Demonstration mode — checkout is simulated and reports are samples. Live payments open at launch.

What is a similarity score? (And what percentage is acceptable?)

When you run a document through a similarity checker, it compares your text against a large database of web pages, journal articles, books and other documents, and reports the percentage of your text that matches something else. That percentage is your similarity score (sometimes called a similarity index).

A similarity score is not a plagiarism verdict

This is the single most important thing to understand. Similarity software finds matching text; it cannot judge intent, quotation or citation. All of these count toward your score, and none of them are plagiarism:

Conversely, a low score does not guarantee innocence: a short, entirely copied paragraph can sit inside a long document and produce a small overall percentage while still being a serious problem. Always read the match list, not just the headline number.

So what percentage is acceptable?

There is no universal threshold, because every university, department and assignment sets its own rules. In practice:

Some institutions publish a hard limit (often around 20%, excluding bibliography and quotations). Your own institution's rule — not a number you read online — is the one that applies to you. Check your handbook, and when you check your document, use a report that excludes nothing you can hide behind: look at where the matches are.

How to lower a score legitimately

Ready to see your own number? Check your document — you will see the word count and price before you pay, and your document is deleted as soon as your report is ready.