“Lossless” describes what a compression method preserves, not how dramatically it shrinks every file. MDN defines lossless compression as allowing the original data to be reconstructed exactly. That is a different claim from promising a particular file size.
Keep preservation and size separate
For a practical comparison, record the original size, compressed size, and method used. Calculate the reduction from those actual values rather than assuming every document or image will behave similarly. Already compressed data may offer little room for another compression step.
Also check what the workflow changes before compression. Resizing an image or deleting metadata is a separate operation. A tool using a lossless codec does not make every preceding edit reversible.
Verify the result you need
Open the output in the intended application and check that it still serves its purpose. For an exact file-preservation task, appropriate integrity checks can compare recovered data with the original. For a web image, dimensions and visual suitability also matter.
The useful question is not simply whether the file became smaller. Ask what was preserved, what was changed, and how the result was checked. That makes compression settings easier to discuss without confusing a technical preservation property with a guarantee of large savings.