The function seems to work in different ways depending on the document being added. My experience with it has been as below. The conversion doesn't seem to be necessary in the first instance, whereas it is in the second.
The issue with this is that getting the number of pages before the conversion is useful when trying to ensure that duplex printing occurs correctly. If there's an odd number of pages, a blank page PDF can be added to effectively pad the document
$client = new \Pdfcrowd\PdfToPdfClient('', '');
$client->addPdfFile('/path/to/adobe-created/pdf');
echo $client->getPageCount(); // Returns correct number
/****************/
$client = new \Pdfcrowd\HtmlToPdfClient('', '');
$client->convertURLToFile('URL', '/path/to/generated/pdf');
$client = new \Pdfcrowd\PdfToPdfClient('', '');
$client->addPdfFile('/path/to/generated/pdf');
echo $client->getPageCount(); // Returns 0