ReviveDB vs BackupDrill — two Supabase backup services that both restore-test, one storing in the EU and one in your bucket

ReviveDB and BackupDrill agree on the important thing: a backup that has never been restored is a guess. ReviveDB restore-tests every database backup; BackupDrill restores the latest snapshot in a scheduled drill — weekly on paid plans, once on the first backup on Free — and both report the result. They disagree about where the copy should live and how much of a Supabase project belongs in it. ReviveDB keeps recovery points on its own EU storage and covers Edge Functions and configuration too; BackupDrill writes into a bucket you own and stops at the database and Storage files. Facts checked in September 2026.

Side by side

	ReviveDB	BackupDrill
Price (as of September 2026)	Free: 1 project, database up to 1 GB, weekly, 3 rolling recovery points. Pro €9/month: 3 projects, 10 GB each, daily, up to 30 days. Scale €29/month: 10 projects, up to 60 days. Prices include VAT	Free: 1 project, weekly, one restore drill on the first backup. Solo $19/month: 5 projects, daily, weekly drills. Team $49/month: 20 projects. Agency $99/month: 100 projects plus client PDF reports
Where the backup lands	ReviveDB's own storage in the European Union — no US or APAC destination, and no bring-your-own bucket	Your own S3, Cloudflare R2, or Backblaze B2 bucket, in whatever region you choose; bytes stream through the worker and are never written to BackupDrill's disks
What gets backed up	Database and Auth, Storage file bytes, Edge Function bundles, and supported project configuration, in one recovery point	The Postgres database (public schema by default), plus Storage files once you add the project's Storage S3 keys. Edge Functions, Auth configuration, and secrets are not backed up
Restore verification	Every database backup is restored into an isolated PostgreSQL instance and compared against a capture-time inventory — schemas, tables, columns, constraints, indexes, views, functions, policies, RLS state, grants; Storage objects and Edge Function bundles are checked by byte count and SHA-256 rather than restored (per its verification page)	Scheduled drills restore the latest snapshot into a throwaway Postgres and verify archive checksum, table counts, populated tables, and sampled Storage file checksums; weekly on paid plans, once on Free. Boundary: grants are not in the backup at all (the dump runs with --no-privileges) and have to be re-applied after any restore, and RLS policies that name Supabase-managed roles are reported as skipped in the plain-Postgres sandbox rather than verified
Connecting a project	Supabase OAuth; a temporary read-only database login is created for each backup and expires after an hour	Supabase OAuth creates one read-only backup role (BYPASSRLS, pg_read_all_data) that stays until you revoke it with the SQL shown on the project page; or a generated SQL snippet; or a pasted connection string
Backup frequency	Weekly on Free, daily on paid plans — daily is the shortest schedule offered	Weekly on Free, daily on paid plans
Usage limits	Metered: backup storage per account (5 / 50 / 100 GB) and processing per month (5 / 20 / 70 GB) by plan	Per-project size caps by plan (10 GB on Solo, 25 GB on Team, 100 GB on Agency); storage is whatever your bucket allows
Restore path	Through ReviveDB's dashboard and docs; recovery points live on ReviveDB's side	Hosted restore wizard restores the database into a fresh Supabase project (Storage files follow with the CLI command it hands you), or the open-source CLI end to end, or plain pg_restore — the artifact in your bucket is a standard pg_dump custom-format file

Facts checked against revivedb.dev and its pricing page in September 2026; both products are young and change quickly, so confirm current limits on their site. Both read your data out of Supabase on every run, which Supabase may bill as egress.

Choose ReviveDB if

Edge Functions and Auth configuration are part of what you would need back. Redeploying functions from a repository is fine for most teams; if yours would rather have them in the recovery point, ReviveDB includes them and BackupDrill does not.
EU data residency is a requirement rather than a preference, and you do not want to run a bucket to get it — ReviveDB’s storage is EU-only by design.
You want the lowest price with restore testing included: €9/month for three daily-backed-up projects, VAT included.
You prefer that no standing database role exists — ReviveDB creates a temporary login per backup and expires it within the hour.

Choose BackupDrill if

You want the copy in storage you control. Every snapshot is a standard pg_dumpcustom-format file — plus your Storage files once you add the project’s Storage S3 keys — in your own S3, R2, or B2 bucket, readable with stock tools if BackupDrill disappeared tomorrow.
You are outside Europe and care where the copy lives: the bucket region is yours to pick. Be precise about what that covers — during a backup the bytes stream through BackupDrill’s worker without being written to its disks, and during a drill the snapshot is downloaded into a temporary sandbox that is destroyed when the drill ends; the lasting copy is only ever in your bucket.
You want more projects per dollar at the top: Team is 20 projects for $49/month, Agency 100 projects for $99/month with client-facing PDF drill reports, which is the agency use case in the name.
You want a restore you can run yourself, from the artifact, with the open-source CLI — the same engine the hosted service runs, MIT licensed.

The one-sentence difference

Both will tell you whether a backup restores — ReviveDB for each one, BackupDrill for the latest on a weekly drill; ReviveDB keeps that backup for you in Europe and covers more of the project, BackupDrill hands it to you in your own bucket and covers the database and, once you add its keys, the files. If you want the second model, start with the quickstart — the free plan covers one project.

FAQ

Does ReviveDB test that backups restore?

Yes. Its site states that every database backup is restored into an isolated PostgreSQL instance and compared against an inventory taken at capture time — down to policies, RLS state, and grants — on every plan including Free, and that Storage objects are checked by byte count and SHA-256 hash rather than restored. BackupDrill restores the latest snapshot in scheduled drills — weekly on paid plans, once on the first backup on Free — checking table counts, populated tables, and a checksum sample of the Storage files (100 by default) against the manifest; because that sandbox is plain Postgres, grants are not in the backup at all (the dump runs with --no-privileges) and have to be re-applied after any restore, and RLS policies that name Supabase-managed roles are reported as skipped in the plain-Postgres sandbox rather than verified. Both verify file integrity; ReviveDB hashes every object, BackupDrill samples.

Where does ReviveDB store my backups?

In the European Union, on ReviveDB's own storage. There is no US or APAC option and no bring-your-own bucket, which is the main structural difference from BackupDrill, where every snapshot is written into an S3, R2, or B2 bucket you own and can inspect, restore from, or delete without asking anyone.

Does ReviveDB back up Supabase Storage files and Edge Functions?

Both, per its site: Storage object bytes and Edge Function bundles are included in each recovery point, along with supported project configuration. BackupDrill covers Storage files once you add the project's S3 keys but does not back up Edge Function source, Auth settings, or secret values — those are outside the database and have to be redeployed from your repository.

Which is cheaper?

ReviveDB: €9/month including VAT for 3 projects with daily backups, versus $19/month for BackupDrill Solo with 5 projects. ReviveDB meters backup storage and monthly processing on top of the project count; BackupDrill's paid limit is per-project size, with storage cost paid to your own bucket provider (Cloudflare R2's first 10 GB is free).

Sources

Facts and prices last verified September 7, 2026 against the sources above. Written by the team behind BackupDrill.