Direct answer
A plagiarism scan is a text-matching process: your document is broken into strings of words and compared against a database of web pages, academic publications, and previously submitted student papers, and every match is returned with its source highlighted. The output is a similarity report, not a verdict — it shows what percentage of your text matched an existing source, and a human still has to judge whether those matches are properly cited [1]. Because the number is a match ratio rather than an integrity ruling, the useful skill is not running the scan but reading the report correctly before you submit.
How Does a Plagiarism Scan Actually Work Under the Hood?
Every plagiarism scanner, from a free online checker to an institutional Turnitin account, performs the same core operation: it normalises your text, slices it into overlapping word sequences, and looks for those sequences inside an index of previously seen documents. The index is the product. Turnitin's matching engine compares your submission against its student paper repository, archived and live web content, and scholarly publication databases, and the institutional administrator decides which of those repositories are switched on for a given assignment [2]. Two scanners given the same essay can therefore produce different percentages simply because their indexes differ in size and scope.
The matching is literal, not semantic. A scanner finds the same words in the same order; it does not reliably recognise that you paraphrased an idea into fresh vocabulary, and it does not know that a quotation is legitimate. That is why the report is built around sources rather than sentences: each detected match is grouped under the document it came from, with the overlapping text highlighted and a link to the original where one exists [2]. When you open a report, you are looking at a list of documents your writing resembles, ranked by how much text they share with you.
Turnitin also exposes exclusion controls precisely because the raw match count is crude. You or your instructor can exclude quotations, the bibliography, or small matches below a set word count, and the percentage recalculates as each category is removed [2]. Watching the number move as you toggle those switches is the fastest way to understand what the score is actually made of — and it is a far more honest reading than treating the headline figure as a grade. A draft whose similarity collapses once quotations and references are excluded is behaving very differently from one whose similarity survives every exclusion.
What Does a Similarity Score in a Plagiarism Report Actually Mean?
A similarity score is the proportion of your document's text that matched something in the index — nothing more and nothing less. Turnitin states plainly that a high similarity percentage does not automatically mean plagiarism, and a low percentage does not automatically mean the work is original [3]. A literature review that quotes three sources extensively and cites them impeccably can score far higher than a paper assembled from unattributed paraphrases, because the metric counts matched words rather than measuring intent.
Inflation comes from ordinary academic writing. Direct quotations, reference lists, standard methodology phrasing, module titles, and common collocations all register as matches, and small matches are sometimes filtered out entirely by a word-count threshold configured at the institutional level [3]. This cuts both ways: the same threshold that hides 40 incidental five-word matches can also hide a short but genuinely copied passage. A score of 18% is therefore not "18% plagiarised" — it is 18% of your word count appearing somewhere else first.
The practical reading is to ignore the headline number on first pass and work the source list instead. For each matched source, ask three questions: is this a quotation I marked, a paraphrase I cited, or text I did not intend to reuse? Only the third category is a problem, and it is a problem whether the overall score is 7% or 47% [3]. Students who fix citations source by source routinely end up with a cleaner submission than students who chase a target percentage by swapping synonyms.
Can Students Scan Their Own Draft With Turnitin Before Submitting?
Usually not through the institutional account. Turnitin access is provisioned by the university, and whether a student can submit a draft depends entirely on how the instructor configured the assignment — some set up draft-check folders or allow resubmission before the deadline, while others give students no self-service route at all [4]. Where a self-check option exists, it is normally limited to the specific assignment and may not permit checking unrelated work.
The gap matters because the moment you most want a scan is the week before the deadline, when a draft still exists as a file on your laptop rather than a submission in the gradebook. Checking early converts an invisible risk into a fixable one: you find the uncited paragraph while you still have time to paraphrase it and add the reference, instead of discovering it in a post-submission report that arrives with a misconduct referral attached [4]. It also removes the guesswork from revision, because you are editing against concrete highlighted sources rather than a vague feeling that a section "sounds copied."
Independent pre-check services fill that gap, and this is where turnitin0 differs from a generic free checker. Because it is not affiliated with Turnitin, LLC, it is not a substitute for your institution's account — but it does return the same two PDFs a professor sees in the LMS: a similarity report with matched sources highlighted, and an AI writing report, produced without adding your file to Turnitin's student paper database. Documents must be English, between 300 and 30,000 words, and under 20 MB; turnaround is under 15 minutes in 98% of cases, with delivery guaranteed within 30 minutes even in queue spikes. Pricing is pay-per-use — $3.80 for a single check, or $2.75 per check in a 10-check pack — with no subscription, and the service has delivered 100,000+ reports to 20,000+ students across the US, UK, Canada, Australia, New Zealand, and Ireland.
Reading the report is only half the job — the other half is seeing your own draft inside the exact report layout your marker will open, with every matched source flagged and every AI-suspected passage marked, before the deadline makes the result permanent. That is what turnitin0 is built for: a real Turnitin AI and similarity report on your file in minutes, so the corrections happen while they still count.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is a plagiarism scan the same thing as a plagiarism check by my university?
Not necessarily. The matching logic is similar, but the databases searched depend on how the account is configured, so percentages can differ between a third-party scan and your institution's report [2].
What similarity percentage is considered safe?
There is no universal threshold. Turnitin notes that high scores can be fully cited and low scores can still contain unattributed text, so the source list matters more than the number [3].
Does a scan detect paraphrased plagiarism?
Only partially. Scanners match word sequences, so heavy paraphrase into new vocabulary often escapes detection even when the underlying idea is unattributed [2].
Will scanning my draft add it to a student paper database?
That depends on the service. Institutional submissions generally do enter the repository, while turnitin0 checks without adding your file to Turnitin's student paper database.
Can I scan a document shorter than 300 words?
Not with turnitin0, which requires English documents between 300 and 30,000 words and under 20 MB. Very short texts also produce unreliable match ratios because there is too little text to compare [3].