Move files between clouds without downloading them first
Why download-and-reupload fails on large folders, what actually works, and where a dedicated transfer tool earns its keep.
6 min read Updated
You have a folder in Google Drive, Dropbox, or OneDrive. You want it in another cloud. The obvious path — download everything, then re-upload — is also the slowest, most fragile one. This guide walks through the alternatives honestly: what works for small jobs, where each approach breaks, and what to reach for when none of them are enough.
What people usually try first
The default plan is to right-click the folder, hit Download, wait for a zip, and re-upload it on the destination. For a few hundred megabytes, that’s fine. For anything bigger, three things tend to go wrong.
First, the zip itself fails. Google Drive’s web UI silently splits very large folders into multiple archives or stops mid-export. Dropbox refuses to zip past a size limit. OneDrive’s “Download” can quietly skip shared items you don’t fully own.
Second, the upload chokes on home internet. A 200 GB folder over a 50 Mbps upload link is roughly nine hours of uninterrupted streaming — which means it’s actually a week of restarts, because anything from a Wi-Fi blip to a tab close kills it. Resume support across providers is uneven.
Third, you lose metadata. Comments, share permissions, version history, “Last modified” timestamps — most of it doesn’t survive a download/re-upload round trip. For a personal photo dump that’s annoying. For a company migration it’s a real problem.
What works better
A handful of approaches sidestep the computer entirely. They’re worth knowing on their own merits — not every job needs a paid product.
- Provider-native migration tools. Google Workspace, Microsoft 365, and Dropbox each ship admin-level migration tooling for their own platform. They’re the right answer when you’re staying within one vendor (e.g., consolidating two Google Workspace tenants). Outside that lane, they don’t help.
rclone. The open-source command-line tool that mounts and copies between most cloud providers, server-to-server, without round-tripping through your machine. It’s powerful, free, and battle-tested. The trade-off is that it’s a CLI: you write a config file, you understand OAuth scopes, you read the docs. For technical users, it’s often the right choice. For most people, it isn’t.- Google Takeout / equivalents. Useful for a one-shot personal export, especially before leaving a service. The export is asynchronous (sometimes a day or more), the archives are split by size, and you still have to upload them somewhere. Fine for backups; painful for migrations.
- Server-to-server hosted tools. Mover.io was the canonical example before Microsoft acquired it and shut it down to non-Microsoft destinations. MultCloud, CloudHQ, and similar services still exist; they vary widely in pricing, provider coverage, and how transparent they are about where your data flows.
For a one-off transfer of moderate size, one of those will probably do the job.
When that still isn’t enough
The honest answer to “which is best” is “it depends on the job.” A few situations push past what the simple answers cover:
- Large folders with mixed file types — hundreds of thousands of small files, where per-file API overhead matters more than raw bandwidth.
- Recurring transfers, not a one-time migration — you want a backup that runs every Sunday, not a tool you click once.
- Privacy or compliance constraints — you can’t have files traversing a vendor’s hosted infrastructure, even briefly.
- Cross-organization moves — a company switching from Google Workspace to Microsoft 365, with permissions and shared drives to preserve.
- Verification needs — you need a manifest of what was copied, what was skipped, and why, so the migration doesn’t end with “I think it’s done.”
If any of those describe your situation, a CLI plus a cron job can still work — but you’re now maintaining transfer infrastructure as a side project. That’s the gap dedicated tools try to fill.
How Syncologic does this
Syncologic will move files between clouds without ever pulling them through your computer. The workflow is the same regardless of provider pair:
You connect a source and a destination — Google Drive, OneDrive, Dropbox, S3-compatible storage, SFTP, your server, or Nextcloud. You scope access to the folder you actually care about, not the whole drive. You preview what will be transferred before anything moves: count, total size, conflicts, files that will be skipped. Then you choose where the transfer runs.
There are three options for that. On Syncologic’s infrastructure (Cloud Runner) is the convenient default — short-lived credentials, isolated job state, no setup on your side. Through your browser tab (“This Device”, Browser Runner) streams the bytes between source and destination so nothing lives on a hosted runner; the trade-off is that the tab will need to stay open. On your own hardware (Private Runner) is a small binary you run yourself on a server, NAS, or VPS — the bytes flow source → your hardware → destination, and Syncologic only sees metadata.
Once it’s running, you’ll watch progress live and get a completion report at the end: what copied, what was skipped, what failed and why. The same workflow will handle a one-time migration, a scheduled backup, or a cross-cloud sync.
Caveats
A few things to be straight about. Syncologic is pre-launch — you can join the waitlist, but you can’t run a transfer today. Provider coverage starts with the seven listed above; we’re not promising every cloud on the internet. The Browser Runner trades convenience for transparency: it works, but a one-terabyte job over a residential link will still take hours, and you’ll need to keep the tab open. And no transfer tool — ours or anyone else’s — preserves every piece of provider-specific metadata across vendors. Some things (Google Drive comments, OneDrive sharing intricacies) live in the source provider and don’t have a destination equivalent.
If your job is small, one-off, and within a single vendor, you may not need any of this. If it’s anything more — large, recurring, sensitive, or cross-vendor — that’s the case Syncologic is being built for. Drop your email below and we’ll let you know when it’s ready.