"Make it smaller without making it worse" is the eternal document request. The trick is that size and quality are not really at war, most bloat lives in places you can trim without anyone noticing. Here is how to shrink files intelligently rather than bluntly.
Target the images, not the text
Text is tiny; images are where the weight lives. The smartest reduction re-encodes oversized images down to the resolution the page actually displays. A scan captured at 600 DPI shown at normal size loses nothing visible when brought to 200 DPI, but the file can drop dramatically. This alone solves most size problems.
Match the strength to the content
- Image-heavy or scanned documents tolerate stronger compression gracefully, there is a lot of redundant data to remove.
- Text-heavy documents need a gentle touch that preserves the crisp, selectable text layer. Aggressive, rasterising compression would turn readable type into a soft picture.
Remove before you compress
The cheapest size reduction is deleting what you do not need. Blank pages, duplicate scans, and appendices the recipient will not read all add weight. Trim them first, then compress what remains, you avoid spending effort shrinking pages you were about to remove anyway.
Clear the hidden weight
Files assembled from many sources carry duplicated resources and leftover editing data. A good optimiser tidies this internal clutter losslessly, real savings with zero visible change.
Stop at "good enough"
There is no prize for the smallest possible file, only for one that fits and still looks right. Start on a balanced setting, check the result, and stop the moment it is under your limit and still crisp. Chasing extra megabytes past that point only risks quality you did not need to spend.
