Backup Plan Checklist: What to Write Down Before an Audit

Backup Plan Checklist: What to Write Down Before an Audit

Table of Contents

Veeam research found that 89% of organizations hit by ransomware had their backup repositories targeted by a threat actor, and only 32% of those organizations used repositories configured as immutable, according to a Veeam analysis of its 2025 ransomware research. The backup copy is now part of the attack surface. An auditor will ask you to prove it is governed.

The question usually arrives in one sentence: show me who owns the restore and when it was last tested. Most teams can answer from memory; few can hand over a document that names the owner and the date.

Backup retention is the written rule for how long each backup copy is kept, where it is kept and who may delete it. It is one field in a backup plan. It only means something beside the owner, the scope, the restore order and a dated test.

This checklist gives you a one-page backup plan with seven fields: owners, scope, schedule and backup retention, restore order, isolation, test date and audit evidence. Each field has a blank to complete and the evidence an auditor expects to see. Retention cells stay empty on purpose, because the right number comes from your own legal, contractual and recovery requirements, not from a default.

What a Backup Plan Must Record Before an Audit

A backup plan is a decision record. It states which data is protected and who is accountable. It sets how often copies are made and how long they live. It lists the order in which systems return after an incident, and how you know the copies work. An auditor does not grade the technology; the check is that a named person made each decision, wrote it down and tested it.

The guidance behind this is consistent. The NIST contingency planning guide, SP 800-34 Rev. 1 remains the foundational reference for documenting backup and recovery strategy. ISO/IEC 27001 includes a backup control that expects copies to be maintained and tested. Both frameworks ask for a documented, repeatable process, and neither names a particular product.

Sophos points the same way: backups should be tested regularly and stored offline or in immutable formats, which turns two good habits into two lines of your plan.

Use this seven-field structure as the page you print and sign:

  • Owners: one named person and one deputy for each decision.
  • Scope: systems and data classes in, systems and data classes out, with sign-off.
  • Schedule and retention: frequency, copies kept, location and deletion rights.
  • Restore order: rank, dependency, recovery targets and restorer.
  • Isolation: where the protected copy lives and what can reach it.
  • Test date: last restore test, result and next date.
  • Audit evidence: the log, report or record that proves each line.

If you cannot name the record that proves a line, the line is not finished.

Owners: Name One Person for Each Backup Plan Decision

Shared ownership is the most common failure in a backup plan. When four teams touch backup, none is accountable for the restore. Write one name per decision and one deputy, so the plan survives a resignation or a holiday.

The NIST SP 800-53 contingency planning control family is the control home for this work. It treats backup and recovery as a managed capability with assigned responsibility, not an administrative afterthought.

Complete this table with names, not teams:

Role Name Deputy Evidence an auditor can see
Backup owner (jobs, repositories, schedule) ____ ____ Job configuration export, change records
Data owner (what the data is and how long it must be kept) ____ ____ Signed retention requirement
Restore approver (authorizes a production restore) ____ ____ Approval record from the last test
Security contact (isolation, credentials, incident link) ____ ____ Access review for the backup identity domain

The last row deserves attention. Sophos reports that 79% of ransomware attacks now originate from compromised identities. If the same administrator identity controls production and backup, an attacker who takes one takes both. The current edition of NIST guidance on ransomware risk management addresses backing up data, securing the backups and testing restoration, so record who holds the credentials for the backup environment and how those credentials are separated from production.

Scope: Define What Your 3-2-1 Backup Plan Protects

Scope answers two questions: what is protected, and what is deliberately not. The second matters more to an auditor, because an exclusion without a signature looks like an oversight.

A 3-2-1 backup plan keeps three copies of data, on two different media types, with one copy off site. It is a useful minimum for copy count. It says nothing about who may delete those copies. It says nothing about how long they live. It cannot tell you whether an attacker with stolen credentials can reach them. Those gaps belong in your written plan.

Fill in the scope block:

System or data class Included? Copies (3-2-1 check) Exclusion reason and approver
____ Yes / No ____ ____
____ Yes / No ____ ____
____ Yes / No ____ ____

Isolation and immutability each get one line in the plan. The isolation line records which copy cannot be reached from the production network or identity domain. The immutability line records which copy cannot be modified or deleted during its retention period. The CISA #StopRansomware Guide says to maintain offline, encrypted backups of critical data and regularly test their availability and integrity in a disaster recovery scenario.

The Veeam analysis cited earlier makes the case in numbers: with 89% of victims seeing their repositories targeted and 32% using immutable repositories, a copy that is merely stored elsewhere is not a protected copy. For the implementation detail, see the StoneFly guide to immutable file-level WORM storage and the StoneFly post on object lock for backup copies.

Bar chart: 89% of ransomware victims had backup repositories targeted and 32% used immutable repositories.
Veeam research found 89% of ransomware victims had backup repositories targeted, while only 32% used immutable repositories.

Schedule and Backup Retention: Write the Retention Schedule Down

This is the field where most plans are weakest. Backup retention is usually set once, by a default, and never tied to a requirement. A retention schedule fixes that by tying every number to a reason.

A retention schedule lists each backup tier, how many copies are kept, where they sit, who may delete them and the requirement that drives the number. Leave the number cells blank until the data owner has supplied the requirement.

Tier Copies kept Location Who may delete Driving requirement
Daily ____ ____ ____ ____
Weekly ____ ____ ____ ____
Monthly ____ ____ ____ ____
Yearly ____ ____ ____ ____

Backup tools commonly implement tiers as a grandfather-father-son rotation. It keeps weekly, monthly and yearly full backups for a set number of weeks, months and years. The Veeam long-term retention policy documentation describes how Veeam Backup & Replication configures these restore points. Each tier has its own retention field, so each needs its own row in your schedule.

What Is a 30 Day Retention Period?

A 30 day retention period is a rule that keeps each backup copy for 30 days before it may be deleted or overwritten. It is a deletion rule, not a promise that every file from day 31 is recoverable. Whether a recovery point exists depends on the schedule, the job history and the tier. The plan records all three together.

Platform Defaults Are Not Your Plan

Platform defaults answer a different question from your plan. Amazon documents that automated backups for Amazon RDS default to a 7 day retention period when created from the console and 1 day from the API or CLI, with a configurable range of 0 to 35 days, and 0 disables automated backups.

Microsoft draws a similar line for its SaaS platform. The Microsoft 365 Backup FAQ, updated in November 2026, states that legal holds retain data, but that feature is optimized for export, not for mass restore. Holding data and restoring data are different capabilities. Record the platform default beside your own backup retention decision. The gap then stays visible.

Restore Order in Your Backup Plan: Decide Before the Incident

Restore order is the field that decides how long an outage lasts. During an incident, every executive believes their system is first. A written order removes the argument.

The CISA guide advises you to prioritize restoration and recovery based on a predefined critical asset list that includes systems critical for health and safety, revenue generation or other critical services, as well as the systems they depend on. Teams skip the dependency clause. An application cannot start without the identity service and network it relies on. Those return first.

Complete one row per system, ranked:

Rank System Depends on RTO RPO Restorer Restore location
1 ____ ____ ____ ____ ____ ____
2 ____ ____ ____ ____ ____ ____
3 ____ ____ ____ ____ ____ ____

RTO is the recovery time objective, the longest the business accepts a system being down. RPO is the recovery point objective, the longest span of data the business accepts losing. Write both as numbers agreed with the system owner, and write the person who performs the restore by name.

Four-stage restore-order diagram: identity and network first, then core infrastructure, tier-1 applications and remaining systems.
Restore order runs from identity and network, through core infrastructure and tier-1 applications, to the remaining systems.

Recovery through backups is now the main route out of an attack. The Sophos State of Ransomware 2026 survey of 2,158 respondents found that backup-based recovery reached 66% of encrypted-data cases, up 12 percentage points from 2025. Speed varies: the Sophos July 2026 findings report that 55% of organizations recovered within one week and 16% in less than a day. If your plan has no restore order and no dated test, you have no evidence you will be in the fast group.

How StoneFly Supports Backup Retention Isolation and Restore Testing

Two lines of the plan are hard to satisfy with process alone: the isolation line and the test line. A document can say that a copy is isolated, but the architecture has to make it true.

The StoneFly DR365V is a Veeam Ready air-gapped and immutable backup and disaster recovery solution. It offers three repository types: regular, immutable and a Veeam-aware Air-Gapped Vault® repository. The immutable repository blocks modification, overwrite and deletion during the configured retention period. The Air-Gapped Vault® repository stays isolated and is attached only for authorized backup or restore jobs, so the isolation line in your plan describes a controlled state rather than a hope.

For the test line, the DR365V includes automated recovery testing and on-demand isolated sandbox testing. Record the date and result of each test in the audit evidence column.

Teams that already run Veeam can keep their environment. The StoneFly DR365VIVA adds Air-Gapped Vault® and immutable storage to an existing Veeam environment. Veeam continues to manage job configuration, scheduling, retention, backup-copy operations and recovery orchestration. For teams that prefer a hosted model, StoneFly Veeam Cloud Backups combines Veeam backup technology with StoneFly cloud storage and optional backup testing and recovery validation.

Test Date and Audit Evidence for Your Backup Plan

A backup that has never been restored is an assumption. The CISA guide calls for regular testing of the availability and integrity of backups in a disaster recovery scenario, and Sophos adds that backups should be tested regularly. Your plan needs a dated record, not a statement of intent.

Complete the test block:

Field Entry
Last restore test date ____
Systems restored ____
Result (pass, partial, fail) ____
Signed by ____
Next test date ____

Then list the evidence an auditor can open without asking you for it:

  • Backup job logs showing success, failure and retry history for the period.
  • Restore test reports with the system, time to restore and verification result.
  • Change records for any edit to schedule, scope or retention.
  • Access reviews for the backup identity domain and repository credentials.
  • The signed retention requirement from each data owner.
Four retention tiers, daily, weekly, monthly and yearly, each with fields for copies kept, location, deletion rights and driving requirement.
The retention schedule gives daily, weekly, monthly and yearly tiers each a row for copies, location, deletion rights and requirement.

Conclusion: Keep Backup Retention Current on a Fixed Date

A backup plan earns its value when someone else can read it. Seven fields cover the decisions an auditor will ask about: owners, scope, schedule and backup retention, restore order, isolation, test date and evidence. Each one needs a name, a number or a date, and a record that proves it.

Put a review date on the page and treat it like any other control. Backup retention changes when a contract changes, a regulator updates a rule or a system moves. Restore order changes whenever a dependency changes. Review the plan on a fixed date each year, and after every major infrastructure change.

If you want to discuss how isolation and restore testing fit your plan, talk to a StoneFly engineer.

Related Products

StoneFly DR365V Veeam Ready Backup & DR Appliance

Unified Storage and Server (USS™) Hyperconverged Infrastructure (HCI)

Unified Scale-Out (USO™) SAN, NAS, and S3 Object Storage Appliance

Subscribe To Our Newsletter

Join our mailing list to receive the latest news, updates, and promotions from StoneFly.

Please Confirm your subscription from the email