N-Gram

What is an N-gram?

An n-gram is a contiguous sequence of n words taken from a body of text. A unigram (1-gram) is a single word. A bigram (2-gram) is two consecutive words. A trigram (3-gram) is three consecutive words. The term comes from natural language processing (NLP), where n-gram analysis is used to understand patterns in how language is used in a given corpus - whether that corpus is search queries, product reviews, ad copy, or keyword lists.

N-gram analysis in PPC

For paid search marketers, n-gram analysis is a technique for auditing PPC campaigns at scale. Instead of reviewing every individual search term in a Google Ads account - which can number in the thousands for active campaigns - n-gram analysis aggregates performance data by word or phrase frequency across all search terms. This surfaces patterns that are invisible at the individual term level: a two-word phrase that appears across 200 search terms and consistently has a high conversion rate, or a word that appears across 150 terms and has zero conversions despite significant spend. These patterns guide negative keyword additions and positive bid adjustments far more efficiently than manual term-by-term review.

The practical workflow is: export your search terms report from Google Ads, run n-gram frequency and performance analysis (typically in Excel, Python, or a dedicated tool like Search Terms Manager), identify high-CPL or zero-conversion phrases to add as negatives, and identify high-converting phrases to add as exact or phrase match keywords. Doing this regularly - monthly or quarterly - is one of the most reliable ways to improve CPA on Google Search campaigns by progressively eliminating wasted spend on irrelevant queries.

N-gram analysis in SEO and content

In SEO, n-gram analysis of competitor content and top-ranking pages helps identify which phrases and word combinations Google associates with a given topic. By analysing the n-grams that appear frequently across the top 10 results for a target keyword, you can identify related terms and semantic phrases that should be included in your own content to signal topical depth and comprehensiveness. This is part of what is broadly called semantic SEO or entity-based optimisation. N-gram analysis also informs keyword research by surfacing long-tail phrase variations that might not appear in standard keyword tools but have measurable search volume in aggregate.