Google Takeout Fixer: restore photo dates, GPS and captions
Photos show the wrong date after Google Takeout? This puts the real date taken, the location and, if you want, the caption back inside every photo and video.
Drop your Takeout .zip files or the extracted folder
Add every part at once (takeout-…-001.zip, -002.zip …): a photo's JSON can sit in another part.
Free: scan and preview everything, fix up to 25 files. Day Pass or Pro: unlimited. See plans
Runs in your browser. Your photos never leave this device.
The only request during a job checks your plan. It sends the tool name and a file-count range, never file names or contents.
Check it yourself: open your browser's network tab while the tool runs, or switch off Wi-Fi after the page has loaded. The job still works.
Omnvert is not affiliated with Google, Meta or Microsoft. Works with Google Takeout exports.
This tool processes on your device; your file is not uploaded for processing.
About
A Google Photos Takeout gives you your photos back, but not in the shape you expect. The date taken, the location and the caption sit in a separate .json file next to each photo, and many photos lost those details inside the file when they were uploaded or edited. Import the folder into Apple Photos, Immich, Synology Photos or OneDrive and years of pictures pile up on the day you downloaded them.
This tool reads every part of your Takeout together, pairs each photo and video with its JSON sidecar and writes the date taken back into the file: DateTimeOriginal with its time-zone offset and, when the file has none, the GPS position for JPEG photos, and the creation time for MP4 and MOV videos. It handles the cases that break simple scripts: sidecar names cut at about 51 characters, the newer .supplemental-metadata.json names, counters that move from IMG_1234(1).jpg to IMG_1234.jpg(1).json, edited copies that share the original's JSON, live photo pairs, and JSON files that ended up in a different ZIP part from their photo. The image and video data are never re-encoded: only the header bytes change, and the result is checked by reading it back.
Limits, plainly: HEIC, PNG and WebP files are copied as they are for now and get their date as the ZIP entry time and through the set-dates scripts; writing inside them comes later. GIF, BMP and RAW files never get a date written inside. A camera date already in a photo is kept unless you choose to replace it. A file with no JSON gets a date from its name when the name carries one (IMG_20190312_142233, PXL_…, Screenshot_…); a folder name like "Photos from 2019" alone is never used as a date, only reported.
Everything runs in your browser. Archives are read in place, so a 50 GB Takeout is never copied or uploaded, and the only request during a job checks your plan with the tool name and a file-count range. Scanning and the full preview are free; writing is free for up to 25 files per job, and a credit, a Day Pass or a subscription covers the whole Takeout.
How it works
- 1Add every part of your Takeout (the .zip files) or the folder you extracted them to.
- 2Wait for the scan: it reads the archives, parses every JSON and checks the dates already in your photos. Nothing is written yet.
- 3Check the summary and the preview table, pick the time zone and the output options, and download the list of files without a date if you want.
- 4Choose a folder or a ZIP download and write the fixed files, then import them into Apple Photos, Immich or Synology Photos.