Direct answer
Direct Answer - There is no single "good" Turnitin similarity score, because the percentage measures how much of your text matched sources in Turnitin's database — it is not a plagiarism verdict [1]. In practice, most universities treat anything under roughly 10–25% as unremarkable, but the number only matters alongside the highlighted matches that produced it [2]. A score of 30% made up of properly cited quotations and a reference list is usually fine, while a 12% score with one uncredited paragraph lifted from a journal article is a genuine problem.
What Similarity Percentage Is Considered a Good Score on a Turnitin Report?
Turnitin's own documentation is unusually direct on this point: the similarity score is the percentage of a paper's text that matches its repository of student papers, publications, and internet sources, and it is designed to be read as a starting point for interpretation rather than a conclusion [1]. The company states plainly that there is no fixed percentage that indicates plagiarism, and that a high score may reflect legitimate quotation while a low score may still contain unoriginal work that simply was not indexed [2]. That is why your institution — not Turnitin — decides what number triggers a conversation.
The informal bands most commonly cited by instructors and academic skills teams cluster in a fairly narrow range. Scores below about 10% are usually read as normal academic writing, since standard phrasing, module titles, and assignment wording will always match something [2]. Scores in the 10–25% band are typically treated as unremarkable provided the matches are attributed, and scores above roughly 25% tend to prompt a closer look at the match overview rather than an automatic penalty [1]. These are conventions, not rules, and a department that sets a 15% ceiling is exercising its own academic judgement.
The more useful habit is to stop reading the headline number and start reading the sources behind it. Turnitin's report breaks matches down by source and shows you exactly which strings of text triggered each one, so you can tell a cited block quote from an unattributed paragraph in seconds [1]. A score is only "good" or "bad" in relation to that breakdown — which is why two papers with identical percentages can have completely different integrity outcomes [2].
Why Does a Turnitin Similarity Score Look High When the Writing Is Genuinely My Own?
Turnitin matches text, not intent, so anything you have quoted, paraphrased with attribution, or reproduced from a template will still register as a match [3]. Direct quotations are the most obvious case: a well-used block quote from a key source is exactly what a good essay should contain, yet it will appear in the report at full length. Paraphrases count too, because Turnitin's algorithm compares strings of words rather than ideas, and a close paraphrase that keeps the original's phrasing will overlap heavily even when it is properly cited [3].
Reference lists and standard academic furniture are a second common culprit. Bibliographies, in-text citations, module codes, headings, and conventional methodology sentences such as "this study uses a qualitative approach" all tend to match existing sources, and across a long reference list those matches accumulate quickly [3]. Turnitin offers filters that can exclude bibliographies and small matches, but those filters are applied by the person viewing the report, so your instructor may see a number that differs from the one you saw [3].
A third pattern is the dispersed match: many small overlaps with many different sources, none of them large enough to look alarming on its own. Turnitin's guidance notes that a paper can reach a moderate overall percentage purely through cumulative short matches in common academic phrasing [3]. When you open the match overview, look at whether the highlighted text is concentrated in a few substantial blocks or scattered as single sentences — the first pattern deserves attention, the second usually does not.
How Can I Check My Own Turnitin Similarity and AI Score Before I Submit?
Turnitin's student-facing guidance explains that whether your draft is stored in the student paper repository depends on how it was submitted, which is why many students prefer a pre-submission check that does not add their file to the database [4]. A non-repository check lets you see the same similarity report your instructor would see, without your draft becoming a future match for someone else's paper. Turnitin also advises students to review the similarity report and the AI writing indicators together, since the two reports answer different questions about a draft [4].
The AI side has its own quirk worth knowing before you panic. When Turnitin's AI writing detection is not confident enough, the report displays an asterisk-percentage (*%) rather than an exact figure, because the signal falls below its 20% confidence threshold [4]. Seeing *% therefore means low-confidence detection, not a hidden high score. Reviewing both reports side by side before submission gives you time to fix a genuinely unattributed passage or rework a heavily flagged section while you can still act on it [4].
A practical pre-submission routine is short: run the check, open the match overview first, then read the sources list, then decide whether any highlighted passage is unattributed rather than merely quoted [4]. If the matches are all cited material and your own analysis, the number is doing its job. If a paragraph is highlighted and you cannot point to the source you credited, that is the passage to rewrite — regardless of what the overall percentage says.
Reading your own report is far easier when you can see the real thing rather than a mock-up. turnitin0 gives you the actual Turnitin similarity and AI writing reports — the same cover page, score band, flagged passages, and source summary your professor opens in the LMS — so you can interpret your percentage with the evidence in front of you instead of guessing at what the number means.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is 20% a bad Turnitin similarity score?
Not by itself. A 20% score is inside the range most institutions treat as routine, and it only becomes a concern if the match overview shows an unattributed passage rather than cited quotations or reference-list entries [2].
Can I get 0% on a Turnitin similarity report?
It is possible but not necessarily desirable. A 0% score can mean your paper matched nothing in the repository, but it can also mean you cited very little — Turnitin notes that a low score is not proof of good academic practice [2].
Do quotations count toward my similarity score?
Yes. Direct quotations and close paraphrases register as matches even when correctly attributed, which is why Turnitin offers filters that can exclude quoted material and bibliographies when the report is viewed [3].
Does the similarity score tell me anything about AI writing?
No. Similarity and AI writing detection are separate reports; the AI indicator shows *% when detection confidence falls below Turnitin's 20% threshold, so the two figures should always be read independently [4].
Should I check my draft before submitting it?
Reviewing your similarity and AI reports before final submission lets you correct unattributed passages while you can still act, and a non-repository check avoids adding your draft to the student paper database [4].