A TIFF is a container, not a single picture. It can hold one image or hundreds, in any of six common compression schemes, and it records the DPI the scanner used. Converters that read only the first image, or that re-encode a black and white scan as JPEG, are throwing away pages and sharpness that were there in the original.
One file, many pages
Most image formats hold exactly one image. TIFF does not. The format was designed in 1986 around a list of image directories, and a file can carry as many of them as it likes. That is the reason it survived: when a document scanner takes in a forty page contract, it can write one TIFF with forty images inside rather than forty separate files, and when a fax arrives the whole transmission lands as a single file.
This is also where most conversions go wrong. A converter that treats TIFF as a single image reads the first directory, ignores the rest, and hands back a one page PDF. Nothing warns you. The file downloads, it opens, it looks correct, and pages two through forty are simply gone.
If you have converted TIFFs elsewhere and kept the PDFs, open one and check the page count against the original. A multi-page TIFF that came back as a single page was silently truncated, and the original is the only place those pages still exist.
What the compression tells you about the file
TIFF does not define one compression method, it allows several, and which one a file uses tells you what kind of document it holds.
- CCITT Group 3 and Group 4. One bit per pixel, black or white, nothing in between. This is fax and it is most document scanning. Group 4 is the more efficient of the two and dominates.
- LZW. Lossless, general purpose, common in files that came out of design and imaging software.
- PackBits. A simple run length scheme, old, still found in files from Apple software and older scanners.
- Deflate. The same compression as ZIP. Lossless, efficient, common in newer files and in anything Adobe wrote.
- Uncompressed. Exactly what it sounds like. Large, and usually a sign the file came out of a microscope, a scientific instrument or a very old workflow.
The one that causes trouble is an obsolete variant from the early 1990s that stores an old style JPEG inside the TIFF. It was replaced almost immediately and almost nothing writes it any more, but files from that era still circulate. This tool names it rather than failing with a generic error, so you know to re-export rather than assume the file is corrupt.
The DPI question, and why your page comes out the wrong size
A TIFF records the resolution it was scanned at. A standard fax page is 1728 pixels across at 200 DPI, which works out at 8.64 inches, close enough to a sheet of US Letter. That number is written into the file.
A converter that ignores it has to guess, and the usual guess is to stretch every image to fill A4. A page scanned at 300 DPI and a page scanned at 200 DPI then come out the same size on screen even though the originals were different, and anything you measure off the page is wrong. That matters more than it sounds for scanned plans, forms and anything that will be printed at scale.
Read the resolution tags
XResolution, YResolution and the unit, straight out of the file's directory.
Divide pixels by DPI
1728 pixels at 200 DPI is 8.64 inches. That is the page width, not a guess.
Convert to PDF points
PDF measures in points, 72 to the inch, so 8.64 inches becomes 622 points.
Fall back only when it has to
Some files record no resolution at all. Those get A4, and the tool says so rather than pretending.
Why a converted fax looks worse than the original
This is the failure people notice, because it is visible. A fax is one bit per pixel. Every pixel is either black or white, the edges of the letters are exact, and the file is small because there is so little to store.
JPEG was designed for photographs. It works by discarding the high frequency detail the eye is least likely to miss, which is precisely the sharp black to white transition that makes text readable. Run a fax through it and the letter edges pick up grey fringing and faint blocks, thin strokes soften, and the file often ends up larger than the bilevel original because JPEG has nothing to exploit in an image with two colours.
Re-encoded as JPEG
Grey fringes around every letter, blocking in the white space, thin strokes softened, and frequently a bigger file than the original. Irreversible: the detail is gone from the PDF.
Embedded losslessly
Every pixel exactly as scanned, edges still crisp at any zoom, and a small file because two colour images compress extremely well. What this tool does when it detects a black and white page.
What this tool does
It expands every image directory in the file, so a forty page TIFF gives you forty PDF pages and the count is shown before you convert. It reads the resolution tags and sizes each page to the paper the scan came from. It checks the bit depth of each page, and where a page is black and white it is embedded without loss, while photographic pages are embedded as JPEG at the quality you pick. Files with several TIFFs dropped at once are joined in the order shown, and you can remove any of them first.
All of it runs in the browser. That is worth saying plainly for this format in particular, because TIFF is what medical imaging, legal discovery and archive scanning tend to produce, and those are exactly the files that should not be uploaded to a stranger's server to save a step.
Questions people actually ask
What people ask with a folder of scans and faxes.
01My TIFF has 12 pages but the PDF has 1. What happened?
The converter read only the first image directory. This is the single most common TIFF conversion fault and it fails silently. Convert the original again with something that reports a page count before it builds the file.
02Can I convert the other way, PDF to TIFF?
Not on this page. If what you want is an image out of a PDF, the PDF to Images tool does PNG and JPEG, which is what almost every destination actually wants now.
03Why is my converted file so large?
Usually a photographic or greyscale scan at high DPI, where there is genuinely a lot of information. Try the Small quality setting. If the pages are text, check whether the source is actually greyscale rather than black and white, because a greyscale scan of a text document is several times the size for no readability gain.
04Will the text in the PDF be searchable?
No, and it was not searchable in the TIFF either. A scan is a picture of text, not text. Run the result through OCR to add a searchable layer underneath the image.
05Does it handle TIFFs from a microscope or a satellite?
Often, but that corner of the format uses tags and bit depths that general purpose decoders skip, and 16 bit or floating point scientific TIFFs are outside what this tool claims to do. If a page comes back looking wrong, that is the likely reason.