PDF guide
Compress a PDF while keeping the features you need
By MultiZest · · 4 min read
A PDF can contain text instructions, images, embedded fonts, annotations, forms, and signatures. Two documents that look similar may store their pages very differently. Compression is therefore a choice about what information you are prepared to change.
Find out whether the page contains real text
Open the PDF and try selecting a sentence. Search for a word you can see. If selection and search work, the document likely has a text layer. A scanned page can also have an OCR layer, so looking like a scan does not necessarily mean it is only an image.
Decide which features matter before compressing. A casual visual reference might only need legible page images. A searchable report, accessible document, fillable form, or signed agreement can need much more. File size is not the only acceptance criterion.
Start with structure optimization
The structure option rewrites the PDF’s object storage without intentionally rasterizing its text into page images. It is the sensible first attempt when text selection matters. It may create only a small saving if the document was already stored efficiently.
A larger output is not a successful compression. MultiZest compares sizes and keeps the original when its generated copy is larger. Review the result even when it shrinks: rewriting a PDF can still affect document-level features such as forms or signatures.
Use page-image compression for the right scan
The image method renders each page and places the rendered image into a new PDF. Lower render scale and JPEG quality can reduce the data used by large scans. The tradeoff is that page content becomes a picture.
Selectable text, interactive links, form fields, annotations, and existing signature features are not preserved by this method. It does not perform OCR afterward. If those features are required, use another PDF workflow that can compress images while preserving the necessary document structure.
Check a representative set of pages
Do not judge a fifty-page scan only by its clean first page. Background noise and faint printing can make later pages much harder to compress.
- Inspect a page with the smallest text.
- Look at a page containing a chart, stamp, or fine line work.
- Check the darkest scan and the lightest scan.
- Search and select text again if those features matter.
- Confirm the page count and page order.
- Check the actual downloaded size against the receiving system’s limit.
If the file still exceeds the limit
Ask whether the recipient accepts separate files. Extracting only the required pages can remove unnecessary content without lowering the quality of the pages you keep. If every page is required, lower scale or quality gradually and recheck readability.
There is no universal “compress to 1 MB” setting. The attainable size depends on page count, dimensions, content, and required quality. Keep the original and avoid overwriting a legally or operationally important document with a flattened copy.
Choose a difficult page as your quality sample
A cover page usually tells you little about compression damage. Inspect the densest table, smallest footnote or faintest stamp instead. If those remain readable, then review the rest of the document for layout and interactive requirements.
Check normal reading size as well as close zoom. A recipient should not need extreme magnification merely to read an ordinary paragraph in an email attachment.
Consider a smaller selection before harsher compression
If only one section is required, extracting it can reduce the attachment without lowering the quality of every page. Include the headings, units and notes needed to interpret the selected information.
Keep the full original and confirm that extraction is appropriate for the request. A smaller distribution copy should remain an understandable document, and it should not be confused with a signed or interactive original that serves another purpose.