What Is an Originality Report and How Should You Read It?

An originality report shows what percentage of your submitted text matches sources in a plagiarism checker’s database. It highlights specific passages, links them to their sources, and assigns an overall similarity score. A high score does not automatically mean plagiarism: properly cited quotes and common phrases can inflate it. Review each match individually before drawing conclusions.

What is an originality report?

An originality report is the output a plagiarism checker produces after scanning your text. It does three things: assigns a similarity score to the whole document, highlights the specific passages that matched external sources, and links each match to the source it came from.

The report doesn’t tell you whether you plagiarized. It tells you where your text overlaps with other text. You read the matches and decide which ones are problems.

Originality reports are generated by institutional tools like Turnitin and SafeAssign, by LMS-embedded checkers in Canvas and Google Classroom, and by standalone tools like Quetext. The report structure is similar across platforms: a score, colored highlights, and a source list.

What does my similarity score mean?

The similarity score is a percentage that reflects how much of your document matched sources in the checker’s database. A 20% score means 20% of your text registered as matching.

One important detail most students don’t know: the score is a weighted average across all matched segments, not a raw word count. A 20% score doesn’t mean 1 in every 5 words is copied. It means the weighted similarity across flagged passages totals 20%. A single long matching block contributes more to the score than a short one.

What the score includes by default:

• Exact matches to published sources
• Near-exact or paraphrased matches (in tools with fuzzy matching)
• Your reference list and bibliography, if not excluded
• Block quotes and properly cited passages, if not excluded

What it does not tell you:

• Whether any given match is intentional or accidental
• Whether a match is properly cited
• Whether a match is a common phrase or technical term with no meaningful origin

A 25% score on a research paper with 10 cited block quotes is a very different situation from a 25% score on a personal essay with no references. The number without the detail is close to meaningless.

What is a good originality score?

There is no universal threshold. Different institutions set different standards, and the acceptable range depends on the type of work.

General benchmarks used across academic institutions:

• 0 to 15%: Low similarity. Typical for original writing with minimal direct quotation.
• 15 to 25%: Moderate similarity. Common in well-cited research papers. Usually acceptable if matches are properly attributed.
• 25 to 40%: Elevated. Warrants a closer look. May reflect heavy quotation, missing citations, or paraphrasing without attribution.
• Above 40%: High. Most institutions treat this as requiring review. Doesn’t automatically mean misconduct, but it requires explanation.

One counterintuitive point: a 0% score is not always a good sign. A well-researched paper with proper citations will naturally show some similarity because it quotes and references real sources. A 0% score on a research assignment can indicate the writer didn’t engage with external sources at all, which may itself be a problem.

The score is a starting point for review, not a grade.

How do I read the color codes on my originality report?

Color coding tells you the type and degree of match at a glance. Different platforms use different systems.

Turnitin’s color bands indicate similarity percentage ranges across the document:

• Blue: no matching text found
• Green: 1 to 24% similarity
• Yellow: 25 to 49%
• Orange: 50 to 74%
• Red: 75 to 100%

These colors reflect the document-level score, not individual match severity.

Quetext’s ColorGrade system works differently. Instead of document-level bands, it color-codes individual highlighted passages by match type:

• Red: exact or near-exact match (90 to 100% similarity to a source)
• Orange: close or fuzzy match (paraphrased or structurally similar text)

The distinction matters in practice. A red highlight means the passage closely replicates a source and needs either a citation or a rewrite. An orange highlight means the passage is similar enough to warrant review: it may need a citation, or it may be a coincidental overlap with common phrasing. Quetext’s two-color system tells you which category each match falls into, so you know where to focus.

SafeAssign uses a percentage range per individual match, displayed in the report’s source list rather than in document-level color bands.

Why does my score say 25% when I only copied one sentence?

Because the score is a weighted average, not a word count.

If one matched segment is long (say, a 100-word block quote), it contributes more to the weighted average than a 10-word sentence. A single long passage can push the document score well above what the word count alone would suggest.

The reverse is also true: many short matches spread across a long document may produce a lower score than their frequency would imply, because each individual match is small relative to the document’s total length.

The most useful thing to do when a score surprises you is to open the matched passages individually rather than focus on the number. Quetext’s interactive snippet text viewer shows you the flagged text side-by-side with the matching source, so you can judge whether the match is a citation issue, a common phrase, or something that needs rewriting.

How Quetext’s ColorGrade compares to Turnitin’s color bands

Turnitin’s colors describe the overall score bracket the document falls into. They don’t differentiate between match types within the document. A document in the yellow band (25 to 49%) could have those matches distributed across a dozen properly cited quotes, or concentrated in two uncited paragraphs. The color alone doesn’t tell you which.

Quetext’s ColorGrade colors each individual match by type. Red passages are exact or near-exact: they closely replicate the source. Orange passages are contextually or structurally similar: they may be paraphrased, or they may share phrasing with the source without being copied. The distinction changes what you do next.

For exact matches (red): check whether the passage is cited. If it is, the match is expected and may be excludable. If it isn’t, add the citation.

For fuzzy matches (orange): read the passage against the source. If the idea came from the source, add a citation. If the overlap is coincidental phrasing or common terminology, no action is needed.

DeepSearch, Quetext’s detection engine, runs contextual analysis and fuzzy matching alongside exact string comparison. That’s why it surfaces orange matches that a keyword-only checker would miss entirely.

Originality reports in Canvas and Google Classroom

Canvas and Google Classroom both surface originality reports to students when instructors enable the feature.

In Canvas, the originality report is generated through a third-party integration, most commonly Turnitin, and appears directly in the submission view. Students see a similarity score and colored highlights using Turnitin’s color system.

In Google Classroom, Google’s own originality check tool scans submitted work against web content and a database of student papers. The report appears in the student’s assignment view and shows highlighted matches with source links. The feature is available to Google Workspace for Education accounts.

Both LMS reports work the same way as standalone checker reports: a score, highlights, and a source list. The interpretation is identical. If you received a report through Canvas or Google Classroom and want a more detailed breakdown before your instructor reviews your work, you can run the same document through Quetext’s free plagiarism checker for a sentence-level ColorGrade analysis.

Can I exclude my references from my originality report score?

Yes, and you should.

Reference lists, bibliographies, and properly formatted block quotes are expected to match their sources. Including them in the similarity calculation inflates the score without indicating any integrity issue. Most institutional tools offer exclusion options.

In Turnitin, instructors can configure the assignment to exclude quoted material and bibliography. Students can’t adjust these settings themselves, but they can ask instructors to enable them before submission.

In Quetext, you can exclude specific URLs from your results. If a high-traffic reference is inflating your score, removing it from the match list shows you the score without that source’s contribution.

What is worth excluding:

• Your full reference list or bibliography
• Block quotes with proper in-text citations
• Boilerplate text such as institutional headers or required assignment prompts

What is not worth excluding:

• Passages you suspect may be unintentionally paraphrased
• Uncited material you’re unsure about

The goal of exclusions is to see the score that reflects genuinely unattributed content, not to suppress the number.

How to fix a high originality report score

A high score has two possible causes: too many uncited matches, or too many legitimate matches (quotes, references) that haven’t been excluded. The fix depends on which it is.

For uncited matches:

Open each flagged passage. If the text restates or copies a source idea without a citation, add one. Quetext’s Citation Assistant generates APA, MLA, and Chicago citations from a source URL directly within the report, so you can fix the citation without leaving the tool.

Adding a citation often resolves the match without requiring any rewrite. The passage itself is fine; the attribution was missing.

For paraphrased matches (orange highlights in Quetext):

Read the flagged passage against the source. If the idea clearly came from the source, add a citation. If the passage is paraphrased so closely that it reads as the source’s phrasing in different words, rewrite it more independently, then cite it.

For reference list inflation:

Exclude the bibliography from the score calculation (see above) to see where your actual similarity sits.

For common phrase matches:

Some matches flag standard academic phrases or technical terms that appear across many documents with no real source. Review these individually: if the phrase has no meaningful origin, it’s a false positive and doesn’t need action.

A common mistake is deleting or rewriting highlighted text to lower the score when adding a citation would resolve the match without changing the argument. Rewriting costs you the substance of the point. A citation preserves it and removes the flag.

Common false positives: what inflates your score without meaning plagiarism

Several types of content reliably inflate similarity scores without indicating any integrity issue:

Properly cited quotations. A block quote with an in-text citation and a bibliography entry is not plagiarism. It will still appear as a match unless excluded.

Reference lists. Every source in a bibliography will match its indexed entry. This can add several percentage points to the score by itself.

Common academic phrases. Phrases like “this paper examines,” “the results suggest,” and “as shown in Figure 1” appear across millions of documents and will sometimes match. They carry no meaningful attribution requirement.

Technical terminology. Domain-specific terms used the same way across a field (“the p-value was below 0.05,” “using a double-blind randomized design”) will match without indicating copying.

Institutional boilerplate. Assignment prompts, course headers, or required formatting instructions included in a submission will match their source documents.

Reading each match type individually, rather than reacting to the total score, is the only way to separate real issues from noise.

What should I do if my originality report score is too high?

Work through it systematically. A high number is a prompt to investigate, not a conclusion.

1. Open each flagged passage. Don’t try to fix the score as a whole. Address each match individually.
2. Check whether the passage is cited. If the match is properly cited, it may be excludable. If it’s uncited, add the citation.
3. Check whether the match is a false positive. Reference list entries, common phrases, and technical terms can all be excluded from the calculation.
4. For paraphrased orange matches, assess the similarity. If your text closely follows the source’s structure or phrasing, rewrite the passage more independently and add a citation.
5. Exclude what should be excluded. Remove your bibliography and any properly cited block quotes from the score, then re-read the adjusted result.
6. Re-run the check. After making changes, run the document again to see whether the score has moved.

Quetext’s free plagiarism checker lets you re-scan as many times as you need. The ColorGrade breakdown updates with each scan, so you can track which passages you’ve resolved.

Does a 0% originality score mean my paper is perfect?

No. A 0% score means the checker found no text in your document that matched its source database. For a personal essay or creative piece, that’s expected. For a research paper, it can indicate a problem.

A well-researched paper will quote and reference sources. Those quotes and references will generate matches. A research paper with a 0% similarity score may have engaged with no external sources, cited nothing, or used sources that aren’t indexed in the checker’s database. None of those situations is necessarily fine.

Treat 0% the same way you treat any other score: ask what the paper actually contains, and whether that’s what it should contain.

How to generate and download your originality report in Quetext

After running a scan at quetext.com/plagiarism-checker:

1. Review your ColorGrade results in the editor view
2. Open each highlighted match to see the source and match percentage
3. Use the Citation Assistant to add or generate citations directly within the report
4. Use the interactive snippet text viewer to compare your text against the matched source side-by-side
5. Download the full originality report as a PDF (Pro feature) to share with an instructor or keep for your records

The downloadable report includes the similarity score, all matched passages with their sources, and the ColorGrade breakdown. It’s formatted for submission or documentation if you need to show an instructor the state of your work at a specific point.