MEGABIT Insights · Cybersecurity & Infrastructure
In many SMEs, the business continuity plan was never formalised and system recovery was never tested. When something goes wrong, and it does, the consequences are severe. This guide explains what’s at stake, what today’s tools allow you to do, and how your company can be prepared already in 2026.
1. The change that’s happening, and what it means for an SME in 2026
For years, business continuity was treated as a concern for large companies. SMEs often assumed, tacitly, that an external hard drive was enough, or that “the cloud takes care of it”. In 2026, that assumption is dangerous and, in some sectors, already illegal.
Three converging forces have made operational resilience an unavoidable priority for any company, regardless of size:
- The threat has grown. Ransomware, supplier failures and human error are now frequent causes of downtime, and SMEs are no longer a secondary target: attacks are largely automated and do not discriminate by company size.
- Regulation has arrived. DORA came into force on 17 January 2025. NIS2 was transposed into Portuguese law via Decree-Law No. 125/2025. The GDPR, under Article 32, already required documented restoration capability. The requirements are concrete, have consequences, and directly or indirectly cover many Portuguese SMEs.
- The insurance market has changed. Cyber insurers tend to request evidence of immutable backups and documented restore tests before issuing or renewing policies. Without this, coverage may be refused or limited. Confirm the specific conditions with your broker.
Many statistics circulate about companies closing down after data loss, but most lack a verifiable primary source, so we don’t use them here. The reasoning doesn’t need them anyway: every hour of downtime costs unrealised revenue, stalled work, waiting clients, and in some cases, unmet legal obligations. The arithmetic becomes clear once each manager runs it with their own numbers.
2. What business continuity and disaster recovery actually mean
The terms are often used as synonyms, but they have distinct scopes. Understanding the difference helps set priorities.
Business Continuity Planning (BCP)
The Business Continuity Plan (BCP) is the strategic document that defines how the company continues to operate during and after a disruption, be it a cyberattack, a power failure, a flood or the unavailability of a critical supplier. It covers processes, people and communication, not just technology.
Disaster Recovery (DR)
Disaster Recovery is the technological component of the BCP. It defines how information systems, servers, databases and applications are restored after a failure. It is put into practice through tools, technical procedures and regular testing.
The two fundamental indicators: RPO and RTO
Any conversation about DR comes down to two concepts:
- RPO (Recovery Point Objective): how much data the company accepts losing. If backups run hourly, the RPO is 1 hour, meaning that in case of failure, at most one hour of work is lost.
- RTO (Recovery Time Objective): how long the company accepts being down until systems come back online. The value should be defined per system, based on the impact of downtime, and the chosen solution must be able to meet it; automated and tested recoveries are far faster and more predictable than manual processes.
In a managed services contract, it’s advisable that the RTO and RPO for each critical system be written as explicit targets. What used to be an aspirational goal should now be a contractual requirement.
The 3-2-1-1-0 rule
The classic 3-2-1 rule (three copies, on two types of media, one offsite) has evolved. The current version, now used as best practice in the industry, is the 3-2-1-1-0 rule:
- 3 copies of the data
- 2 different media types (for example, local NAS and cloud)
- 1 offsite copy
- 1 offline or air-gapped copy, inaccessible from the network (protection against ransomware)
- 0 errors verified after automated restore, meaning backups must be tested and verified
The final zero is the most overlooked and the most critical. A backup that was never tested is a promise, not a guarantee.
Today’s tools, by name
The backup and DR market has matured significantly. The main solutions used in SME contexts in 2026 are:
- Veeam Backup & Replication v12.2: native support for immutability on S3-compatible storage, integration with Microsoft Azure Stack HCI, and protection for Kubernetes workloads. Flexible licensing per socket or per workload via VUL (Veeam Universal License). Veeam Backup for Microsoft 365 v8 specifically protects Exchange Online, SharePoint, Teams and OneDrive, filling the retention gap left by Microsoft itself.
- Acronis Cyber Protect 16: combines backup, antivirus and EDR in a single console. The Cyber Protect Essentials plan is aimed specifically at SMEs. The Cloud version is widely used by Portuguese MSPs in resale mode.
- Nakivo Backup & Replication v11: strong in VMware vSphere 8 and Microsoft Hyper-V environments, with a physical appliance option for companies without a dedicated IT team.
- Azure Site Recovery (ASR): Microsoft’s cloud-native solution for replicating on-premises virtual machines to Azure, with an RTO of minutes. Integrated into Azure Arc for hybrid management.
- Zerto 10: journal-based replication with an RPO of seconds, currently positioned by HPE (which acquired the company in 2021) for the mid-market segment and SMEs with VMware infrastructure.
- Datto SIRIS 5 and Barracuda Backup: appliance-plus-cloud solutions with a strong presence in the European MSP channel. Datto was acquired by Kaseya in 2022.
For destination storage with data residency in the European Union, OVHcloud and Cloudflare R2 (no data egress costs) are gaining share among SMEs sensitive to data location.
MEGABIT, as a Sophos partner and cybersecurity specialist for businesses, regularly works with this ecosystem of tools, tailoring the choice to the size, sector and criticality of each organisation.
3. Practical scenarios: what can happen and how it’s resolved
The following scenarios are illustrative. They represent typical situations faced by companies in different sectors, without corresponding to specific client cases.
Scenario A: accounting firm with GDPR obligations
Imagine an accounting firm with 12 employees, regulated by the Ordem dos Contabilistas Certificados (OCC), that processes tax and financial data for dozens of clients. Files are kept in TOCONLINE, with occasional manual copies to a shared folder.
In a typical double-extortion ransomware attack scenario, data is exfiltrated and encrypted. The shared folder, synced with OneDrive, is also encrypted. Microsoft retains previous versions, but only for a limited period and without a guarantee of full recovery.
With an architecture based on Acronis Cyber Protect 16, with an immutable copy in Azure Blob Storage (Cool tier), and documented quarterly restore tests, the company is able to recover to the pre-attack state, meet the CNPD notification obligation within the 72-hour deadline required by the GDPR, and present documented evidence in case of an audit.
Scenario B: retail company with a local ERP
In a retail company with 20 employees, the Primavera BSS v10 ERP runs on a local physical server. A hardware failure on the server, without warning, brings the entire operation to a halt: no invoicing, no stock access, no order history.
With an architecture using Veeam Agent to run hourly incremental backups to a local NAS, with daily replication to OVHcloud Object Storage, the RPO is 1 hour and the RTO is 2 hours. The server is restored on a replacement machine or a temporary VM while the hardware is repaired.
Scenario C: medical clinic with data residency requirements
In an illustrative scenario of a medical clinic with a PACS (medical imaging) server and contractual obligations preventing data from leaving national territory, the solution involves replication with Zerto to a second site in colocation in Lisbon. The RTO required by the insurer is under 30 minutes. Data never leaves Portugal, meeting the contractual requirement with the SNS and CNPD guidelines.
Scenario D: industrial SME with a production line
In an industrial company with a production line controlled by SCADA systems, downtime is not just financial, it’s also a safety issue. The DR plan includes backup of Siemens TIA Portal configurations, regular snapshots of supervisory VMs via Nakivo, and storage of PLC configurations in a private Git repository on self-hosted GitLab. In the event of a failure, control systems are restored from validated versions, without relying on human memory or outdated documentation.
The CrowdStrike case: a warning about third-party dependency
In July 2024, a faulty update to the CrowdStrike Falcon agent caused massive downtime in Windows systems globally. SMEs that relied exclusively on this endpoint vendor had no clear recovery path for hours. The incident demonstrated that a DR plan must account for IT supply chain failures, not just internal failures, and that supplier diversification and rollback procedures are essential components of any robust plan.
4. Best practices and adoption checklist for 2026
Implementing a continuity and DR plan doesn’t require a large upfront investment. It requires method. The sequence below is practical guidance for SMEs starting out or reviewing their plan.
Step 1: Identify what cannot stop
Make a list of the systems and data critical to daily operations: ERP, CRM, email, client files, invoicing system. For each one, define a realistic RPO and RTO. Be honest about the financial and operational impact of every hour of downtime.
Step 2: Implement the 3-2-1-1-0 rule
Make sure each critical system has three copies, on two distinct media types, with at least one offsite copy and one immutable or air-gapped copy. Choose a tool suited to your infrastructure: Veeam, Acronis, Nakivo, or a solution managed by an MSP.
Step 3: Specifically protect Microsoft 365
If your company uses Microsoft 365, don’t assume your data is protected by Microsoft. Implement a third-party backup solution, such as Veeam Backup for Microsoft 365 v8, which covers Exchange Online, SharePoint, Teams and OneDrive. Microsoft guarantees the platform, not your data.
Step 4: Document the plan
A DR plan that exists only in one person’s head is not a plan. Document restore procedures, emergency contacts, access to tools, and who is responsible for each step. The document should be accessible even when the main systems are unavailable (a printed version or offline copy).
Step 5: Test, at least once a year
An untested backup is a promise without a guarantee. Tools like Veeam SureBackup and Acronis Disaster Recovery Runbooks enable automated tests without affecting production. Document the results, including for GDPR or DORA audit purposes.
Step 6: Review annually
Infrastructure changes, threats change, regulation changes. The DR plan should be reviewed at least once a year, or whenever a significant change occurs in systems, suppliers or the regulatory context.
Quick checklist
- RPO and RTO defined for each critical system
- 3-2-1-1-0 rule implemented
- Microsoft 365 backup with a third-party solution
- Immutable backups configured (an insurer requirement in 2024-2025)
- DR plan documented and accessible offline
- Restore test performed and documented in the last 12 months
- Responsible parties identified for each step of the plan
- Critical suppliers included in the plan (supply chain risk)
- Compliance verified against GDPR, NIS2 or DORA as applicable
5. The most common mistakes, and how to avoid them
Confusing backup with disaster recovery
Having backups is necessary, but not sufficient. DR also requires the ability to restore quickly, on alternative infrastructure, with tested procedures. A company that has backups but has never tested the restore process may take days to recover, even if the data is intact.
Treating OneDrive as a backup solution
OneDrive is a sync and collaboration tool, not a backup solution. Files deleted, corrupted or encrypted by ransomware can propagate to the cloud within seconds. Microsoft’s retention is limited, varies by service and plan, and does not cover every scenario. This is one of the most common mistakes among SMEs that adopted Microsoft 365 without proper technical guidance.
Ignoring cloud data egress costs
SMEs that replicate to AWS or Azure without considering data transfer costs during the recovery phase may face unexpected bills at precisely the worst moment. Providers like OVHcloud and Cloudflare R2 eliminate egress fees, which can be decisive when choosing an architecture.
Not accounting for external supplier failures
The CrowdStrike incident of July 2024 is the most recent example, but not the only one. Dependence on a single security software vendor, a single cloud provider, or a single MSP without a documented SLA, creates single points of failure that a robust DR plan must address.
Underestimating the impact of VMware/Broadcom
After Broadcom’s acquisition of VMware, licensing changed in 2024 with the end of the perpetual model and mandatory migration to subscriptions. SMEs with old VMware licences face uncertainty about support and compatibility with existing DR tools. Anyone who hasn’t yet reviewed their virtualisation architecture should do so urgently.
Not documenting for regulatory purposes
DORA requires annually tested plans and documentation of ICT incidents. It’s not enough to have a plan, you must be able to prove it exists and works.
6. Executive summary
Business continuity and disaster recovery are no longer optional for Portuguese SMEs. The convergence of a growing threat (ransomware, supplier failures, human error), stricter regulation (DORA, NIS2, GDPR) and an insurance market that now demands proof of resilience has created a new minimum standard.
The tools are available, accessible and suited to the reality of SMEs: Veeam Backup & Replication v12.2, Acronis Cyber Protect 16, Nakivo v11, Azure Site Recovery and Veeam Backup for Microsoft 365 v8 cover most scenarios with different combinations of cost, complexity and level of protection.
The three immediate steps for any SME manager are: define RPO and RTO for critical systems; check whether Microsoft 365 has third-party backup; and schedule a restore test within the next 90 days. The rest builds from there.
Frequently asked questions
My company uses Microsoft 365. Isn’t that enough to keep my data protected?
No. Microsoft guarantees platform availability, but not the recovery of data deleted by the user or by ransomware beyond a limited period, which varies by service and plan. Data deleted, corrupted or encrypted by malware may be unrecoverable without a third-party backup solution such as Veeam Backup for Microsoft 365 v8 or Acronis Cyber Protect Cloud.
What do RPO and RTO mean, and why do they matter for my SME?
RPO (Recovery Point Objective) is the maximum amount of data your company accepts losing, measured in time. If you back up hourly, your RPO is 1 hour. RTO (Recovery Time Objective) is the maximum acceptable time to get systems back up and running after a failure. The value should be defined per system, based on the impact of downtime; automated and tested recoveries are far faster and more predictable than manual processes. Defining these two values is the first step of any DR plan.
Do DORA and NIS2 apply to my company?
DORA, in force since 17 January 2025, applies to regulated financial entities, including fintechs, asset managers and payment providers. NIS2, transposed into Portuguese law by Decree-Law No. 125/2025, covers sectors such as healthcare, energy, transport and digital service providers, but indirectly affects SMEs that are suppliers to essential entities, through supply chain requirements. The GDPR, under Article 32, requires any company processing personal data to have documented restoration capability, which includes practically all SMEs.
How MEGABIT can help your company
MEGABIT provides IT support for companies in Porto and Greater Porto, specialising in infrastructure, backup, cybersecurity and Microsoft 365. As a Sophos partner and Microsoft partner (MCSE, MCSA and MCTS certified), our team is qualified to assess the maturity of your continuity plan and propose solutions suited to your company’s reality, with no fixed monthly fee and no tie-in contract.
If your company doesn’t yet have a formalised DR plan, if the last restore test was over a year ago, or if you have doubts about compliance with GDPR, NIS2 or DORA, the next step is a no-obligation conversation.
Our Hour Bundle model gives you flexible access to specialised expertise, using support when you need it, with no fixed structures.
Contact us: comercial@megabit.pt
Want to be notified when we publish our next article or video?
Free. Get notified as soon as we publish. No spam, easy to unsubscribe.


