I have recently purchased the Pro version of this application. Upon testing I have found the following issues  
 
a) When testing the same uploaded MS word document on four occasions the resulting reports were markedly different - why?
 
b) After making a self compiled plagiarised report from 5 different websites. Only one of these "copied & pasted" paragraphs was detected the others were coded GREEN - Why?  
 
This is way lower than the 97% accuracy rate described. No settings were changed and the performance level was set at "advanced check" for each check. Can this be explained rationally or is the program defective?
 
As a school Principal I have to be certain of the credentials of this program - at present it has a very big question mark over its capabilities!!!!!!!!!!

I would prefer a refund to my credit card please. This program cannot be viewed as reliable for its stated purpose when one student's work gets four different results, and one purposfully plagiarised document does not have the content detected as plagiarised (despite the fact that it was copied and pasted directly from a website). When assessing student plagiarism we need to be as fair as possible. At present this lacks consistency, and therefore fairness - in education this is not acceptable. Additionally, we do not have the time to forward individual student files for checking.

Name = Tony Fynn (Please refund ASAP).

E-mail provided in the Private Meassafe on this forum. We will contact you about the refund shortly after we receive the documents.
 
Thank you for your patience.

Hello,
 
I've issued a refund per your request.
 
Secondly - judging by the values on your previous post - I'm afraid that you do not quite understand Plagiarism Detector
logics.
 
Apart from marking Plagiarism Plagiarism Detector also marks "referenced" part. Any SIMILAR part is considered
referenced if at least one hyperlink points to the source. In your experiment "61% quoted" means that you HAVE NOT removed
the hyper links from the text. THAT IS WHY Plagiarism Detector shows 0% Plagiarism and 61% quoted.
 
In case you remove the hyperlinks it will show 61% Plagiarism or even more (PD ha a sophisticated method to heuristically detect "linked" fragments)
 
Unfortunately I have not received the e-mail with your test document  .
 
The last thing to say - Plagiarism Detector may show DIFFERENT percentage on the same source and there IS a good reason for this -
Every time a request goes to the Search Engine this is an asynchronous reuest - that is they return in different time,
thus the final similarity matrix may be SLIGHTLY different at each rescan.
 
Theoretically, if for some reason a critically important source fails (due to the request timeout or a search engine timeout)
this may cause the whole document to be reported as 100% green.
 
Still - this is a general plagiarism detection issue - it is not a bug in any respect. In case you find any program doing it  
better - feel free to post a note here - our RnD and I will have a closer look at it.
 
By the way - currently our RnD is working at the high precision cache mechanism that will
slightly aid the problem of missed out source, but I repeat that this is more a theoretical problem than a technical bug.
 
So far I believe that Plagiarism Detector is the best software available.
Good luck!