Google Timeline Visualiser

A single-file, fully client-side web tool that turns a Google Timeline export into ride stats, a map, elevation/speed profiles, shareable ride-card PNGs, and GPX/CSV/JSON exports — built for motorcycle touring but useful for any trip.

→ Live: https://jameslashmar.github.io/Google-Timeline-Visualiser/

Drop your Timeline.json (or a .gpx / .csv), pick a date range, and get:

• Distance, moving time, average & max speed, elevation gain
• Bend counts (L/R split) and a twistiness score (°/km), with a best-10 km rolling window — optional "match to roads" recovers the corners sparse GPS misses
• A speed-coloured map (real basemap tiles) and an elevation + speed profile chart
• Ride-card PNGs in three layouts (vertical / square / widescreen), with an optional rider name — share to WhatsApp, Instagram, etc.
• Clean GPX / CSV / Timeline-slice JSON exports for just the dates you chose

…plus:

• Tour summary — pick a multi-day range for a per-day breakdown (km, twistiness, top speed, bends) and trip totals; click any day to jump to it.
• Best bit — one tap frames the twistiest ~10 km of the ride on the map.
• Stops — coffee/fuel/lunch stops with durations, pulled straight from the data (no lookups).
• Rally roadbook — export rally-style pace notes (distance · direction · severity · angle) as a text file.
• Compare riders — drop a mate's GPX/CSV to overlay their track and duel the twistiness scores.
• Weather — optional keyless historical weather (temp / rain / wind) for the ride, via Open-Meteo.
• Flythrough clip — an MP4 of the route drawing on over the map at constant speed, with the full ride-card stats, for Stories/Reels.
• Per-day GPX (ZIP) — one GPX per day in the range, ready to sync against action-cam footage.
• Instant reopen — your last import is cached locally (IndexedDB), so revisiting restores it without re-loading the big file (clearable, never leaves your device).

Everything parses in your browser. Your location data is never uploaded — the only network calls are optional and disclosed (map tiles, and the opt-in road-matching request).

Ride cards

Shareable PNGs with the road-matched route, place names (amber = towns you rode through), per-day weather and the full stat block — in vertical, square and widescreen:

Flythrough

A 30 fps MP4 of the route drawing itself on at constant speed with a follow-cam, count-up stats and day-by-day weather, ending on the full-tour overview (sped-up preview):

---

Getting your Timeline.json out of Google

Google moved Timeline on-device in 2024 — the web Timeline is gone and Google Takeout no longer includes it. The only export lives on your phone, and it dumps your entire history as one big Timeline.json (there's no date filter — this tool does that part).

Android (Pixel and most phones):

1. Open Settings (the phone's Settings app, not Google Maps).
2. Go to Location → Location Services → Timeline (older builds: Settings → Location → Google Location History/Timeline).
3. Tap Export Timeline data.
4. Save the Timeline.json (to Files, Drive, or share it to your computer).

iPhone (Google Maps app):

1. Open Google Maps → tap your profile picture → Your Timeline.
2. Open the settings (gear / "Location & privacy settings").
3. Choose Export Timeline data and save the Timeline.json.

iPhone without Google Maps — Apple Watch / Fitness workouts:

Apple has no Timeline-style export — Significant Locations never leaves the phone. But if you record your rides as workouts (Apple Watch, or the Fitness app on the phone), their GPS routes can:

1. Open the Health app → tap your profile picture → Export All Health Data.
2. Unzip the export — the workout-routes/ folder has one .gpx file per workout.
3. Drop those GPX files straight into the tool.

Watch workouts log GPS about once a second — far denser than Timeline's ~1/minute — so bend counts and speed colouring come out more accurate, no road-matching needed.

The file can be ~50–60 MB. That's normal — it's your whole history. This tool loads it locally and slices out only the dates you ask for; nothing is sent anywhere.

Then just open the live site and drop the file in.

---

Also accepts GPX and CSV

You don't need Google Timeline at all — you can drop:

• .gpx — from an action cam (e.g. Insta360), bike GPS, an Apple Watch workout (see the Health export steps above), or any GPS logger. These log far more often than Timeline (often ~1–10 Hz vs ~1/min), so bend counts and speed colouring are much more accurate — no road-matching needed. Per-point speed is read from the file or derived from the track.
• .csv — re-imports this tool's own CSV export (timestamp, lat, lng, speed_kmh, altitude, source). Common column aliases (latitude, lon, elevation, …) are understood too.

Everything downstream — stats, map, profile, ride cards, exports — works identically whatever you drop in.

---

Privacy

• No server, no analytics, no build step — it's one HTML file.
• Parsing and all exports happen in your browser tab.
• The only network calls, all optional and disclosed in the footer:
  • Map tiles (OpenStreetMap / CARTO) for the map view — switch to Abstract view for a fully offline session.
  • Road matching (opt-in button) sends the selected range's coordinates to the public FOSSGIS Valhalla server to snap the track to real roads.
• Your Timeline.json / GPX / CSV is never uploaded.

---

Self-hosting (and the Google Drive loader)

The tool is one file — to host it yourself just serve index.html from any static host or locally:

python3 -m http.server 8000 --bind 127.0.0.1
# then open http://localhost:8000

Serve from a folder that does not contain your Timeline.json if others are on your network.

Google Drive loading (pick your Timeline.json straight from Drive) is included in the source but is only shown when you self-host — it's hidden on the public GitHub Pages site because it needs your own Google Cloud OAuth keys, which is more setup than a casual visitor wants. When you run the source locally (localhost / file://), the "Load from Google Drive" button appears. To use it you'll need, in Google Cloud Console:

• Drive API and Picker API enabled
• An OAuth client ID (Web application — choose User data, not a service account) with your serving origin (e.g. http://localhost:8000) as an authorised JavaScript origin
• An API key (for the picker)
• Your Google account added as a test user on the OAuth consent screen

Enter both keys in the UI (they're stored only in your browser's localStorage). The scope is drive.file — the app can only see the file you pick, never your whole Drive.

---

Tech notes

• Leaflet for the map (lazy-loaded from CDN), CARTO/OSM/Esri/OpenTopoMap tiles.
• Road matching via public Valhalla (trace_route, motorcycle costing).
• Three built-in themes (Dash carbon-fibre / Roadbook paper / Dark).
• No dependencies to install; no bundler.

License

MIT — see LICENSE.