There are two distinct processes that the software undertakes during operation; scanning for files and compressing files. When scanning, paths of supported files that haven't already been processed are added to a queue for processing. During the optimization phase, files are pulled from the queue and compressed.
The order in which the files are processed is detailed below:
General file processing order
1. There's a concept of "priority" which is based on the file extension.
2. The priority order is JPEG, PowerPoint, Word, PDF, Excel, TIFF, PNG then Zip.
3. Files are sorted by priority and then in descending order of size.
The software doesn't wait for the scan to complete before it starts optimising files.
So if it happens to find some Word documents before anything else, it'll probably process them first in descending size order as it finds them.
Over a large set of files, you'd expect the JPEGs to mostly be optimised first and in descending size order, followed by PowerPoint files etc. but the actual order of processing will depend on the order the scanner finds files and the relative speed of scanning vs optimisation.