CYBBACK|for Salesforce
Quick User Guide

Your first Salesforce backup
in about 20 minutes.

From sign-up to an encrypted, restorable snapshot of your Salesforce org - records of every object, files and metadata - within a share of your API quota that you control.

AudienceSalesforce admins & IT
Reading time~ 10 minutes
Setup time~ 20 minutes
VersionEN - 2026
What you'll learn

6 steps to a production-ready backup

1
Get started

Sign up - trial or paid plan

Every plan covers every application, Salesforce included; only the storage changes.

  1. Open cybback.com/essai-gratuit for a 14-day trial (no credit card), or pick a plan on cybback.com/tarifs.
  2. Confirm your e-mail, then open Dashboard → Salesforce.
2
Prepare Salesforce

Create the user, the permission set and the External Client App

CYBBACK connects with one secret that backs up and restores. Its rights come from a dedicated user, for example backup-svc@acme.com, never a person's own account: its actions then show clearly in the Salesforce audit trail.

The user and its permissions

  1. Create the user backup-svc.
  2. Create a permission set CYBBACK Backup and assign it to backup-svc. Backup: API Enabled, View All Data, View Setup and Configuration, Query All Files. Restore: Modify All Data, Modify Metadata Through Metadata API Functions, Customize Application. Optional (original dates and owners): Set Audit Fields upon Record Creation, Update Records with Inactive Owners.
  3. Never use the System Administrator profile: the permission set gives exactly the rights needed.
Open point: the licence of backup-svc

The free Salesforce Integration licence may refuse Modify All Data. Backup works with it, but restoring may require a full Salesforce licence for backup-svc. After validation, CYBBACK shows a badge for the backup rights and one for the restore rights.

The External Client App

  1. Setup → External Client App Manager → New External Client App. New connected apps are disabled by default since Spring '26: use an External Client App.
  2. Enable OAuth with the scopes Manage user data via APIs (api) and Perform requests at any time (refresh_token, offline_access).
  3. Enable the Client Credentials flow, then in the policies set Run As = backup-svc. Restrict access to backup-svc (admin approved users are pre-authorized).
  4. Copy the consumer key and the consumer secret (shown after a verification code), and note the My Domain URL of the org (Setup → My Domain).
3
Connect the org

Paste the secret, check the budget, pick the scope

  1. Open app.cybback.com/dashboard/salesforce-backup → Configuration.
  2. Paste the My Domain URL, the consumer key and the consumer secret, then click Validate: CYBBACK shows the org, its type (production or sandbox), the backup-svc user, the daily API quota and the permissions found for backup and restore.
  3. Choose the scope: every object (new objects picked up automatically) or Limit to certain objects; options for files, metadata and field history.
  4. Click Save. The secret is encrypted (AES-256-GCM) before being stored.
app.cybback.com/dashboard/salesforce-backup
Dashboard
Configuration
Salesforce configurationEncrypted at rest
  ✓ acme · production · backup-svc · backup OK
One org = one connection

Add the production org, then each sandbox with its own URL and its own External Client App. A sandbox can be backed up, or only be used as a restore target.

API budget

By default CYBBACK uses at most 20 % of the daily API quota of the org and keeps a reserve of 30,000 calls (never more than half the quota) for your other integrations. Both values can be changed for each org in the form. A large org may need several nights for its first full backup.

4
Bring Your Own Bucket

Use your own S3-compatible storage

Want full data sovereignty? Point CYBBACK at your S3-compatible bucket - AWS S3, GCS, Scaleway, OVH, Wasabi, MinIO. Your data, your provider, your region.

  1. In your S3 provider, create a private bucket (no public access, versioning recommended).
  2. Create an access key / secret key with permissions limited to that bucket.
  3. Open Dashboard → Storage, select My own S3 storage, click Test connection, then Save.
5
First backup

Run your first backup

  1. Back on the Dashboard tab, click Start a backup. A daily schedule is created automatically after the first backup.
  2. Follow the progress live in the history (one line per object, plus metadata and files) or on the Events page.
  3. Read the API quota consumption block: calls used, allowed budget, quota left in the org after the backup.
Partial is not failed

When the API budget is reached, the backup stops in time and is marked partial (amber): the remaining objects and the files are picked up by the next backups, nothing is lost. The next backups only read the records that changed.

Not backed up

Passwords and sessions, sharing rows, Big Objects, external objects and earlier file versions.

6
Restore

Restore - records, objects or metadata

  1. Selection. Browse the backup at its date: list of objects, paginated records with search, a record with all its fields. Compare with Salesforce shows each field in the backup and in the live org. Tick records, a whole object or metadata components. Show the recycle bin to see deleted records.
  2. Options. Dry run (on by default: the target org is read, never written), target org (the same one or another connected org), existing records (changed fields only by default, overwrite with explicit confirmation, or skip), recycle bin first, recreation of purged records, parents and master-detail children, files, original dates and owners.
  3. Execution. The worker undeletes what is still in the Salesforce recycle bin, recreates the rest in two passes (parents first, references remapped) and reports every record. You can cancel; a re-run never duplicates.
Dashboard → Salesforce → Restore
1. Selection
2. Options
3. Execution
Restore options1 account · 3 contacts · 2 opportunities
Dry runNothing is written to Salesforce
 
What cannot be restored identically

Recreated records get new ids. Your validation rules, flows and triggers stay active: a refusal is reported record by record and the restore is marked partial. Users, profiles and sharing are not restored, field history is exported only, files come back in their latest version. Metadata restorable: custom objects and fields, record types, global value sets, page layouts and tabs. Up to 50,000 records per restore.

You're all set.

Need help? Our team replies within one business day on every plan.

Contact support →