Thanks patteo. This is interesting and I'm giving it a go now. My "workflow" for creating pdfs from ppts is to create a backup of my ppt and compress all of the graphics to webview (96dpi), print to FinePrint using 4-UP Landscape and then print to ScanSoft PDF Converter 4 Pro. The settings for Scansoft are usually set to webview as well. What's bizarre is that printing directly from Powerpoint to Scansoft using the handouts view (4 per sheet) will result in a significantly smaller pdf file than the source ppt file, printing from FinePrint results in a significantly larger pdf file than the source ppt file (I expect the FinePrint file to be larger than the powerpoint handout file, but not than the original ppt file!). Note, though, that there is no appreciable difference in terms of compression no matter what route I take to creating my pdf (i.e. if I simply create a pdf from my original, full sized ppt file resulting in the largest possible pdf I still see no more than 2% compression with PDF-DocPro). Anyway, I'm testing the script you linked to now.
On a related note, I wrote to the developer of PDF-DocPro, explaining that I have had no luck with their product compressing ANY PDF more than 2%, whether I've created the PDF at full compression, with none at all, or anything in between. They wrote back that the product will not compress a pdf file that is created by a PDF converter software that automatically compresses to a maximum level?!
This prompted me to download and run a trial of PdfCompressor 3.1 (the $300 Gorilla mentioned in jdds original post) and ran it on a PDF that was 9.72 MB in size. PDF-DocPro had compressed it 2% (to 9.42 MB) while PdfCompressor managed to squeeze it to 4.72 MB. Significant. However, it achieved this by compressing the graphics further, resulting in heavy pixelation... So, it *is* possible to further compress my PDF's, but I'm dealing with the law of diminishing returns here. Patteo's on the right track - the source file needs to be optimized.
The script is still running...