Free tool · runs in your browser
Google Ads N-Gram Tool
Drop your search terms report and see which words and phrases spend without converting, across every search that contains them. Tick the bad ones, copy them as negative keywords, paste into Google Ads. No script, no Google Sheet, nothing uploaded.
- Reads the raw export, any format
- 1- to 4-word n-grams
- Copy as negatives in one click
"free" · "jobs" · "diy" · "salary"
"free" · "jobs" · "diy" · "salary" · "course"
.csv, Excel .csv, .xlsx or .tsv, straight from the Google Ads download button. Title rows and totals are handled.
Nothing is uploaded. The file is read in your browser and never leaves your computer.
- 0search terms analysed
- 0total cost in the report
- 0spend on terms with no conversion
- 0conversions in the report
| N-gram | Terms | Impr. | Clicks | Cost | Conv. | Cost / conv. | CTR | Conv. rate | Conv. value | ROAS |
|---|
Export guide
How to export the search terms report from Google Ads
Four clicks. The tool does the rest, so do not tidy the file before you drop it.
- 1
Open the search terms report
In Google Ads go to Campaigns, then Insights and reports, then Search terms. The report shows every search that triggered your ads.
- 2
Pick the date range
Use at least 30 days, ideally 90. Fewer days means fewer clicks per word and noisier results. Leave the columns as they are; the defaults already include clicks, impressions, cost and conversions.
- 3
Download it
Click the download icon above the table and pick any format: .csv, Excel .csv or .xlsx. Do not open and re-save it in a spreadsheet first; the tool reads the raw export, title rows and totals included.
- 4
Drop the file here
Drag it onto the box at the top of this page or click to browse. The tool finds the header row, the currency and the columns on its own, then shows the n-grams sorted by cost.
Why n-grams
What an n-gram analysis shows that the search terms report hides
The report lists every search on its own row. The word that is wasting your budget is spread across hundreds of those rows, a few clicks each, and never looks like a problem on any one of them.
Small leaks, everywhere
"plumber jobs", "plumbing jobs near me", "jobs for plumbers", "plumber job openings". Four rows, a handful of clicks each, none of them worth a second look. Together they are one word, jobs, with a real cost and no conversions.
Judge the word, not the row
The n-gram analysis adds up clicks, cost and conversions for every search term that contains a word or phrase. Sort by cost, filter to zero conversions, and the leaks are the first rows on the screen. Longer n-grams work the other way: 3- and 4-word phrases that convert are keywords you should be bidding on.
Reading the results
Three passes, then paste the negatives
1 word, sorted by cost, zero conversions
The broad waste: intent words that never buy. Check the Terms column to see how many searches each one sits in before you block it.
2 words, same filter
Phrases that only mean trouble together. "near me" in a city you do not serve, "repair manual", "how to". Phrase-match negatives are the safe default here.
3 and 4 words, sorted by conversions
This is where the winners show up. Phrases with a good cost per conversion that are not yet keywords belong in their own ad group with their own ad.
Questions
Google Ads n-gram analysis, answered
What is an n-gram in Google Ads?
An n-gram is a run of n words taken from a search term. "cheap plumber near me" contains the 1-grams cheap, plumber, near and me, the 2-grams cheap plumber, plumber near and near me, and so on. An n-gram analysis adds up the clicks, cost and conversions of every search term that contains each n-gram, so a word that appears in hundreds of different searches gets judged on its total, not one row at a time.
How is this different from the Brainlabs n-gram script?
The script pulls your search terms into a Google Sheet and builds the n-grams there. It needs to be installed in the account, scheduled, and it no longer runs in the new Google Ads scripts environment without changes. This tool takes the report you can already download and does the same maths in your browser. Nothing to install, nothing to schedule.
Is my search terms data uploaded anywhere?
No. The file is read by JavaScript in your browser and never leaves your computer. There is no server behind this page. You can turn off your internet after the page has loaded and it still works.
Which n-gram size should I use?
Start with 1-grams to catch the broad waste: words like free, jobs, diy, salary, course. Then 2-grams for phrases that only mean trouble together, such as near me when you do not serve that area, or repair manual. 3-grams and 4-grams are mostly for spotting good phrases you should add as keywords.
Which columns does the export need?
Search term, Clicks, Impressions and Cost are enough. Add Conversions to see what converts and what does not, and Conversion value if you track revenue. The default columns in the Google Ads search terms report already include all of these except conversion value.
How do I add the n-grams as negative keywords?
Tick the n-grams you want to block, pick a match type, and press Copy negatives. Then in Google Ads open the campaign or a negative keyword list, choose Add negative keywords, and paste. Phrase match blocks every search that contains the words in that order. Exact match blocks only that precise search. Broad negatives block any search containing all the words in any order.
Does it work with Microsoft Ads or other exports?
Yes, as long as the file has a search term column and the usual metric columns. Microsoft Ads uses Spend instead of Cost and Revenue instead of Conversion value; both are recognised. Excel exports (.xlsx), Excel-flavoured CSVs and tab-separated files all work.
Free lead gen strategy
Found the waste? We fix the rest.
Negatives are one leak. Fill this in and we look at your whole account, your pages and your tracking before we talk. On the call you get a lead generation strategy for your business and a recommendation: coaching, done for you, or neither.
- No obligation, no slide deck
- For service, finance, legal and insurance businesses
- Johan replies personally