Why Is My PDF Lagging When I Scroll? A Measured Fix
Why is my PDF lagging when I scroll? Scanned pages are big images your viewer redraws constantly. We measured what actually shrinks them, and what does nothing.
Why is my PDF lagging when I scroll? Almost always because each page is a large photograph rather than text, and your viewer has to decode that photograph every time the page comes into view. Scanned documents are the usual culprit, and the giveaway is that a much larger text document on the same machine scrolls perfectly.
The fix is to reduce how much image data sits on each page. What is worth knowing before you try is that the obvious safe-looking option does nothing at all here, and we measured that rather than guessing.
Pixels cost you, megabytes do not
A text PDF stores instructions: use this font, place these glyphs at these positions. Drawing a page from instructions is cheap, which is why a three hundred page report can be small and scroll instantly.
A scanned PDF stores a picture of each page. At 300 DPI, an A4 page is roughly 2481 by 3507 pixels, which is about 8.7 million pixels for one page. Your viewer decodes that image, scales it to your window, and throws it away when you scroll past. Do that repeatedly on a phone or an older laptop and the stutter you are seeing is the decode cost, not the download.
This is why file size is a poor guide. What matters is pixels per page, and the way to fix scroll lag is to reduce that number.
First, confirm it is actually a scan
Everything below assumes your pages are images. That is worth confirming in a few seconds, because if the document is real text then re-rendering it is the wrong move and will cost you selectable text for no benefit.
The quickest check is to try selecting a word on the page. If your cursor highlights text, the page is text. If it draws a rectangle over the area instead, the page is a picture. For a second opinion that does not depend on your viewer, drop the file into PDF to Text: it reads the text layer directly, so words coming back means there is real text in the file and nothing coming back means the page is an image.
Two useful outcomes. A document that returns text and still scrolls badly has a different problem, covered further down. A document that returns nothing is a scan, and the measurements below apply to it.
What we measured, including the setting that did nothing
We built a twelve page scanned document at 300 DPI, put it through the live compressor at two different settings, and inspected the resulting files rather than trusting the labels.
| Version | Size | Page image | Megapixels |
|---|---|---|---|
| Original | 4.29 MB | 2481 × 3507 | 8.7 |
| Lossless setting | 4.29 MB | 2481 × 3507 | 8.7 |
| Balanced setting | 1.01 MB | 1191 × 1684 | 2.0 |
The lossless setting changed nothing that matters. Same pixels, and a size difference too small to see. We ran it on three scans: ours came back at the same 4.29 MB, a second shed 576 bytes on 2.5 MB, and a third shed 578 bytes on 3.5 MB, which is 0.023 and 0.016 percent. The two nonzero deltas are nearly the same byte count even though the files differ by a megabyte, which is what you would expect if only the structure is being rewritten. Expect no useful change rather than a byte-for-byte copy. That is not a fault, it is what lossless means: it re-packs the document structure without touching the page images, and a scan is already stored as compressed photographs, so there is nothing left to squeeze.
The balanced setting cut the file by about 77 percent and, more to the point, cut each page from 8.7 megapixels to 2.0. That is roughly four times less image data to decode per page, which is the number that governs scrolling.
Do not reach for lossless on a scan
It sounds like the careful choice and it is the correct one on a text document you need to keep selectable. On a scan it gives you back a file of effectively the same size, which reads as the tool having failed when it has done exactly what it says.
Which setting to pick
- Open the file in Compress PDF. The page lists each level with what it does, including the resolution it re-renders at, so you are not guessing.
- If the document is a scan, skip lossless. Across the three test scans above, the largest reduction was 0.023 percent.
- Start with the balanced level. On our test that was the 77 percent reduction, and small print stayed readable.
- Check the result before deciding you need more. Open the output and read the smallest text on the busiest page. If it is clear, stop.
- Only then consider the stronger level. It re-renders lower again, which shrinks further and costs legibility. A scan you cannot read is a worse outcome than one that scrolls slowly.
When the images are not the problem
Two other causes are worth ruling out before you change the file, both cheap to check.
Page count. Some viewers slow down on very long documents regardless of what is on the pages. If you only need part of it, Split PDF gives you a shorter file to work in, and Extract Pages pulls out a specific range.
Your viewer rather than your file. Open the same document somewhere else, a browser tab if you were in a desktop reader or the reverse. If it scrolls smoothly there, the file is fine and the software is the problem. This takes ten seconds and saves you re-rendering a document that never needed it.
What this will and will not promise you
We measured file size and pixels per page, both of which we can state precisely because we inspected the output files. We did not measure frames per second on your device, and no honest guide can, because that depends on your hardware, your viewer and how much memory it has free.
What we can say is that the most common cause of PDF scroll lag is the amount of image data per page, that the balanced setting reduced it about fourfold on a representative scan, and that the lossless setting reduced it not at all. Whether that is enough on your machine is something you will know within a few seconds of opening the result.
Why this runs in your browser
The documents that scroll badly are usually scans, and scans are usually the sensitive ones: bank statements, contracts, identity papers, medical records. Uploading one to an anonymous server to make it scroll better is a poor trade.
The tools here process files inside your own browser tab. The PDF is read by JavaScript on your device, the re-rendering happens in local memory, and the smaller file is written straight back to your downloads folder. Nothing is uploaded, so there is no server copy of a document you only wanted to make easier to read.
Your files never leave your browser
PDF Mavericks processes everything locally. No file is uploaded to any server.
Frequently asked questions
Why is my PDF lagging when I scroll but a bigger PDF is fine?
File size is a poor predictor of scrolling smoothness because what costs your viewer effort is pixels per page, not megabytes. A three hundred page text document can be tiny and scroll instantly, since each page is drawn from font instructions. A twelve page scan can be far heavier to move through, because every page is a photograph that has to be decoded before it appears.
Does lossless compression fix a slow scanned PDF?
No, and we measured this rather than assuming it. Running a twelve page scan through the lossless setting returned the same 4.29 MB with exactly the same page images, 2481 by 3507 pixels before and after; two further runs on different scans shaved 0.023 and 0.016 percent, which is noise. Lossless compression squeezes the document structure, and a scan is already stored as compressed photographs, so there is nothing left for it to squeeze.
How much does re-rendering actually shrink a scan?
On our test document the balanced setting took a 4.29 MB file to 1.01 MB, a reduction of about 77 percent. More importantly for scrolling, each page image went from 2481 by 3507 pixels to 1191 by 1684, which is 8.7 megapixels down to 2.0. That is roughly four times less image data for the viewer to decode on every page.
Will compressing make my PDF scroll smoothly?
It removes the most common cause, but we cannot promise a specific result on your machine. What we measured is the input to the problem, which is how much image data sits on each page, and that we reduced by about four times. How that translates into smoothness depends on your device, your viewer and how much memory it has available.
Does re-rendering lose the text I can select?
On a scan there is usually no selectable text to lose, because the page is already an image of words rather than words. On a document that does have real text, re-rendering converts the page to an image and the text stops being selectable. That is the trade, and it is why the lossless setting exists for documents where selectable text matters more than size.
Is scroll lag ever caused by something other than image size?
Yes, and it is worth ruling these out before changing the file. A very high page count makes some viewers slow regardless of what is on the pages, and a document with many form fields or annotations has extra work to do on every redraw. Opening the same file in a different viewer separates a file problem from a software one in about ten seconds.
Should I use the highest compression setting for speed?
Not automatically. The stronger setting re-renders pages at a lower resolution again, which shrinks them further but makes small print harder to read, and a scan you cannot read is worse than one that scrolls slowly. Start with the balanced setting and only go further if the document is still heavy and the text survives at the smaller size.
Does compressing a PDF upload it anywhere?
Not with this tool. The file is read by JavaScript inside your own browser tab, the re-rendering happens in local memory, and the result is written back to your downloads folder. That matters for the documents people most often need to shrink, which tend to be scanned statements, contracts and identity papers.