How Does Turnitin Work? From Workflow to Text Scanning
You may see Turnitin in a course upload page and wonder what happens after you submit your file.
Turnitin works by comparing your writing with source databases, producing a Similarity Report, and, when enabled, showing instructors AI writing indicators. Those reports are meant to support review, not replace judgment.
This guide explains the workflow in plain language: what Turnitin checks, how reports are created, and what students should do with the results.

What Is Turnitin?
Let’s explore how Turnitin actually operates behind the scenes and delivers its services.
Turnitin typically works in partnership with educational institutions and is integrated into Learning Management Systems (LMS) such as Canvas, Blackboard, or Moodle. Because of this setup, most individual users—students or teachers—don’t create Turnitin accounts on their own. Instead, institutions provide access as part of their course management.
When a student submits an assignment through the LMS, Turnitin automatically receives the document and runs it through its detection system.
![]()
Besides these, Turnitin offers several other features designed to enhance the learning and teaching experience. These include grammar and spelling checks, inline feedback tools, and citation assistance—all depending on the institution’s specific setup.
How Does Turnitin System Detection Work?
Let’s see how Turnitin really works under the hood and how it actually works.
Turnitin usually partners directly with schools and is built into Learning Management Systems (LMS) like Canvas, Blackboard, or Moodle. Consequently, most individual users—students or teachers—do not set up Turnitin profiles. Their schools will issue login credentials as needed for managing courses.
When the student uploads an assignment to the LMS, it’s automatically uploaded to Turnitin, which checks it for plagiarism.
On top of that, Turnitin has a variety of other tools that aim to improve the teaching and learning process, like grammar and spell-check, inline feedback tools, and citations help—again, all part of the individual school’s usage.

How Does the Turnitin Similarity Report Work?
When an instructor decides to use Turnitin for a course or a specific assignment, they create a "Turnitin Assignment" within their Learning Management System (LMS). Once students submit their assignments through the LMS, the files are automatically sent to Turnitin for text-matching analysis.
Turnitin compares the submitted text against a vast database that includes internet content, academic journals, and previously submitted student papers. Based on this comparison, Turnitin generates an originality report showing a similarity percentage and highlighting the parts of the student's work that match other sources. Both instructors and students can access this report.
All submitted assignments are stored in a central repository, allowing Turnitin to compare future submissions — even those made years later — against previously submitted work to detect repeated or duplicate content.
Does Turnitin Detect Plagiarism?
No. Turnitin does not detect plagiarism. It only identifies matching text between the student’s paper and other sources such as websites, scholarly publications, or earlier student papers.

A high similarity score doesn’t automatically mean plagiarism has occurred. The report simply flags matched text, and it is up to the instructor to decide whether these matches are properly cited or indicate potential academic misconduct.
How Do Instructors Judge Turnitin’s Originality Reports?
Turnitin provides filtering options that let instructors exclude certain matches, like direct quotes and references, which can otherwise inflate the similarity percentage.
For example, a paper with many long, correctly cited quotations may show a high similarity score. Filtering out these quotations gives instructors a clearer view of the student's original writing. However, excessive filtering might leave the report showing mostly excluded quotations, making it harder to evaluate the student’s actual contribution.
How Does Turnitin Detect AI-Generated Writing?
Turnitin added AI writing detection in 2023 as schools began reviewing more assignments created with tools such as ChatGPT and other large language models.

When the feature is enabled and the file qualifies, Turnitin may generate an AI writing report for instructors. The report usually separates likely AI-generated writing from likely AI-paraphrased writing.
1. AI-Generated Content: How Is It Detected?
Turnitin reviews qualifying prose in overlapping text segments so that sentences are evaluated with nearby context rather than in isolation.
In simple terms, the model assigns probability-style scores:
Each sentence in context is scored between 0 and 1:
0 means the text looks more likely to be human-written.
1 means the text looks more likely to be AI-generated.
A middle value suggests uncertainty or mixed signals, not proof by itself.
Turnitin then combines the sentence-level signals into an overall estimate for qualifying prose. In the report, that may appear as an AI writing percentage.
For example, the report may estimate that a specific share of the document appears likely AI-written.
The highlighted sections help instructors decide where to review more closely, but they still need assignment context, drafts, and student explanation.

2. AI-Paraphrased Content: What Is It and How Is It Detected?
After likely AI-generated text is identified, Turnitin may evaluate whether that text also appears AI-paraphrased.
AI paraphrasing happens when generated text is reworded or rephrased by another tool while keeping much of the same structure, logic, or rhythm.
The report handles this as a second layer of review:
The system reviews AI-paraphrasing only in text already treated as likely AI-generated.
Those passages are evaluated by a separate AI paraphrasing model.
The model checks whether the likely AI text was rewritten by another AI tool rather than drafted directly.
This second layer also uses probability-style scoring:
A high score means the sentence is more likely to have been AI-paraphrased.
A low score means the model sees fewer signs of AI rewriting.
Together, the categories help instructors ask two different questions:
Does this passage look AI-generated?
And if so, does it also look AI-paraphrased?
Turnitin Can Get It Wrong Too
That said, Turnitin's AI detector has accuracy limits. There are instances of false positives, where real student work is identified as likely AI-generated.
If you or your professor think that Turnitin's AI score is wrong, here is what to consider:
🧾 What Can Cause False Positives?
A few writing patterns can make human writing look more AI-like:
Lists or bullet points with little sentence variety.
Repeated phrasing or paragraphs that mirror each other.
Heavy paraphrasing that changes wording but adds little original analysis.
Short documents or limited qualifying prose, where there is not enough context for a confident judgment.
📣 What Can You Do?
If you think the result is wrong, respond with evidence rather than panic:
Save drafts, outlines, version history, source notes, and any assignment planning.
If you can see the report, note the exact passages that were flagged and why they may look formulaic or heavily paraphrased.
For instructors especially, Turnitin's AI score should start a review, not end it. The percentage should be considered alongside the student's writing history, assignment requirements, drafts, and school policy. Human judgment remains essential when deciding whether a passage is original, AI-assisted, or simply written in a style the model misread.
How Do Other Turnitin Features Work?
Turnitin is widely recognized for similarity checking and AI writing reports, but it also includes tools for grading, feedback, draft review, and academic-integrity investigation. These features support the teaching side of the workflow, not just the detection side.
GradeMark™ – Streamlined Feedback and Grading
GradeMark™ is Turnitin's grading and feedback workspace. It lets instructors review a submission inside the browser and attach feedback directly to the student's text. Common uses include:
Inline Comments: Instructors can highlight a sentence or paragraph and leave targeted feedback.
QuickMarks: Reusable comments can be added quickly for common writing issues.
Rubrics: Educators can grade against published criteria for transparency and consistency.
General Comments: Instructors can leave overall feedback in text or voice form.
GradeMark supports a more organized feedback process, especially when many papers need the same rubric-based review.
Draft Coach™ – Real-Time Writing Support

Turnitin's Draft Coach™ is designed for work before final submission when a school enables it. Students may access it through supported writing environments and use it to review:
Similarity Checks: Identify matching text that may need quotation, paraphrase revision, or citation.
Citation Support: Review citation formatting and missing source details.
Grammar Suggestions: Improve clarity, structure, and readability before submission.
Draft Coach is most useful when students treat it as revision support rather than a last-minute score check.
Authorship Report – Investigating Academic Integrity
This feature provides metadata and writing analytics so you can try to deduce authorship when you suspect contract cheating or collusion. It shows:
Time stamps of document creation and edits.
Revision history insights.
Writing behavior patterns, like typing bursts or copy-paste events.
The Authorship Report is especially useful in higher-stakes academic integrity investigations.
Document Metadata and Analytics
Turnitin also captures all metadata associated with each submission—such as file upload date/times, document properties, and version histories—that institutions can use to investigate any suspect activity that suggests foul play, plagiarism, or non-original authorship.

FAQ
What is an acceptable similarity score on Turnitin?
There is no universal “acceptable” or good Turnitin similarity score—it varies by institution. Generally, scores under 15-20% are considered safe, but always check your school’s specific guidelines.
How does Turnitin tell if you’ve plagiarized?
Turnitin doesn’t decide if you plagiarized. It compares your text against a large database of web pages, academic papers, and prior submissions to highlight matching or similar content.
Is 30% similarity on Turnitin acceptable?
A 30% similarity score is relatively high and might raise concerns. However, some matches could be properly cited quotes or common phrases. The context matters, so review the detailed report carefully.
I didn’t plagiarize, so why is my Turnitin similarity so high?
High similarity can come from common phrases, well-cited quotes, references, or repeated technical terms. This doesn’t necessarily mean plagiarism.
How can I use Turnitin for free plagiarism detection as a student?
Turnitin is usually accessed through your school or institution. There is no official free individual version, but some schools provide student accounts or integrate Turnitin within their platforms.
How does Turnitin work? Do professors always know if you plagiarized?
Turnitin generates a similarity report for professors to review. It’s a tool to highlight potential issues, but professors use their judgment when interpreting the report.
Can Turnitin detect if I reuse my own previous work?
Yes, Turnitin checks submitted papers against a database of earlier student work, so reusing your own papers without citation can be flagged as self-plagiarism.
Can I check my work before submitting it to Turnitin?
Some institutions allow students to submit drafts or use tools like Draft Coach to check their work beforehand, but access depends on your school’s settings.
How long does it usually take for Turnitin to generate a report?
Turnitin report time varies widely. It can range from a few minutes to several hours depending on current workload, with initial submissions possibly taking longer.
A Quick Summary
Turnitin works by comparing submitted writing with source databases, generating a Similarity Report, and, when enabled, adding AI writing indicators for qualifying prose. It can show matches and likely AI signals, but it does not decide plagiarism or misconduct by itself.
Understanding the workflow helps students submit cleaner drafts, cite sources accurately, and respond calmly when a report needs explanation.