A Google Photos export through Google Takeout often contains duplicates because a photo that sits in an album is exported twice: once in its “Photos from [year]” folder and once in the album’s folder. Photos you edited in Google Photos add an extra version with “-edited” in the name. You can avoid most of this by exporting only the year folders, but if you already have a doubled export on your PC, DupeDoctor from Automata Labs finds the byte-for-byte copies by content, keeps at least one of every set, sends extras to the Recycle Bin, and can replace album copies with hardlinks so every folder keeps its photo without using the space twice.
Find the doubled photos in your Takeout export by content, and reclaim the space safely.
Get it from Microsoft
Why does Google Takeout create duplicate photos?
Takeout exports Google Photos as folders. There is one folder per year, named like “Photos from 2020”, which together hold every photo in your library. There is also one folder for each album you created. An album is a collection that points to photos in your library, but in the export each album folder gets its own copies of the files.
A Google Photos Community answer in this Takeout thread puts it plainly: “Note that if you have albums, two copies of those photos will be downloaded.” The same answer explains that the year albums “contain all the master” photos, and that ticking both the year albums and your own albums “will output duplicate copies.”
That is why an export can be much larger than your Google storage figure suggests. Every photo that sits in two albums can appear three times.
What else is in a Takeout export?
Another community answer, in a thread about JSON files without images, lists what you get when you select only one set of folders: “a single copy of each photo, one json file per photo, and one filename-edited.jpg for each” edited photo. A later thread adds two practical points:
- The JSON files contain the date, location and descriptions added in Google Photos. You can delete them if you do not need them, but keep them until you have decided, because tools that restore dates read them.
- Extract every zip to the same folder. Large exports are split across many zips. Each batch needs to be extracted into one folder for the structure to be complete.
These are community answers rather than official Google documentation, but they match what most people find when they open an export.
How do I export from Google Photos without duplicates?
The community steps for a new export:
- Go to Google Takeout and deselect everything.
- Select Google Photos.
- Click the button that reads All photo albums included and deselect all.
- Select either all the year albums (“Photos from 2020” and so on) or the albums you created, not both.
- Create the export and download every zip.
- Extract all of the zips into the same folder.
Choose the year albums if you want everything. Choose your own albums only if every photo you care about is in one of them, because photos outside an album will be left out.
How do I clean up an export that already has duplicates?
If you already downloaded everything, you do not need to export again. You need to find which files are true copies and decide what to keep. There are three kinds of “duplicate” in a typical export, and they need different handling:
| What you see | What it is | Safe to remove? |
|---|---|---|
| Same photo in “Photos from 2020” and in an album folder | A byte-for-byte copy | Yes, if one copy is kept, or replace with a hardlink |
photo.jpg and photo-edited.jpg | Your original and the version you edited in Google Photos | Only if you are sure which you want. They are different files |
photo.jpg.json and similar | Metadata sidecar with dates, location and descriptions | Keep until you have restored dates or decided you do not need them |
The first kind is the space hog. The second kind looks like a duplicate but is not. Treating them the same is how people lose their edits.
How do I clean a Takeout export with DupeDoctor?
DupeDoctor is an offline duplicate file finder for Windows 11. According to its Microsoft Store listing:
- Exact duplicates are confirmed by content, not by name or size. A size grouping prunes non-matches, then a BLAKE3 hash verifies true byte-for-byte copies, so renamed files are still caught.
- Look-alike photos are found with perceptual hashing, catching the same image at different resolutions, crops or re-encodes. They are never auto-selected, because similar is not identical.
- Deletion always goes to the Recycle Bin. There is no permanent-delete path.
- At least one file in every duplicate set is always kept. The app will not let you empty a group.
- Hardlinks can replace redundant copies on exact-duplicate sets, reclaiming the space while keeping a file in every location.
- Move instead of delete, to a folder of your choice.
- Auto-keep rules such as keep newest, oldest or shortest path pre-select the redundant copies in exact sets.
- A CSV or JSON report of everything found.
- No network client, no telemetry and no account.
Keep one copy of every photo, keep your album folders, and keep your edits.
Get it from Microsoft
A step-by-step cleanup for a Takeout folder
- Extract everything first. Put all the zips into one folder, as above. A half-extracted export gives misleading results.
- Scan the whole Takeout folder. Point DupeDoctor at the folder that holds the year and album folders.
- Review exact duplicate sets. Each set shows the paths, so you can see the year-folder copy next to the album copies.
- Choose a keep rule. Pick newest, oldest or shortest path, whichever keeps the copy you want, then check a few sets to confirm it picks the one you expect.
- Decide: delete or hardlink. If you want album folders to keep working as folders, replace the extras with hardlinks. If you only care about the year folders, send the extras to the Recycle Bin.
- Leave the “-edited” files alone unless you have looked at them. They are not exact copies, so if they appear at all it is as look-alikes, and DupeDoctor never auto-selects look-alikes.
- Export the report as CSV for a record of what changed.
- Empty the Recycle Bin later, once you have checked the result.
If you are moving the photos into another library, our guide to cleaning up EXIF data in a Google Photos export covers the dates and metadata side.
Hardlinks or delete?
| Send extras to Recycle Bin | Replace with hardlinks | |
|---|---|---|
| Disk space reclaimed | Yes | Yes |
| Album folders still contain their photos | No | Yes |
| Works across drives | Yes | No, hardlinks only work within the same NTFS volume |
| Easy to undo | Restore from Recycle Bin | Files remain in place, but they now share one copy |
| Best for | Moving to a new photo library | Keeping the Takeout structure as an archive |
A hardlink is a second name for the same file data. Editing the file through one folder changes it in every folder. That is fine for an archive, but if you plan to edit photos, decide on one master copy instead. Our guide to replacing duplicate files with hardlinks explains the trade-offs.
Common mistakes
- Scanning before all zips are extracted. Some sets will look unique when their copies are still in an unopened zip.
- Deleting “-edited” files as duplicates. They are your edits. Review them as look-alikes.
- Deleting the JSON files too early. They hold dates and locations some tools use to restore metadata.
- Using a name-only duplicate finder. Name matching misses copies that were renamed and flags different photos that share a name. Content hashing avoids both.
- Running it on the only copy of your library. Keep the original zips until you are happy with the cleanup.
Frequently asked questions
Why is my Google Takeout bigger than my Google Photos storage?
Usually because photos in albums are exported twice, once in the year folder and once in each album folder. Edited versions and JSON files add more.
Does Google Takeout include albums?
Yes. Each album you created gets its own folder, alongside the “Photos from [year]” folders. Photos in albums are copied into both.
Are the “-edited” photos duplicates?
No. They are the edited versions you saved in Google Photos. They look like the original but contain different data.
Can I delete the JSON files from Google Takeout?
You can if you do not need them. They contain the date, location and descriptions from Google Photos, so keep them until any metadata restore is done.
Is it safe to delete duplicates found by DupeDoctor?
Its safety rules are enforced in code: deletions go to the Recycle Bin, at least one file in every set is always kept, and look-alikes are never auto-selected. See also our guide to deleting duplicate files safely.
Try DupeDoctor
Get DupeDoctor from the Microsoft Store. It is a one-time purchase, and the Store listing shows the current price. For backups beyond Takeout, see finding duplicates across backup drives and finding duplicate photos on Windows.
One clean copy of every photo, with your albums and edits intact.
Get it from Microsoft
Google, Google Photos and Google Takeout are trademarks of Google LLC. Microsoft and Windows are trademarks of the Microsoft group of companies. DupeDoctor is not affiliated with or endorsed by them.
Last updated 19 October 2026. Published by Automata Labs, a software company in Canada.