Google Takeout JSON Viewer

What is actually inside a Google Takeout JSON file?

Below is a real-shape Location History record from a Google Takeout export. The left side is the raw JSON that ships in the ZIP. The right side is what that record actually means.

{
  "latitudeE7": 487856612,
  "longitudeE7": 23415879,
  "accuracy": 10,
  "timestampMs": "1710440627182",
  "source": "WIFI",
  "deviceTag": 1234567890,
  "platformType": "ANDROID"
}

Place
  48.7857° N, 2.3416° E
  Accuracy ±10 meters
  Source: WiFi-based reading

When
  March 14, 2024 at 6:23 PM UTC

Device
  Android phone (one of yours)

The translation rules at play in this single record:

• latitudeE7 / longitudeE7: divide by 10,000,000 to get decimal degrees.
• timestampMs: Unix epoch in milliseconds. Divide by 1000 and convert with your favorite date library.
• accuracy: meters of error radius. Lower is better.
• source: which sensor produced the reading (GPS, WIFI, CELL, UNKNOWN).
• deviceTag: an internal hash identifying the device. Same hash means same physical device across records.

A full Takeout export contains hundreds of thousands of records like this one across dozens of files. Reading them by hand is not realistic.

See your real Takeout in plain English

The breakdown below is sample data shaped like a real export, parsed for readability. Drop your own ZIP onto the tool to see your numbers.

How to parse Google Takeout JSON manually

If you want to write your own parser, the basic shape is straightforward in any language. In Python:

import json
from datetime import datetime, timezone

with open("Records.json") as f:
    data = json.load(f)

for record in data["locations"]:
    lat = record["latitudeE7"] / 1e7
    lon = record["longitudeE7"] / 1e7
    ts_ms = int(record["timestampMs"])
    when = datetime.fromtimestamp(ts_ms / 1000, tz=timezone.utc)
    print(f"{when:%Y-%m-%d %H:%M} - {lat:.4f}, {lon:.4f}")

That handles the happy path. The complications come fast:

• Schema drift. Google has shipped at least four different shapes for Location History since 2014. Older files use timestampMs, newer files use timestamp as ISO 8601. Some have a top-level locations array, others wrap differently.
• Multiple files per category. Search Activity is split across MyActivity.json files per product (one for Search, one for Maps, one for Image Search, etc.). YouTube has separate files for watch history, search history, comments, and subscriptions.
• Missing fields. Older accounts have records without source, accuracy, or deviceTag. Newer accounts have those fields but may lack older fields like place visits.
• Non-JSON data. Gmail ships as MBOX. Chrome can ship as both JSON and HTML. Photos has its own quirks. A general-purpose Takeout parser handles each format separately.
• Encoding. Some files are UTF-8 with BOM, some without. Some have non-printable characters in places. Robust parsers normalize on read.

If you want to ship a parser, plan for several hours of work plus ongoing maintenance every time Google adjusts the export format. If you just want to read your data, the tool below saves you the time.

Read your Takeout in 60 seconds without writing code

TakeoutReader is a browser-based parser. Drop your ZIP onto the page and the report renders immediately. The parser handles the schema drift, the multi-file layouts, the format conversions, and the missing-field edge cases so you do not have to.

• Six categories supported: Location History, Search Activity, YouTube History, Chrome History, Android Activity, Gmail.
• Output formats: PDF, plain text, Markdown, CSV.
• Zero upload. Parsing happens entirely in your browser via JSZip. You can verify in DevTools by watching the Network tab.
• No account, no login, no extension.

For a full audit of what Google has on you, see what does Google know about me. For a focused look at GPS records, see our Google Location History viewer. For activity-specific exports, see the Google Activity viewer hub.

Common file formats inside Google Takeout

A typical export contains several formats depending on which Google products you included:

• JSON: Location History (Records.json, Semantic Location History/), My Activity (MyActivity.json per product), YouTube (watch-history.json, search-history.json, subscriptions.json), Chrome (Bookmarks.json, History.json), Android Activity, Drive metadata.
• HTML: Pre-rendered summaries Google generates for some categories (My Activity, YouTube). Faster to skim but less complete than the JSON twins.
• MBOX: Gmail. Plain-text email format readable by any mail client (Thunderbird, mutt) but not by web browsers without conversion.
• CSV: Some products ship CSV alongside JSON (Saved, Tasks).
• KML: Location History also ships a Timeline KML format usable in Google Earth and other GIS software.
• Original media: Photos and Drive contain the raw files (JPG, PDF, etc.) plus per-file metadata.

TakeoutReader currently focuses on the six JSON-based categories with the broadest user demand. Other formats are passed through unmodified.

Frequently asked questions