A practical backup and disaster recovery (DR) plan for an Indian SME rests on one rule and two numbers. The rule is 3-2-1: three copies of important data, on two different media, with one copy off-site and ideally offline or immutable. The two numbers are how much data you can afford to lose (recovery point objective, RPO) and how long you can be down (recovery time objective, RTO). Cloud backup is now affordable at small-business scale. IDrive's pricing page, for example, shows a free 10 GB Basic tier and paid Personal, Team and Business plans priced in US dollars. The more expensive failure is untested restores. CERT-In's April 2022 Directions list ransomware among reportable incidents and require reporting within 6 hours, so recovery has to be planned, not improvised.
Executive summary
- Sync is not backup. Google Drive, OneDrive and similar tools replicate deletions and ransomware-encrypted files. You need independent, versioned copies.
- Set RPO and RTO per system, not per company. Accounting and ERP data usually need a daily or better RPO. An archive folder may be fine with weekly.
- Official prices are mostly in USD. IDrive publishes USD plans with first-year discounts. Veeam, Acronis and many enterprise tools did not publish SME prices on the pages we could access, so quotes are needed.
- Compliance links backup to incident response. CERT-In requires 6-hour incident reporting and 180-day log retention within India. The DPDP Act requires reasonable security safeguards for personal data, which includes being able to restore it without exposing it.
- Test restores are the KPI that matters. A backup never restored in a test should be treated as unproven.
Market context
Regulatory drivers:
- The CERT-In Directions of 28 April 2022 apply to service providers, intermediaries, data centres, body corporate and government organisations. They require incident reporting within 6 hours, secure retention of ICT logs for a rolling 180 days within Indian jurisdiction, and NTP time synchronisation with NIC/NPL-traceable servers. For backup planning, this means your logs need protecting too: an attacker who deletes logs also removes your evidence.
- The DPDP Act, 2023 and the DPDP Rules, 2025 put security and breach-handling duties on data fiduciaries. The Rules follow an 18-month phased timeline (PIB). Backups are copies of personal data, so they need the same access control, encryption and retention discipline as production systems.
- Where you use Indian cloud or VPS providers, the same Directions require those providers to keep validated subscriber information for 5 years. Expect KYC when you sign up for backup storage from an Indian provider.
Threat context. We do not quote private-survey statistics on ransomware in India. Qualitatively, SMEs are exposed through phishing, unpatched remote access and shared admin passwords. Ransomware increasingly targets backup systems first, which is why immutable or offline copies matter.
Evaluation framework
App Advisor's methodology for SME backup and DR:
| Criterion | Weight | What we look for |
|---|---|---|
| Recovery capability | 25% | Granular file restore, full system (bare-metal) restore, VM restore, restore speed |
| Ransomware resilience | 20% | Immutable storage, offline/air-gapped copy, MFA on the backup console, anomaly alerts |
| Coverage | 15% | Laptops, servers, NAS, Microsoft 365/Google Workspace, databases, SaaS apps |
| Cost predictability | 15% | Per device vs per GB, egress/restore fees, currency, renewal |
| Compliance and data location | 15% | Storage region choices, encryption keys, audit logs, retention policies |
| Ease of operation | 10% | Central console, reporting, alerting, India-hours support |
Vendor comparison
Tool types
| Need | Tool type | Examples on App Advisor |
|---|---|---|
| Laptops and desktops | Cloud endpoint backup | IDrive, Acronis Cyber Protect |
| Servers and virtual machines | Image-based backup and replication | Veeam, Acronis, Iperius Backup |
| Microsoft 365 / Google Workspace | SaaS backup | Spanning Backup, Backupify |
| Cloud workloads | Native cloud backup | AWS Backup |
| Managed service providers | Multi-tenant backup | Comet Backup |
Official prices we could verify
| Product | Plan | Official price shown | Notes | Source |
|---|---|---|---|---|
| IDrive | Basic | Free, 10 GB, no credit card required | Single user | IDrive pricing |
| IDrive | Personal 5 TB | $9.99/month, shown discounted to $6.99/month | Discount applies to the first year; standard pricing after | IDrive pricing |
| IDrive | Team 5 computers, 5 users (5 TB) | $9.99/month, shown discounted to $6.99/month | First-year discount | IDrive pricing |
| IDrive | Business 500 GB (unlimited users, servers, NAS) | $19.99/month, shown discounted to $13.99/month | First-year discount; larger tiers available | IDrive pricing |
| Veeam | Microsoft 365 backup / Data Platform | Not published on the page we could access | Quote | Veeam |
| Acronis Cyber Protect | — | Not published on the page we could access | Quote / partners | Acronis |
| AWS Backup | Pay-as-you-go by region | Varies by service and region; use the pricing page with the India region selected | Metered | AWS Backup pricing |
IDrive's page states that "the discounted price applies to the first year of the selected plan" and that standard pricing applies afterwards. Budget on the standard price.
Browse the backup software category, the cloud storage category and the IT management category.
Setting RPO and RTO by system
RPO (recovery point objective) is the maximum age of data you can accept after a restore. RTO (recovery time objective) is the maximum time a system can be unavailable. The example below shows how a typical Indian SME might set them. These are illustrative starting points to discuss with business owners, not standards.
| System | Business impact if down | Example RPO | Example RTO | Suggested approach |
|---|---|---|---|---|
| Accounting / ERP (Tally, cloud ERP) | Invoicing, GST filing and payments stop | 1–4 hours | 4–8 hours | Frequent backups plus a local copy for fast restore |
| Customer communication disrupted | 24 hours | 4 hours | Native retention plus SaaS backup if required | |
| File server / shared drives | Project work stops | 24 hours | 1 day | Nightly versioned backup, immutable copy |
| CRM / helpdesk (SaaS) | Sales and support slow down | 24 hours | 1 day | Vendor export or SaaS backup |
| Website | Lead flow drops | 24 hours | 4–8 hours | Host snapshots plus off-site copy |
| Laptops | Individual productivity | 24 hours | 1–2 days | Back up business folders only |
| Security and system logs | Investigation impossible | Near real-time | Not user-facing | Central log store, 180-day retention in India |
Ransomware-resilient backup design
Ransomware operators try to find and delete backups before encrypting production systems. Design with that assumption:
- Separate identities. Backup admin accounts are not domain admin accounts and use their own MFA.
- Immutability. At least one copy is stored where it cannot be changed or deleted for a set period, even by an administrator.
- Offline or logically air-gapped copy. A copy that is disconnected from the network, or in a separate account with no standing access.
- Alerting. Alerts on failed jobs, sudden drops in backup size, mass deletions and changes to retention settings.
- Clean restore. Restore into an isolated environment first, and scan before reconnecting to the network.
Backing up Microsoft 365 and Google Workspace
Suite vendors provide retention features and version history, but those features are designed around their own service. Many businesses add a third-party SaaS backup because of:
- Accidental or malicious deletion that passes retention windows before anyone notices.
- Departing employees whose licences are removed, deleting their data.
- Legal holds or audits needing point-in-time restores of mailboxes or drives.
- Account compromise, where an attacker deletes data from inside a valid account.
Decide based on your retention policy and risk assessment, and document the decision.
The disaster recovery runbook
A DR runbook should fit on a few pages and be printed or stored outside the systems it protects. Include:
- Contact list: owner, IT partner, backup vendor support, internet service provider, and the CERT-In reporting route.
- Decision tree: who declares a disaster, and when to fail over or restore.
- Restore order: identity and network first, then accounting/ERP, email, file shares, and finally laptops.
- Credentials access: how to reach backup consoles if the main password manager or email is unavailable.
- Communication templates: messages for staff, customers and suppliers.
- Evidence preservation: which logs and disk images to secure before rebuilding.
Total cost of ownership
Illustrative model — assumptions stated. A 40-person CA firm with 40 laptops, one file server (1.5 TB) and Microsoft 365 mailboxes. Assumptions:
- Laptop user data averages 40 GB per person (1.6 TB total), mostly documents. Server data 1.5 TB.
- Option A: cloud backup for laptops and server plus a local NAS copy. Cloud subscription cost is an assumed ₹25,000/month, pending quotes. NAS hardware is an assumed ₹1,50,000 one-time, depreciated over 3 years.
- Option B: local NAS only, with weekly USB drives taken home by a partner. Hardware is an assumed ₹1,80,000 one-time over 3 years.
- Staff/partner time: 4 hours/month for A, 10 hours/month for B (manual rotation and checks), at an assumed ₹1,200/hour.
- Downtime cost: an assumed ₹60,000 per working day of full outage (billing staff idle plus penalties). Assumed recovery after a ransomware event: 1 day for A, 5 days for B, because offsite USB copies are older and slower to restore.
- One ransomware-type event assumed in 3 years for both options.
| 3-year cost line | Option A: 3-2-1 with cloud | Option B: NAS + USB rotation |
|---|---|---|
| Subscriptions | ₹25,000 × 36 = ₹9,00,000 | ₹0 |
| Hardware | ₹1,50,000 | ₹1,80,000 |
| Staff/partner time | 4 × ₹1,200 × 36 = ₹1,72,800 | 10 × ₹1,200 × 36 = ₹4,32,000 |
| Downtime from one incident | 1 × ₹60,000 = ₹60,000 | 5 × ₹60,000 = ₹3,00,000 |
| 3-year total | ₹12,82,800 | ₹9,12,000 |
| Data loss exposure (RPO) | Hours | Up to 1 week |
Interpretation:
- On this model the manual option is cheaper in rupees, but it can lose up to a week of client work. For a CA firm in filing season, that risk may be unacceptable. Put a value on lost work before deciding.
- The assumed cloud subscription is the biggest line. Get real quotes, and back up only business data (not whole laptops) to reduce storage.
- A hybrid is common: cloud backup for the server and critical folders, and a local NAS for fast restores.
Questions to ask backup vendors
- Which storage regions are available, and can we choose India?
- Is immutable storage supported, and for how long can it lock backups?
- Are there charges for restores, downloads or data egress?
- How is encryption handled, and who holds the keys?
- What happens to our backups if the subscription lapses, and for how long are they kept?
- Can we do a full system restore to different hardware or to a virtual machine?
- What reports show job success, protected devices and storage used?
- Is the price in the order form fixed after any first-year discount ends?
Common mistakes
- Backing up to a USB drive that stays permanently connected, so ransomware encrypts it too.
- Backing up only the file server and forgetting the accounting database, or the reverse.
- Keeping backup console credentials in the same email account an attacker could compromise.
- Assuming the SaaS vendor "backs up everything" without reading the retention terms.
- Never timing a full restore, so the RTO is a guess.
Implementation roadmap
Weeks 1–2: Business impact analysis
- List systems: accounting/ERP, file server, email, CRM, website, laptops.
- Assign RPO and RTO to each with the business owner, not just IT.
- Identify personal data sets and retention rules.
Weeks 3–4: Design
- Apply 3-2-1, and add immutability or an offline copy for at least one set.
- Separate backup admin credentials from domain/admin accounts; enforce MFA.
- Choose storage regions; confirm location in writing where it matters.
- Include logs in scope, so 180 days of logs survive an incident.
Weeks 5–7: Deploy
- Roll out agents to servers first, then laptops.
- Back up Microsoft 365 or Google Workspace with a SaaS backup tool if your risk assessment requires it.
- Configure alerts for failed jobs and unusual deletion volumes.
Weeks 8–9: Test
- Restore a single file, a mailbox, and a full server to a test environment.
- Time each restore against the RTO. Adjust the design if you miss it.
Ongoing
- Monthly job review; quarterly restore test; an annual DR exercise combined with the CERT-In incident drill.
Risks and compliance checklist
- 3-2-1 with an immutable or offline copy in place.
- Backup console protected with MFA and separate credentials.
- CERT-In 6-hour reporting: the DR plan starts the reporting clock on ransomware and data-loss incidents.
- 180-day ICT logs within India backed up or otherwise protected from deletion.
- NTP-synced timestamps on backup and security logs.
- DPDP Act: backups encrypted; access restricted; retention aligned with purpose; deletion requests handled across backups per documented policy.
- Data localisation: storage region confirmed where contracts or sector regulators require India.
- Restore tests documented quarterly with actual times.
- Renewal and currency: USD-priced plans budgeted at post-discount standard prices plus a forex buffer.
KPIs to track
| KPI | Suggested target |
|---|---|
| Backup job success rate | Above 98% per week, with failures fixed within 24 hours |
| Systems meeting RPO | 100% of critical systems |
| Tested restore time vs RTO | Within RTO for every critical system |
| Restore tests per quarter | At least 1 file, 1 mailbox, 1 full system |
| Endpoints protected | 100% of company laptops |
| Immutable/offline copy age | Within policy (for example, under 24 hours for critical data) |
How to choose
- Microsoft or Google suite users: decide whether native retention meets your RPO and legal needs. If not, add SaaS backup.
- Laptop-heavy teams: cloud endpoint backup (such as IDrive or Acronis) with central management.
- Server and VM workloads: image-based tools (such as Veeam or Acronis) with replication to a second site or cloud.
- Workloads on AWS: native AWS Backup with cross-account copies, in the India region.
- No IT staff: a managed backup service with monthly restore reports.
Related guides: cybersecurity for Indian SMEs, cloud storage pricing in India and how to choose the right IT infrastructure.
FAQs
What is the 3-2-1 backup rule?
Keep three copies of data, on two different types of media, with one copy off-site. Modern guidance adds that at least one copy should be offline or immutable, to resist ransomware.
Is Google Drive or OneDrive a backup?
Not by itself. Sync tools replicate deletions and encrypted files. Version history helps, but an independent backup is safer.
How much does IDrive cost?
IDrive's pricing page shows a free 10 GB Basic tier, and paid plans in USD such as Personal 5 TB at $9.99/month shown discounted to $6.99/month, and Business 500 GB at $19.99/month shown discounted to $13.99/month. Discounts apply to the first year.
What RPO and RTO should an SME set?
It depends on the business. Many firms set a same-day or better RPO and a 1-day RTO for accounting and billing systems, and longer targets for archives. Agree them with business owners.
Do CERT-In rules affect backup?
Indirectly but importantly. Ransomware and data breaches are reportable incidents with a 6-hour deadline, and ICT logs must be kept for 180 days within India. Your backup and DR design should protect logs and support fast investigation.
Should backups be stored in India?
CERT-In requires logs to be kept within Indian jurisdiction. For other data, check contracts and sector regulations. Choose an India region when in doubt, and confirm it in writing.
How often should we test restores?
At least quarterly for critical systems, and after any major change.
What is an immutable backup?
An immutable backup is stored so that it cannot be changed or deleted for a set retention period, even by someone with administrator access. It protects against ransomware operators who try to delete backups before encrypting production systems.









