Zoom to GDrive — Transfer Pipeline Walkthrough

Overview

Zoom to Google Drive Transfer

Automated pipeline that downloads Zoom recordings, converts VTT transcripts to Google Docs, and smart-names files using Google Calendar event matching. One command, all recordings transferred.

280

Recordings Transferred

92.9%

Calendar Match Rate

13 mo

Jan 2025 — Feb 2026

What It Does

Downloads all Zoom cloud recordings
Converts VTT transcripts to Google Docs
Smart-names files from Google Calendar
Organizes by year/month folders
Logs every transfer to a spreadsheet
Handles Zoom's 30-day API pagination

Key Files

File Structure
`zoom_to_drive.py`	Main orchestrator
`zoom_client.py`	Zoom API wrapper
`gdrive_client.py`	Google Drive + Docs API
`naming_engine.py`	3-tier smart naming
`naming_rules.json`	Time-based naming rules
`state_manager.py`	Transfer state tracking

💡

Fully automated: Run python zoom_to_drive.py and it handles everything — API pagination, file downloads, naming, Google Drive upload, transcript conversion, and logging.

Pipeline

Transfer Pipeline

End-to-end flow from Zoom cloud storage to organized Google Drive folders. Each recording goes through naming, download, and upload stages.

Pipeline Flow

Zoom APIList recordings

→

Smart NameCalendar lookup

→

DownloadMP4 + VTT

→

UploadGoogle Drive

→

ConvertVTT → Google Doc

→

LogTransfer sheet

Per-Recording Steps

Check state Skip if already transferred (state_manager tracks by meeting UUID + file type)
Determine smart name 3-tier naming: Calendar match → Time rules → Zoom topic fallback
Create folder structure Year/Month folders created on demand (e.g., Zoom Recordings/2026/02 - February)
Download from Zoom MP4 video + VTT transcript (if available). Uses download_url with access_token.
Upload to Google Drive Resumable upload for large files. Sets file name to smart name.
Convert VTT transcript Parse VTT, strip timestamps, format as readable text, create Google Doc.
Log transfer Append row to Transfer Log spreadsheet with date, name, source, status.

⚠️

Resumability: If the script is interrupted, re-running it skips already-transferred files. The state manager persists progress to a local JSON file.

Intelligence

Smart Naming (3 Tiers)

The naming engine uses a 3-tier fallback system to give each recording a meaningful name instead of Zoom's default UUID-based filenames.

Tier 1: Google Calendar Match (92.9%)

Looks up the recording's start time against Google Calendar events. If a calendar event overlaps the Zoom meeting time window, uses the event title as the file name.

✅

Example: Zoom meeting at 2:00 PM → Calendar event "Client Call - GloboTech" at 2:00-2:30 PM → File named 2026-02-15 - Client Call - GloboTech.mp4

Tier 2: Time-Based Rules (4.8%)

When no calendar match is found, falls back to naming_rules.json. Rules map time windows and days to recurring meeting names.

{
  "rules": [
    {
      "name": "Morning Standup",
      "days": ["Monday", "Wednesday", "Friday"],
      "time_start": "09:00",
      "time_end": "09:30"
    },
    {
      "name": "Weekly Team Sync",
      "days": ["Tuesday"],
      "time_start": "14:00",
      "time_end": "15:00"
    }
  ]
}
        

Tier 3: Zoom Topic Fallback (2.3%)

Last resort: uses the Zoom meeting topic as the file name. This is Zoom's default meeting name (e.g., "Shane's Personal Meeting Room" or "Quick Sync").

⚠️

Common issue: Generic topics like "Zoom Meeting" or "Personal Meeting Room" are not descriptive. Consider adding a naming rule for recurring time slots to avoid this tier.

Naming Tier Stats (280 recordings)

Naming Source Distribution
Tier 1 — Calendar	260 recordings	92.9%	Primary
Tier 2 — Time Rules	13 recordings	4.8%	Fallback
Tier 3 — Zoom Topic	7 recordings	2.3%	Last Resort

API

Zoom API Handling

Zoom's API limits recording list queries to 30-day windows. The pipeline chunks date ranges automatically and handles pagination, rate limits, and auth token refresh.

30-Day Chunking

Start DateJan 1, 2025

→

Chunk 1Jan 1-30

→

Chunk 2Jan 31 - Mar 1

→

...Auto-chunked

→

End DateFeb 28, 2026

13 months = ~14 API chunks, each ≤ 30 days

API Configuration

Zoom API Settings
Auth	Server-to-Server OAuth (JWT deprecated)
Max range	30 days per request
Page size	300 recordings per page
Rate limit	10 requests/second (heavy)
Download URLs	Require `access_token` param
Token TTL	1 hour, auto-refreshed

Error Handling

401 Unauthorized: Auto-refresh OAuth token and retry
429 Rate Limited: Exponential backoff (1s, 2s, 4s, 8s)
404 Not Found: Recording expired or deleted — skip and log
500 Server Error: Retry up to 3 times with 5s delay
Download timeout: 5-minute timeout for large files, retry once

🚨

Zoom recording retention: Cloud recordings are auto-deleted after 30 days on free/pro plans. Run transfers regularly to avoid losing recordings. The pipeline tracks the last successful run date to start from there.

Integration

Google Drive Integration

Recordings are uploaded to an organized folder structure. VTT transcripts are converted to searchable Google Docs. Uses Google Drive API v3 and Google Docs API.

Folder Structure

Zoom Recordings/
  2025/
    01 - January/
      2025-01-15 - Client Onboarding Call.mp4
      2025-01-15 - Client Onboarding Call (Transcript).gdoc
      2025-01-22 - Team Planning Session.mp4
    02 - February/
      ...
  2026/
    01 - January/
      ...
    02 - February/
      2026-02-15 - Client Call - GloboTech.mp4
      2026-02-15 - Client Call - GloboTech (Transcript).gdoc
        

Upload Process

Check/create year foldere.g., "2026"
Check/create month foldere.g., "02 - February"
Resumable uploadLarge files use resumable protocol
Set metadataName, MIME type, parent folder

VTT Transcript Conversion

Parse VTT fileExtract text, strip timestamps and metadata
Format contentClean paragraphs with speaker labels
Create Google DocUpload as Google Docs MIME type
Name with "(Transcript)"Same smart name + suffix

Google API Authentication

Auth Configuration
credentials.json	OAuth client credentials (from Google Cloud Console)
token.json	Auto-generated refresh token (persisted between runs)
Scopes	`drive.file` + `docs` + `calendar.readonly`
Project ID	851550492652 (shared with Accounting tools)

Tracking

Transfer Log

Every transfer is logged to a Google Sheets spreadsheet for auditing and tracking. Includes source details, naming tier used, and transfer status.

Log Columns

Transfer Log Spreadsheet
Transfer Date	When the transfer was executed	Date
Meeting Date	Original Zoom meeting date/time	Date
Original Name	Zoom's default file name (UUID-based)	Text
Smart Name	Final name assigned by naming engine	Text
Naming Tier	Which tier was used: Calendar / Rules / Topic	Category
File Type	MP4, VTT, M4A (audio-only), etc.	Text
File Size	Size in MB	Number
Duration	Meeting duration in minutes	Number
Drive Link	Direct link to uploaded file in Google Drive	URL
Status	Success, Skipped, or Error (with reason)	Status

💡

State vs Log: The state_manager.py tracks which files have been processed (local JSON). The Transfer Log spreadsheet is the audit trail (cloud). Both are updated together, but the state file is what prevents duplicate transfers.

Quick Start

Quick Start Guide

Get up and running with the Zoom to Google Drive transfer pipeline.

Prerequisites

Python 3.10+
Zoom Server-to-Server OAuth app (in .env)
Google Cloud project with Drive, Docs, Calendar APIs enabled
credentials.json from Google Cloud Console

Setup Steps

Install dependencies pip install -r requirements.txt
Configure .env Set ZOOM_ACCOUNT_ID, ZOOM_CLIENT_ID, ZOOM_CLIENT_SECRET
Place credentials.json Google OAuth client credentials in the project root
First run (auth flow) python zoom_to_drive.py — opens browser for Google OAuth consent on first run
Subsequent runs Same command. Picks up from last transfer, skips already-done files.

Common Commands

# Transfer all recordings (from last checkpoint)
python zoom_to_drive.py

# Transfer specific date range
python zoom_to_drive.py --from 2026-01-01 --to 2026-02-28

# Dry run (list what would transfer, no downloads)
python zoom_to_drive.py --dry-run

# Force re-transfer (ignore state file)
python zoom_to_drive.py --force
        

DO

Run transfers regularly (Zoom deletes cloud recordings)
Add naming rules for recurring meetings
Keep credentials.json and token.json gitignored
Use --dry-run before large batch transfers
Check the Transfer Log for any Error status rows

DON'T

Don't delete the state file unless you want full re-transfer
Don't share credentials.json or token.json
Don't run multiple instances simultaneously (state conflicts)
Don't exceed Zoom API rate limits (10 req/s)
Don't manually rename files in Drive (breaks dedup)

🎉

Walkthrough complete! You now understand the full Zoom to Google Drive pipeline — from API chunking and smart naming through upload, transcript conversion, and logging. 280 recordings and counting.