How you actually shoot
Drop a folder of photographs and see which focal lengths, apertures and ISOs you really use — read from the files themselves, on your own machine.
Analyse a batch of photos
Your photos never leave this device
Focal length
Lenses
By number of frames.
Cameras
By number of frames.
Aperture
The f-numbers recorded in the files.
ISO
Shutter speed
Grouped by what each range does to motion rather than by value.
Time of day
From each file's capture timestamp, in the camera's own local time.
What is read, and what is not
Each file is opened in the browser and its first 256 KB is examined for an EXIF block. That is where EXIF lives — immediately after the start-of-image marker — so reading further would cost memory without finding anything. The image data itself is never decoded.
Read: camera make and model, lens model, focal length, aperture, shutter speed, ISO, capture timestamp, orientation, pixel dimensions, and whether GPS coordinates are present.
Not read: the picture. No thumbnail is decoded, no pixels are examined, and nothing about image content is derived. Filenames are used only to report which files could not be parsed, and are not included in any figure.
Nothing is transmitted. No file, no metadata, no aggregate, and no analytics event carrying any of it. The page makes exactly the same network requests whether you analyse four hundred photographs or none.
Formats: JPEG and TIFF, including DNG, which is a TIFF. Most proprietary raw formats are TIFF-based and often work; HEIC does not, because its metadata sits in a container the browser cannot open without decoding the image.
What this is actually for
Most people's sense of how they shoot is wrong, and wrong in a consistent direction: we remember the unusual frames. A photographer who owns a 70–200mm remembers the shots at 200mm and forgets that four fifths of their year was made at 35mm. The distributions below are the correction — not opinions about what you should do, just a count of what you did.
The most common use is before buying a lens. If the histogram piles up at one end of a zoom's range, a prime there may serve you better and faster. If it is spread evenly, it may not. Either way it is a better basis than the review you were about to read.
Reading the histograms honestly
Sample size decides everything. A hundred frames from one afternoon describes that afternoon. A few thousand across a year describes you. Every figure here reports how many files it came from, and below about twenty-five the page says so rather than drawing conclusions.
Selection bias runs through it. These are the frames you kept — or worse, the frames you happened to have in one folder. If you deleted the failures, the distribution describes your successes, which is a different thing from your habits.
A zoom's ends are over-represented. Almost every zoom histogram spikes at both extremes, because people push to the stop rather than choosing a number. A spike at 200mm on a 70–200mm often means "as long as it goes", not "200mm was right".
What it cannot see
The pictures you wanted and did not get. If you shoot at 24mm constantly because your widest lens is 24mm, the histogram shows a preference that is really a constraint — and no metadata can tell the two apart. That is why nothing here recommends buying or selling anything: the tool can count what you did, but the interesting question is what you could not do, and the files are silent on it.
Related
Found GPS you did not expect? The EXIF viewer and remover strips it. To act on what the distributions show, try the field of view calculator and the depth of field calculator.
Questions people actually ask
Are my photos uploaded?
No. Each file is opened by the browser, its first 256 KB examined for an EXIF block, and the result kept in the page. Nothing is sent — not the files, not the metadata, not the summary. The page makes the same network requests whether you analyse four hundred photographs or none.
Why do some of my files show no metadata?
Usually because it was stripped. Messaging apps, social platforms and many editors remove EXIF on export, and screenshots never had any. Some cameras also omit lens data when a manual or adapted lens is fitted, since they have no way to identify it.
Does it work with raw files?
TIFF-based raws generally work, which includes DNG and many manufacturers' formats. HEIC does not: its metadata sits in a container the browser cannot open without decoding the image, which would mean pulling every file fully into memory. JPEG and TIFF are the reliable cases.
How many photos do I need for this to mean anything?
A few hundred across several different outings. A hundred frames from one afternoon describes that afternoon, not your habits. Below about twenty-five files the page says so rather than drawing conclusions from the distribution.
Will it slow down or crash with thousands of files?
It should not. Only the head of each file is read, the buffer is released before the next one, and the loop yields to the browser regularly so the page stays responsive and the cancel button keeps working. Memory use is roughly flat regardless of batch size.
Should I buy the lens my histogram suggests?
This page will not tell you, deliberately. It can count the focal lengths you used, but not the photographs you wanted and could not take. If you shoot at 24mm constantly because 24mm is as wide as you own, the histogram shows a preference that is really a constraint — and the metadata cannot tell those apart.