A small business data backup plan should answer four questions before anything goes wrong: what must be saved, how often it is copied, where protected copies are stored, and exactly how operations will be restored. If ransomware locks a laptop or a storage drive fails, the goal is to recover trusted customer records, accounts and business systems without discovering that every backup was connected to the damaged device.

A practical starting point is the 3-2-1 rule: keep three copies of important information, including the working original; use two storage types or appropriately separate systems; and keep one copy offsite. Add at least one offline or effectively immutable backup that an attacker with ordinary business credentials cannot simply delete. Finally, perform real restore tests. The UK National Cyber Security Centre (NCSC) and the US Cybersecurity and Infrastructure Security Agency (CISA) stress offline protection and testing because ransomware can target accessible backups as well as live files.

This guide is written for a small shop, consultancy, school, restaurant or online business—not just organisations with full-time IT teams. It covers equipment failure, accidental deletion, cloud account compromise and ransomware, with a sample plan you can adapt. Research verified 4 November 2026.

Why a small business needs a backup plan

Important business data often lives in several places: a manager's laptop, a point-of-sale computer, an accounting platform, a cloud drive, staff email and the website hosting account. One damaged laptop may mean losing years of invoices, but an account compromise can spread further when every device uses the same cloud login.

Consider a hypothetical retailer in Nigeria whose checkout computer suddenly fails. If sales and stock records were backed up earlier that morning, the business might replace the device and resume from a known recovery point. Without a backup, staff may need to reconstruct orders, balances and stock counts manually. During ransomware, the stakes are higher: attackers may encrypt or steal records and attempt to destroy network-accessible copies.

The National Institute of Standards and Technology (NIST) recommends planning, maintaining and testing backup files to minimise disruption from ransomware, hardware problems and accidental destruction. Backups are one part of security; they do not replace access controls, updates, phishing defences or incident response.

Step 1: List the information the business cannot afford to lose

Start with business operations, not with buying a drive or subscription. Walk through a normal working day and identify what would stop trading if the relevant device or account became unavailable.

Business informationTypical locationRecovery concern
Sales, POS and payment reconciliationPOS application, cloud platform or local databaseMissing orders, unsettled payments or inconsistent balances
Customer and supplier recordsCRM, spreadsheets, email or ERPLost contacts, commitments and order histories
Accounts, payroll and invoicesAccounting software and document foldersTax, reconciliation and employee-payment disruption
Stock, product listings and pricesInventory database, marketplace or online shopOverselling and difficult stock reconstruction
Website, domain and app configurationHosting platform, source repository and administrator accountsSlow service restoration without configurations and access
Staff files and correspondenceLaptops, cloud drives, collaboration toolsLost contracts, case histories or approvals

For each system, record an owner, how its data can be exported or backed up, whether it contains personal information, and any vendor contract or retention requirement. A cloud application saying it “stores everything securely” is not proof that your business can independently export and restore the records after an account dispute or outage.

Step 2: Decide how much downtime and data loss are acceptable

Two useful planning measures are recovery point objective (RPO)—the maximum amount of recent work a business is prepared to lose—and recovery time objective (RTO)—the desired time to restore an essential service. They are targets for planning and testing, not guarantees.

For example, a busy shop might set an illustrative RPO of one hour for sales transactions and an RTO of four hours for its checkout system. That means the backup or transaction-log strategy should be able to recover close to the previous hour, while the recovery process should be designed and tested to meet the four-hour target. A consultant's archive of old project files may reasonably need less frequent backups and tolerate a longer restoration period.

SystemIllustrative RPOIllustrative RTOPossible approach
Sales/POS records1 hour4 hoursFrequent consistent backups and transaction recovery
Accounts and invoices1 day1 working dayDaily backup and tested export
Website content1 day1 working dayApplication, database and media backups
Archived documents1 weekSeveral daysRegular encrypted versioned backup

These values are examples, not universal standards. An online store with hundreds of daily orders may need substantially shorter recovery points. The correct frequency follows business impact, the cost of lost records and the recovery capability actually available.

Step 3: Apply the 3-2-1 backup rule—and add ransomware protection

The 3-2-1 rule means three copies in total: one production copy and two backups, across two different types or suitably separate storage systems, with at least one offsite. The Netherlands National Cyber Security Centre explains the model and stresses testing, storage location and retention.

A simple setup could use the working business computer, an encrypted external drive kept disconnected after each backup, and a separate offsite cloud backup account with versioning and appropriately protected deletion controls. The point is avoiding a single failure that destroys every copy.

Offline means the copy is not routinely accessible from the affected network. Immutable means changes or deletion are restricted for a defined period. They are different protections. An always-connected external drive can be encrypted by ransomware; a cloud backup controlled by the same compromised administrator account can also be deleted unless its retention and access controls resist that action.

Choose protection suited to your budget and threat model. Keep backup administration credentials separate from normal staff accounts, enforce multi-factor authentication, restrict deletion rights and consider write-once or immutable retention when the service genuinely supports it. Encryption protects confidentiality, but only if recovery keys are kept safely and remain accessible during a crisis.

Step 4: Understand why cloud sync is not automatically a backup

Services that synchronise folders across devices are convenient. They can also synchronise accidental deletion or encrypted files. File history and trash recovery may help, but their retention limits and permissions are not always suitable for an independent recovery plan.

The NCSC's cloud-data backup guidance tells small organisations to ensure that essential cloud data can be exported and restored, rather than relying solely on built-in file versions or deleted-item recovery.

Before calling a cloud tool your backup, ask: Can I restore a deleted folder from months ago? Can I recover without access to the main compromised account? Can I export email, database records or application settings? Are the copies protected from mass deletion? Is the restore procedure documented and tested? A “yes” to automatic syncing alone does not answer those questions.

Step 5: Configure the right backup type and schedule

A full backup copies the selected data set. An incremental backup usually captures changes since the previous backup, while a differential backup captures changes since the last full backup. Each can reduce storage or processing costs but changes how restoration works. The Canadian Centre for Cyber Security's ransomware playbook explains these approaches.

For many small businesses, automation is safer than relying on someone remembering to plug in a drive every Friday. Configure notifications when jobs fail and set retention rules that maintain several independent points in time. A sample policy could keep daily backups for 30 days, weekly copies for eight weeks and monthly copies for twelve months, subject to actual legal, contractual and business requirements. Those periods are examples, not prescribed Nigerian retention rules.

For databases, avoid casually copying active database files as though they were ordinary documents. Use the platform's documented consistent backup method, and include any required transaction logs, media uploads, application configuration and encryption keys. An invoice export alone may not rebuild the accounts system; a raw database file may be unusable without the correct software version or recovery logs.

Also back up the information needed to operate the backups: recovery instructions, contact details, software licences, trusted system images and a secure method of accessing recovery keys.

Step 6: Protect backup accounts, devices and recovery keys

Separate backup access from day-to-day work. A staff member who creates invoices should not be able to erase all historical backups. Limit administrator accounts, use strong unique passwords and MFA, apply software updates and review backup provider alerts.

Encrypt portable drives and offsite copies that contain sensitive business or customer information. Keep encryption keys separate from the protected data, with an authorised emergency-access procedure. If the only recovery key is on the laptop that fails, the encrypted backup may become unusable.

For physical drives, label them, rotate them safely and store the offsite copy away from theft, heat, flooding and electrical damage. Never leave the offline drive permanently plugged into the same computer. For cloud services, inspect whether version retention, account recovery and immutable deletion protection are genuinely supported—not just advertised.

Step 7: Test restoring files, applications and complete operations

A green “backup completed” notice only shows that a job claims to have run. It does not prove the data is readable, complete or useful. NIST's data-integrity recovery guidance emphasises restoring information accurately after destructive events.

Start with a harmless file restore into a separate folder, then try a representative application or database restoration in an isolated test environment. Check dates, record counts, permissions, attachments and whether a user can complete a real task such as generating an invoice.

A manageable example schedule is a weekly sample-file restore, a monthly test of an important application, and a quarterly end-to-end recovery exercise. The right frequency depends on risk and available staff; these are suggested practices rather than a regulatory timetable.

Record the test date, backup point used, responsible person, how long the restore took, errors discovered and what was corrected. If recovery takes twelve hours but your RTO is four, improve the plan before a real incident.

What to do immediately after a ransomware attack

If files are suddenly encrypted, a ransom note appears or multiple systems behave abnormally, do not connect a clean backup drive to the affected machine or start restoring data before the incident is contained.

  1. Isolate affected devices: disconnect compromised computers from Wi-Fi, wired networks and shared drives where safe. Ask your IT responder for help before powering systems off, because shutting them down can destroy useful volatile evidence.
  2. Protect clean copies: secure offline backups and restrict compromised accounts from cloud-backup administration. Do not expose backup credentials on the infected machine.
  3. Identify what happened: document affected devices, first signs, business services interrupted, possible access to customer information and evidence needed for investigation.
  4. Contain and remove the threat: bring in qualified support, change compromised credentials from a clean device, close the entry point and examine connected systems.
  5. Identify a trusted recovery point: choose a backup from before the compromise, check its integrity and scan or validate it as appropriate. The newest backup is not automatically the cleanest.
  6. Rebuild on trusted systems: reinstall or reimage compromised machines from known-good media, patch them and restore into the clean environment.
  7. Validate before returning to service: verify records, application operations, user access and security monitoring, then reconnect systems in stages.

These steps reflect the layered incident-response approach in CISA's StopRansomware Guide . They are a starting framework, not a substitute for professional incident handling when sensitive systems are involved.

Paying a ransom is not a reliable recovery plan. The NCSC warns that payment does not guarantee decryption or deletion of stolen information, and the machine may remain compromised. Independent backups, incident support and appropriate reporting are safer foundations for continuity.

How to recover when a laptop, hard drive or server fails

A hardware breakdown is not necessarily a cyberattack, so the response differs. Disconnect a visibly damaged drive from use; repeated attempts to start a failing disk can worsen the damage. If the records are irreplaceable and no verified backup exists, seek professional data-recovery assistance rather than trying risky repair utilities.

With a tested backup, obtain a replacement device, install a trusted operating system and necessary software, then restore files or applications from the selected recovery point. Verify that restored documents open and that POS, accounting or stock balances reconcile with independent evidence. For hosted services, confirm DNS, credentials, configuration, media files and database versions as needed.

Only discard or recycle damaged storage after checking whether its sensitive data requires secure handling under your business policies and applicable law.

Include a manual business-continuity procedure

Backup copies restore information, but they do not automatically restore customer service. Write down how the business will continue for a few hours without its main software. A retailer might use numbered paper receipts and a controlled temporary stock log; an office may direct customers to a verified telephone number; a restaurant might record orders offline.

When systems return, reconcile these temporary records carefully so transactions are not duplicated or omitted. Assign who can authorise downtime communications and who will confirm that sales, balances and inventory match. The NCSC's small business response-and-recovery guide recommends preparing roles and communications before an incident.

Nigerian businesses: data-protection and breach reporting considerations

When ransomware or a lost device involves personal information, restoring from backup does not settle every legal issue. The Nigeria Data Protection Commission (NDPC) explains business obligations under the Nigeria Data Protection Act 2023. Depending on the circumstances and risk to individuals, section 40 of the Act may require notification to the NDPC within 72 hours of awareness; other notice duties can apply. Not every equipment malfunction is automatically a reportable personal-data breach.

Keep a factual incident timeline, assess whether information was accessed or exposed, consult qualified privacy or legal support and use current NDPC reporting procedures. FlyingEze's guide to assessing Nigerian business obligations under the NDPC offers related context. This section provides general information, not a case-specific legal opinion.

A simple backup plan a small business can copy

Write down the following information in a document available to authorised staff even if the main computer is offline:

Plan fieldExample entry
Essential systemsSales database, invoices, stock, customer records, website
Recovery ownerBusiness manager and designated IT support contact
Backup methodAutomatic consistent backup plus disconnected encrypted copy
Backup locationsSeparate local media and protected offsite backup service
FrequencyBased on RPO; important transactions may need hourly protection
RetentionMultiple daily, weekly and monthly versions under documented rules
Credentials and keysSeparate emergency-access procedure with restricted permissions
Recovery testRestore a sample, test an application and log the result
Incident responseIsolate, preserve evidence, contain, rebuild, restore and reconcile

Assign a person—not just a software tool—to review failed jobs and test reports. Keep a printed emergency contact list or another suitably protected offline copy. Review the plan whenever the business changes software providers, locations, staff responsibilities or important data workflows.

Frequently asked questions

Is one external hard drive enough for a small business?

It is better than having no backup, but it is not a resilient complete plan if the drive stays connected or sits beside the only computer. Add a second protected offsite copy and test both.

Does Google Drive, OneDrive or another sync tool count as backup?

Sync and file history can support recovery, but they may copy deletions or ransomware changes. Keep an independent, restorable backup of critical information rather than relying solely on synchronisation.

How often should I back up business data?

Choose a frequency that matches the amount of data loss you can tolerate. A frequently used sales database may need hourly or more frequent protection, while an archive may need less. Verify the actual recovery point in tests.

Can ransomware encrypt my cloud backups?

It may corrupt synchronised files or destroy reachable backup copies if an attacker gains sufficient access. Strong account separation, immutable retention where available, offline copies and tested restoration reduce this risk.

Can I recover without paying ransomware criminals?

Often, a clean tested backup and a secure rebuilt system offer a recovery route. Recovery still requires investigating the compromise and any stolen information. Paying does not guarantee a successful outcome.

Is backup the same as disaster recovery?

No. Backup preserves recoverable copies of data; disaster recovery includes the people, equipment, applications, credentials, sequence and tests required to resume operations.

Conclusion: The best backup is one you can actually restore

A dependable small business data backup plan does not need to begin with expensive infrastructure. It begins by identifying essential records, defining acceptable downtime and data loss, protecting multiple copies and regularly proving that recovery works.

Use 3-2-1 as a starting point, keep an offline or strongly protected copy, separate backup credentials, automate monitoring and practise restoring data. If ransomware strikes or a laptop fails, those preparations can turn a crisis into a managed recovery process instead of a permanent loss of business records.