Thesis PDFs must fit upload portals with strict size limits — but they also have to stay sharp for reviewers. Learn how smart compression shrinks files while keeping text and figures clean.
Every graduate student knows the scene: the thesis is done, the formatting is perfect, and then the submission portal rejects the file because it is over the size limit. The PDF looks heavy because of the images inside it — screenshots of charts, scanned figures, high-resolution photos. The text itself weighs almost nothing.
Some PDF editors offer an "optimize" mode that only removes metadata and duplicate objects — useful, but limited. Real compression re-encodes the embedded images: each image is recompressed at a quality level that is visually indistinguishable from the original, while the file size drops dramatically. Text stays razor sharp at every compression level, because text is stored as vector data, not pixels.
A thesis built mostly from text and vector figures may already be near-optimal — the tool will tell you when nothing can be gained. A thesis packed with scanned pages or screenshots can often drop by 50–80% at the Standard level, with differences that are hard to notice even side by side.
Because compression only re-encodes images, the text layer stays fully searchable and copyable. Reviewers can still search the document, annotate it and print it at full quality. And since the whole process runs in your browser, the finished thesis never passes through a third-party server — important when your research may not be public yet.