Before you start: a few things to know 

  • Use the institutional login route. Always access iThenticate through the UM Library link, not through the public iThenticate website. Packages purchased directly on the iThenticate website are individual credit subscriptions that do not include AI Writing Detection and are not part of UM’s licence. 
  • Submit complete, near-final texts where possible. Uploading multiple partial versions of the same document may affect the reliability of the results and unnecessary draws on the maximum number of submissions. For PhD theses in particular, uploading the full text as a single document is recommended.
  • Check that your file type is supported. iThenticate accepts common document formats, including Word (.docx) and PDF. Check the iThenticate guide for a current list of supported file types.
  • Your documents are not shared. Unpublished manuscripts submitted to iThenticate through UM’s institutional account are stored only in your private account. They are not added to a public or shared database and are not accessible to other users, publishers, or members of the public.

Quick guide

Step 1. Log in using your UM credentials. Click Launch to start!

Step 2. In your folders area, select or create a folder for your document. 

Step 3. Click Upload and follow the prompts to enter document metadata and upload your file. 

Step 4. Wait for the report — generation can take a few minutes, depending on document length and server load. 

Step 5. Open the Similarity Report and review each highlighted match in context. Use the source list in the report to identify the matched sources. 

Step 6. If AI Writing Detection is enabled for your account, open the AI Writing tab within the report to review any AI writing indicators separately. 

User guides

Detailed user guides can be found on the iThenticate website.

Go

Options for exclusion

There can be various reasons why you may want to exclude certain sources that your document is compared to or certain parts of your document in the similarity check. You can specify options for exclusion in the Settings.

For exclusion options within a similarity report please see this guide.

Interpreting your results responsibly 

A useful general approach when reviewing a Similarity Report: 

  • Start with the match breakdown, not just the overall percentage. Look at which sources are being matched, how much of each match is from that source, and whether the matches cluster in particular sections. 
  • Exclude what should be excluded. The report allows you to exclude quoted text, reference lists, and small matches; use these options thoughtfully so the figures you act on are meaningful. 
  • Consider your genre and discipline. Review articles will typically show higher similarity than empirical papers. A thesis with a substantial literature review may receive a different score from a short methods paper. 
  • Discuss with a supervisor, co-author, or the Library if you are uncertain about what a result means or what to do with it. 
  • For the AI Writing indicator, flag any passages with a high AI writing score for close reading. Consider whether the writing in those passages accurately reflects the research or needs to be revised for clarity, originality, or attribution.