Text Similarity Checker: Compare Two Passages Instantly

This tool compares two pieces of text and gives you two independent similarity scores instantly, entirely in your browser. It is built for the everyday case of checking whether a rewritten paragraph, a new product description, or a blog draft is too close to something already published, whether that is your own earlier content or a competitor's.

How the similarity score is calculated

Character similarity is based on Levenshtein distance, a well established algorithm that counts the minimum number of single-character edits, insertions, deletions or substitutions, needed to turn one string into the other. That edit count is converted into a percentage relative to the length of the longer text. Word overlap uses Jaccard similarity instead: each text is split into unique lowercase words, and the score is the size of the shared vocabulary divided by the size of the combined vocabulary. The two scores answer different questions, one cares about exact wording and order, the other only cares about which words appear at all.

What counts as duplicate content

Search engines do not publish a fixed similarity percentage that defines duplicate content, and their internal comparison methods are far more advanced than a straightforward character or word diff. What tends to matter in practice is whether two of your own pages are competing for the same search intent while saying largely the same thing, which risks splitting ranking signals between them instead of consolidating them onto one strong page. Use this tool as a fast first check, not as a substitute for a full content audit.

Where your text goes

Nowhere. Both comparisons run as JavaScript directly in your browser tab. Nothing you paste is sent to a server, logged, or stored anywhere, which means it is safe to paste unpublished drafts, client copy under NDA, or anything else you would not want leaving your machine.