Dear Madeleine,
Would it be possible to get the scans resent to you? Unless they are scanned well (i.e. not at an angle, flat, clear text, etc.), then the character recognition is unlikely to work correctly. Looking at the PDFs you mention in your review, I can see they appear skewed and distorted. Did you receive the scanned files like that? It is possible they were scanned with the original pages presented to the scanner at an angle, or partially folded open from a bound copy?
(What kind of output did you get from the OCR program? A text file or Word document or somesuch? It seems the PDFs I saw in your review were uploaded with the content as images contained within a PDF, rather than converted text in a PDF.)
There are other OCR programs available, so it may be worth trying one of those. Many will allow you to upload the PDFs you receive from the Inter-Library Loan service, and then convert them online for you e.g.
http://www.free-ocr.com/
http://www.freeocr.net/
https://onlineocr.net/
https://ocr.space/
https://www.iskysoft.us/lp/pdf-editing/
http://www.paperfile.net/
https://www.wondershare.net/ad/pdf-editor/
Unfortunately, there’s not a lot else I can suggest. As mentioned, the reliability of the OCR conversion will depend mostly on the quality of your initial PDF files. (Conversion programs can vary, but generally will be of a certain standard – given decent input. One of the suggestions below should be sufficient in that case, and definitely worth trying before you look at paid solutions.)
Hope you have more success with an alternative program, but let us know if the problems persist. Feel free to send us a sample of the files you receive via the ILL service if you are still having issues with conversion, and we can check it and see how it converts for us. (Out support email address is EPPISupport@ucl.ac.uk.) It would also be useful if you could send us a sample of the OCR output.
Kind regards,
Zak Ghouze