PDF uses (or can use) a number of compressions schemes for different element types (color, grayscale, and monochrome bitmaps as well as text and vector elements). It can also either embed, subset, or ignore the fints used in the docment you are converting. It sounds as though the compression hasn't been optimized (did you have any graphics in your test doc?) and/or the fonts (or a subset) have been embedded in your PDF.