How to organize a Google Takeout photos export
You asked Google to hand back your photos, and what landed in your
Downloads folder is a pile of zip files with names like
takeout-20260914T…-001.zip — a dozen of them, each hundreds of megabytes,
each stuffed with folders you didn't create and the same pictures repeated over and over.
This is normal. Here's what actually arrives in a Google Takeout export, why it's so
awkward to sort by hand, and how to turn it into one clean, dated library.
How to request the export
If you haven't pulled your photos out yet, start at takeout.google.com and be deliberate about it — the default is to export everything in your Google account, which you don't want:
- Deselect all at the top of the product list, so nothing is checked.
- Scroll down and tick only Google Photos. You can drill in and pick specific albums or years, but "all photos" is the usual choice.
- Click Next, then choose .zip as the file type (it opens on any computer) and a per-file size — 2 GB, 4 GB, 10 GB, 50 GB. Google splits your library across as many files as it needs to stay under that size.
- Choose delivery — a download link by email is simplest — and hit Create export. A large library can take hours to a day or more; Google emails you when the archives are ready.
What actually arrives
Once you download and start unzipping, the shape of the thing becomes clear — and it's messier than "here are your photos":
- Multiple archives. Because of that size limit, your library comes as
takeout-…-001.zip,-002.zip,-003.zipand so on. Ten, twenty, thirty files is common for a decade of phone photos. - The same photo, many times over. Inside, Takeout lays your pictures out
by album and in date-based folders like
Photos from 2019— and a shot that lives in three albums is physically copied into all three, plus the year folder. One picture, four or five identical files scattered across the export. - Stray
.jsonsidecars. Next to nearly every image sits a little metadata file —IMG_2043.jpg.jsonand the like — holding Google's own record of the photo. It clutters every folder and means nothing to your operating system. - Photos split across archives. A single album can straddle the size
boundary, so half its pictures land in
-004.zipand half in-005.zip. You can't just open one archive and trust you have a whole event.
The folder dates are lying to you. When you unzip a Takeout archive, every file's date reads as today — the day you exported and extracted it, not the day the photo was taken. Sort that folder by date and a wedding from 2014 and a hike from last week sit side by side, all stamped this week. The real capture date is still inside each photo; it just isn't on the file anymore.
Why this is so painful to sort by hand
None of the usual tricks survive contact with a Takeout export:
- You can't just merge the folders. Drag every
Photos from 2019together and you'll be asked, hundreds of times, whether to replace or keep the identical copies that Takeout duplicated across albums — the manual version of a job no one should do. - Sorting by date doesn't work, because after extraction every file claims the same export date. The one signal you'd want — when the picture was actually taken — is the one the file system no longer shows.
- The duplicates aren't always identical. A photo you also have from an iCloud export, or one a service re-compressed along the way, won't match the Takeout copy byte-for-byte, so "remove exact copies" tools miss it.
- The volume is the killer. You can reconcile a few hundred files by eye across two folders. Across thirty archives and fifty thousand photos, you can't — and that's exactly the size Takeout exports tend to be.
So the export sits in Downloads, half-unzipped, because the honest first step — get every photo into one place, once, dated correctly — is hours of tedious, error-prone work before you've even started organizing.
This is exactly what PictureAttic automates
Point it at the Takeout zips and PictureAttic unpacks every one of them for you, de-duplicates the copies Takeout scatters across albums — the exact duplicates and the re-compressed ones — reads each photo's real EXIF capture date and camera, and files everything by camera and year the way you choose. It runs entirely on your computer, nothing uploaded, your originals read-only. In internal testing; sign up before launch for 25% off.
Get early access