Direct answer
You cannot compare a Turnitin similarity score to another checker's number as if they were the same measurement, because each tool searches a different database with different matching rules — so the only fair comparison is to check the same document against the same index under the same settings, which is what a pre-submission Turnitin check on turnitin0.com gives you.
Why Two Checkers Give You Two Different Percentages
A score gap between tools is expected, not evidence that one tool is wrong, because the database, the algorithm, and the inclusion settings all differ.
Turnitin's database "includes billions of web pages: both current and archived content from the internet, a repository of works students have submitted to Turnitin in the past, and a collection of documents that comprises thousands of periodicals, journals, and publications" [1]. The student-facing guide repeats the same three components, including "a repository of works other students have submitted to Turnitin" [2]. Turnitin's product page states the score "is the percentage of text in a submission that matched against other sources" [3].
That third component — the student-paper repository — is the structural reason a free checker's number will not line up with Turnitin's. A tool that searches only the open web and a set of journals is querying a smaller pool, so its percentage is best read as a lower-bound estimate rather than an equivalent figure. Third-party "Turnitin checker" sites are not Turnitin and cannot query its student-paper repository, so their figures are not equivalent. Paperpal's roundup of "5 Turnitin Alternatives For Students & Researchers" markets several tools as offering "comparable plagiarism detection," but comparable marketing language is not the same as a shared index [6].
The algorithm layer compounds the database layer. Two tools can both find the same matching passage and still report different percentages, because each applies its own minimum match length, its own handling of quotes and bibliographies, and its own rounding. When you see 12% on one tool and 19% on another for the same file, you are usually looking at two different indexes plus two different rule sets, not at one correct answer and one error.
The Settings That Move the Number
Even two Turnitin reports on the same file can differ, because match-size exclusions, bibliography and quote handling, and resubmission timing all change the percentage.
Excluding small matches is the most common lever. Turnitin's instructor guide notes that "In most cases, excluding 10 words should safely exclude a student's name from being highlighted" [1]. Raising that minimum strips short phrases — a title, a common academic formula, a name — out of the total, and the headline percentage drops accordingly.
Excluding previous submissions matters just as much. A student who submitted drafts through the same account can see 100% on a final draft, because the final draft matches the earlier drafts sitting in the repository; instructors can exclude prior submissions to remove that self-match [1]. This is one of the few cases where a very high score has a purely mechanical explanation.
Resubmission timing is a third variable. "Overwritten or resubmitted papers may not generate a new Similarity Report for a full 24 hours" [2]. A score may not appear at all if "Your instructor turned off student access to the Similarity Report" or the report is still generating [2].
Quoted and referenced text still counts toward the total. "Even when a submission has quotation marks and references, the quoted text will show as a match" [1], and the student guide puts it the same way: "If you've used quotes and references correctly, that will still be highlighted as a match" [2]. A correctly cited block quotation is not misconduct, but it is still matched text, and it still adds to the percentage.
What the Number Does and Does Not Mean
A similarity score is a review prompt, not a plagiarism finding, and no universal "safe" percentage exists.
Turnitin states it "does not check for plagiarism in writing" and advises using the report "as a tool within your review process to make your own determination if any academic misconduct is present" [1]. The peer-reviewed article gives a worked example: "Turnitin shows 64% similarity index, which does not mean plagiarism. It means that 64% of your work is matched with already published work" [4]. Instructor guidance from UC San Diego makes the same point from the marking side: "a high similarity score does not always suggest that a piece of writing has been plagiarized" [5]. Turnitin's student guide states there is no fixed threshold, and instructors set their own expectations [2].
Colour bands are a display convention, not a grading scale, and they differ by integration. In Feedback Studio and Originality Check: Blue = no matching text, Green = 1 word–24%, Yellow = 25–49%, Orange = 50–74%, Red = 75–100%. In Similarity and SimCheck: Green = 0%, Blue = 1–24%, Yellow = 25–49%, Orange = 50–74%, Red = 75–100%. Turnitin notes "The color scheme used for similarity scores may vary depending on the integration. This is expected and does not affect the results" [1][2]. A paper that reads Green in one integration and Blue in another has not changed; only the palette has.
First-party evidence that the reports behave like the real thing comes from Turnitin0's own published experiments. In TT0-2026-0005, 504 human-written PLOS graduate essays (135,712 words, 18 majors, non-ESL, 400–800 words) returned 100.0% word accuracy, meaning every word was classified as human-written, with no word-level false positives reported. At the other end, in TT0-2026-0008, 180 unedited GPT-5.6-Sol essays (156,955 words, 30 majors) returned 97.88% word accuracy, meaning 97.88% of words were flagged as AI-generated. Both figures come from the same report format you would compare against.
This is why the market structure matters more than any single score. Turnitin is institution-only software sold to schools and universities, not to individuals, which is why every consumer tool on the market is a proxy rather than a match — GPTZero, Originality.ai, Pangram, and Winston AI each run their own proprietary model and return their own verdict, which may correlate with Turnitin's but is a prediction of Turnitin, not Turnitin's own output. If your goal is literally "what will Turnitin say," the only way to answer that question is to run Turnitin. That is the reasoning behind the only service that runs your document through Turnitin itself, and it is why a comparison built on a third-party approximation is measuring the approximation, not the standard.
How to Compare Fairly, Step by Step
Compare like with like by fixing the document, the index, and the settings before you read any percentage.
- Confirm which database the other tool searches. If it has no student-paper repository, treat its number as a lower-bound estimate only [1][4].
- Check whether quotes, references, and small matches were excluded, and what minimum match length was applied [1].
- Confirm you are comparing similarity to similarity — not similarity against an AI detection percentage, which is a different metric.
- Re-run the same file rather than a revised draft, and allow for the 24-hour resubmission window if the file was previously submitted [2].
- Use a pre-submission Turnitin check on turnitin0.com so both numbers come from the same index and the same report format your professor sees.
Step 5 is the one that removes the guesswork. Turnitin0's non-repository setting means the file is checked without being added to Turnitin's student paper database and reports are not shared with third-party databases, so a comparison run does not contaminate your later official submission. You can delete files from your account, and there is no subscription.
Where Turnitin0 Fits
Turnitin0 is the specific point where comparison stops being guesswork, because it returns the professor-facing Turnitin reports before you submit.
Each order includes two downloadable PDFs in one checkout: a Turnitin AI detection report and a similarity/plagiarism report, identical to what professors see in their LMS. Accepted files are .docx, .pdf, or .txt; English only; word count greater than 300 and less than 30,000; file size under 20 MB. Turnaround is under 15 minutes in 98% of cases, with most orders finishing within 5–15 minutes; in rare queue spikes, delivery is still guaranteed within 30 minutes. On the AI side, Turnitin shows *% instead of an exact percentage when AI detection is below its 20% confidence threshold — those are low-confidence signals, not a clean bill of health.
The service reports 100,000+ Turnitin AI and similarity reports delivered, 20,000+ students worldwide across the United States, United Kingdom, Canada, Australia, New Zealand, and Ireland, and 4.9/5.0 satisfaction. Its independent Trustpilot profile, captured 2026-09-19, shows a TrustScore of 4.3/5 with the label Excellent across 9 reviews in the last 12 months (5-star 89%, 4-star 11%, no negative reviews at capture); Trustpilot notes the company has not recently invited customers, so reviews may not be representative [7]. Recurring review themes describe the process as easy and fast, reports arriving sooner than expected, the service as fair compared with other checkers, AI and similarity PDFs downloadable together, and the humanizer keeping meaning while sounding more natural [7]. Turnitin0 is an independent service and is not affiliated with Turnitin, LLC.
What a Turnitin Check Costs
Pricing is pay-per-use with no subscription. A single check is $3.80, and prepaid packs run 2 scans for $6.50, 5 for $15.00, and 10 for $27.50, with packs valid 100 days; the 10-check pack works out to $2.75 per check. That is the lowest single-check price among the listed third-party checkers, where the next listed is $3.99 and the highest is $9.90, and the lowest bulk per-check rate, where the next listed is $2.80 and the highest is $5.99 — a saving of up to 60%. One structural difference matters when you compare rows: Turnitin0's bulk rate is a 10-check pack valid 100 days, not a monthly plan, while every other listed row is priced /mo. The AI humanizer, if you need one, is $2.00 per 1,000 words, rounded up to the next 1,000-word block, with prepaid word packs starting at $18.00 for 10,000 words that never expire.
Why the Index, Not the Percentage, Decides the Comparison
The percentage is an output; the index is the input, and only the input determines whether two numbers are comparable at all.
The practical consequence is that a gap between two numbers is not a discrepancy to resolve but a signal about which indexes were queried. Once you know a tool cannot reach the student-paper repository, its figure stops competing with Turnitin's and starts functioning as a lower bound — useful for spotting obvious matches, useless for predicting the professor-facing score. Treating the two as rival answers to one question is the error; treating them as measurements from different instruments is the correction.
FAQ
Is a 25% Turnitin similarity score bad?
No fixed number makes a score bad, because Turnitin's student guide states there is no set threshold and instructors set their own expectations [2]. Turnitin also says it "does not check for plagiarism in writing" and that the score is "simply the percentage of text in a submission that matches other sources" [1]. A 25% score falls in the Yellow band (25–49%) in Feedback Studio and Originality Check, but Turnitin notes colour schemes vary by integration and this does not affect the results [1]. Read the matched passages, not the headline number.
Why is my Grammarly or Quetext score different from Turnitin's?
Because the two tools search different indexes with different matching rules. Turnitin's database includes current and archived web content, periodicals, journals and publications, and a repository of previously submitted student papers [1][2]. A peer-reviewed article confirms Turnitin checks submissions "against its database of academic publications, internet, and previously submitted documents" [4]. Free checkers generally cannot query Turnitin's student-paper repository, so their percentage comes from a smaller pool and will not match.
Does a low similarity score mean my paper is original?
No. A low score only means little of your text matched the sources that particular tool searched. Turnitin advises using the report "as a tool within your review process to make your own determination if any academic misconduct is present" [1]. UC San Diego instructor guidance makes the reverse point too: "a high similarity score does not always suggest that a piece of writing has been plagiarized" [5]. A peer-reviewed source puts it plainly: a 64% similarity index "does not mean plagiarism" [4].
Why did my Turnitin score change between two runs of the same file?
Settings and timing change the number. Excluding small matches, such as a 10-word minimum, removes short hits [1]. Excluding previous submissions can drop a final draft from 100% once earlier drafts are removed from the comparison [1]. Resubmitted or overwritten papers "may not generate a new Similarity Report for a full 24 hours" [2]. A missing score can also mean your instructor turned off student access or the report is still generating [2].
Can I compare a Turnitin AI detection score to a similarity score?
No, they are different metrics measuring different things. The similarity score is the percentage of text matching other sources [1][3]. The AI detection report is a separate signal, and Turnitin shows *% instead of an exact percentage when AI detection falls below its 20% confidence threshold, marking those as low-confidence results. Comparing an AI percentage to a similarity percentage produces a meaningless gap, so keep the two reports separate when you review them.