthly routine is enough for a stable account; every two weeks for new campaigns and growing budgets. Exclude terms that accumulate cost without sales over a longer window rather than reacting to single days, and collect them in shared lists that new campaigns inherit.
At what cost should a search term without conversions be excluded?
A practical threshold comes from margin: a search term with no conversions that has spent more than the margin from one average order qualifies for exclusion. Exclude clearly irrelevant terms (different intent, other brands, informational phrases) straight away, without waiting for the threshold.
Do search terms help with product titles?
This is the cheapest source of your customers’ language. Converting search terms whose words are missing from your titles form a ready-made list of feed fixes; the title structure is covered in detail in the guide to Google feed fields. It works the other way too: irrelevant search terms reveal which words in your titles confuse the algorithm.
Cheat sheet for the whole guide
Three rules for working with search terms
1. Read search terms with money and conversions alongside. Sorting by cost and reading conversions over a 30-day window turns a list of phrases into a price list of decisions. Measurement does not see some transactions, so check GA4 coverage (total shop orders vs GA4 purchase transactions for the same period) before excluding generic terms with assists.
2. Three buckets for every search term. Exclusions (different intent, informational queries, someone else’s territory), titles and keywords (customer language that sells) or leave alone (performing correctly). Exclude using shared lists, with a threshold calculated from margin.
3. Routine instead of bursts. A monthly review with a note of decisions, and each month converting search terms feed into the product feed. DataOrganizer reduces the whole process to a single question: search terms, costs, conversions and product titles sit in one place, with the exclusion list ready to download.
DataOrganizer · search terms with costs and conversions
Connect Google Ads and your shop: search terms with costs and conversions sit next to product titles and total shop sales, and the exclusion list and words for your titles come from a single question. Demo account available instantly; you pay only for connected sources.