Text Similarity Checker
Compare two texts and get a similarity percentage based on Jaccard similarity of their word sets.
How to check text similarity online
Paste one piece of text into each box above. The similarity score updates live, showing how much overlap there is between the words used in each text.
How similarity is calculated
This tool uses Jaccard similarity on word sets: each text is broken into its set of distinct words (case-insensitive, punctuation ignored), and the similarity score is the number of words shared by both texts divided by the total number of distinct words across both texts combined, expressed as a percentage. A score of 100% means both texts use exactly the same set of words; 0% means they share no words at all. This word-overlap approach is simple, fast, and works well for comparing general topic and vocabulary overlap, though it doesn't account for word order or exact phrasing.
Example
"The cat sat on the mat" and "A cat sat on a rug" share the words "cat", "sat", and "on" out of 8 distinct words total across both texts, giving a similarity of roughly 37.5%.
Common Use Cases
- Comparing two drafts to see how much their wording overlaps.
- Checking how similar two product descriptions or ad variations are.
- Getting a quick sense of topical overlap between two documents.
FAQs
Does word order matter?
No — this method compares the sets of words used, not their order or the sentences they form. Two texts with the same words in a different order would score 100% similar.
Does this detect plagiarism?
No — this is a simple word-overlap similarity measure, not a plagiarism detector. A high score means the two texts share a lot of vocabulary, not that one copies the other's exact phrasing or structure.
Why is the score lower than I expected for two similar paragraphs?
Jaccard similarity divides shared words by all distinct words combined, so even paraphrased text with the same meaning but different vocabulary choices can score lower than you might intuitively expect.