You don’t get a second chance after a bad restore. This guide shows how to set a practical backup cadence for Umbraco, what to back up and when, and how Azure hosting compares with Umbraco Cloud so you can cut risk without overspending.
At a glance
- How backup frequency ties directly to your RPO and RTO
- What to back up in Umbraco — database vs media — and how often
- Azure Umbraco hosting vs Umbraco Cloud for automation, retention and testing
- Real scenarios with restore times and cost angles
- A clear playbook to test restores and tighten governance
Why backup frequency matters for Umbraco hosting
Frequency sets your Recovery Point Objective (RPO), the data you’re prepared to lose, and influences the Recovery Time Objective (RTO), how fast you can be back online. Daily backups on a busy site still risk hours of lost edits or orders.
A sound plan balances:
- Risk tolerance: acceptable data loss in minutes or hours
- Cost: storage, egress and operations
- Restore speed: smaller, incremental backups restore faster than monolithic snapshots
Declare your RPO and RTO up front. Then buy only the backup frequency and retention needed to hit them.
How often should you back up an Umbraco site
Here’s practical guidance you can tune to your release cadence and change rate:
- Enterprise Umbraco platforms
- Database: Hourly or every few hours
- Media: Daily with incremental or versioned storage for large libraries
- Corporate marketing sites
- Database: Daily
- Media: Weekly or daily if editors upload often
- E‑commerce or campaign sites
- Database: Hourly during peak trading or campaigns
- Media: Daily with fast partial restore capability
Testing matters as much as frequency and a backup only counts once you’ve restored it cleanly and quickly.
Azure Umbraco hosting vs Umbraco Cloud backups
Azure gives you granular control over schedules, retention and geo‑recovery. Umbraco Cloud bakes in automated database PITR and filesystem/blob snapshots with managed retention. Pick based on how often you change content, compliance needs and how hands‑on you want to be.
Feature comparison
Decision criteria | Azure Umbraco hosting | Umbraco Cloud |
---|---|---|
Backup frequency | Fully configurable schedules to align with RPO targets | Automated 35‑day point‑in‑time database recovery and managed filesystem/blob snapshots |
Database backups | Azure SQL automated full, differential and log backups enable point‑in‑time restore within a configurable retention window | 35‑day point‑in‑time database restore and downloadable BACPAC backups |
Media backups | Blob snapshots or versioning with retention policies; supports incremental behaviour at block level | 30‑day filesystem snapshots and 35‑day Blob Storage container snapshots for DR |
Retention | Short‑term PITR typically 7–35 days; LTR up to 10 years if required | Managed by Umbraco Cloud within published policy |
Restore testing | Staging restores and cross‑region drills possible | Managed restores via Umbraco tools and Deploy workflows |
- Azure SQL keeps weekly full, daily differential and log backups every 5–10 minutes for PITR within retention. This underpins sub‑hour RPOs when scheduled correctly. (Source: Microsoft Community Hub)
- Short‑term PITR retention is configurable and commonly 7–35 days; long‑term backup retention can extend to 10 years when enabled. (Source: Microsoft Azure Blog) (Source: Microsoft Community Hub)
- Blob snapshots create point‑in‑time, read‑only copies of media; blob versioning offers an alternative for automatic prior versions. (Source: Microsoft Learn)
- Umbraco Cloud provides 35‑day point‑in‑time database recovery, 30‑day filesystem snapshots and 35‑day blob container snapshots for DR. (Source: Umbraco Docs)
If your editors publish hourly, a daily database backup sets you up to lose a day’s work. Tighten the schedule or accept the risk — but document the trade.
What good looks like for RPO and RTO
- RPO targets
- Content‑led sites: ≤ 24 hours
- Transactional or frequently edited content: ≤ 1 hour
- RTO targets
- Minor rollbacks: ≤ 30–60 minutes to restore to staging, validate, then swap
- Major incident: ≤ 4 hours with a clear runbook
These targets are achievable with Azure SQL PITR and Blob Storage snapshots or versioning, given the built‑in backup cadence and restore mechanics. (Source: Microsoft Azure Blog) (Source: Microsoft Learn)
Real‑world scenarios
Hourly backups protect financial data
A financial services Umbraco platform scheduled hourly Azure SQL PITR. During a failed import, the database rolled back 45 minutes with zero data loss and minimal disruption.
Large media library with incremental behaviour
A global corporate site with a 200 GB media library used Azure Blob snapshots and versioning. Backup windows dropped and partial restores targeted only changed blocks for faster recovery. (Source: Microsoft Learn)
Umbraco Cloud automated restore
A charity on Umbraco Cloud restored to the previous day’s backup in under 30 minutes after a content error, avoiding extended downtime thanks to managed PITR and snapshots. (Source: Umbraco Docs)
ROI and operational impact
Business factor | Azure Umbraco hosting | Umbraco Cloud |
---|---|---|
Data loss prevention | Hourly or sub‑hour RPO with log backups supporting PITR | Daily cadence with 35‑day PITR keeps rollback simple |
Cost control | Tune schedules, tiers and retention to TCO goals | Backups included in plan reduce internal overhead |
Compliance | LTR up to 10 years, regional controls and auditability | Managed retention and DR snapshots within published policy |
Team workload | Self‑managed or managed service from Growcreate | Mostly hands‑off with Cloud’s automation |
Start with the cost of a lost hour of content or orders. Set RPO to cost, not to habit.
How to set a reliable Umbraco backup plan
- Set business‑level thresholds
- Define RPO in minutes and RTO in hours. Put them in your runbook.
- Choose hosting fit
- Need granular control, hybrid setups or long‑term retention audits? Azure fits.
- Prefer baked‑in automation with managed retention? Umbraco Cloud is simpler.
- Pick database cadence
- Azure SQL PITR supports weekly full, daily differential and 5–10 minute logs — align to your RPO. (Source: Microsoft Community Hub)
- Protect media at scale
- Use Blob snapshots or versioning with lifecycle rules to prune old versions. (Source: Microsoft Learn)
- Plan for regional incidents
- Enable cross‑region restore drills on Azure Backup vaults using GRS so you can restore in the paired region without waiting for a formal outage. Be aware of replication latency before secondary restore points become available. (Source: Microsoft Azure Updates) (Source: Microsoft Learn)
- Lock retention to policy
- Map short‑term (7–35 days) and long‑term (up to 10 years) to regulatory needs. (Source: Microsoft Azure Blog) (Source: Microsoft Community Hub)
Restore testing you can run quarterly
Run this in a staging slot or non‑prod environment:
- Select a restore point matching your declared RPO.
- Restore the database to staging. Time the operation end‑to‑end. Log start, ready, swap.
- Restore a media subset using blob snapshot or versioned object copy.
- Validate application health — logins, permissions, content cache, search index, integrations.
- Perform a swap or data merge if needed, then archive the test artefacts.
- Record the actual RTO. If it exceeds target, tighten backup size, add parallel steps or revisit infra.
Governance, risk and compliance guardrails
- Retention & GDPR: Keep only what your policy demands. Short‑term PITR for rapid rollback, LTR for audits. Document purge schedules.
- Isolation: Store backups separate from production access paths. Restrict keys and RBAC.
- Geo‑redundancy: Use GRS for backups and test cross‑region restore drills at least twice a year. Expect secondary RPO lag due to replication. (Source: Microsoft Learn)
- 24/7 runbook: Name the people, tools and timelines. Practice handovers.
Cost levers that move the needle
- Storage tiering for media versions and snapshots
- Shorter retention for non‑prod
- Scheduled throttling of backup jobs during peak hours
- Differential or incremental patterns to cut backup window
How Growcreate manages Umbraco backup strategies
- Custom backup schedules mapped to your edit cadence and risk tolerance
- Geo‑redundant storage with cross‑region drills where required
- Staging restore testing on a schedule, with RTO benchmarks and remediation
- Hybrid options combining Umbraco Cloud daily automation with Azure incremental media protection
- 24/7 SLA‑backed recovery — engineers on call for urgent restores
Our strategic outlook
Backups are now programmatic and testable. Treat them like code. Set bold but realistic thresholds for RPO and RTO, automate the cadence that meets them and schedule restores as drills. Do that and you’ll cut downtime, avoid data loss and build things that grow with you.
Book a quick call to see how a tailored Umbraco hosting backup frequency plan protects your data and keeps your platform resilient.
FAQs
With Azure SQL log backups every 5–10 minutes and an hourly schedule, RPO ≤ 1 hour is realistic for most sites. (Source: Microsoft Community Hub)
If audits or contracts require it, enable up to 10 years LTR on Azure SQL. Keep daily PITR short to control cost. (Source: Microsoft Community Hub)
Use Blob snapshots or versioning. They create point‑in‑time copies of files and support targeted restores without pulling the entire library. (Source: Microsoft Learn)
35‑day point‑in‑time database restore plus 30‑day filesystem and 35‑day blob snapshots for DR. (Source: Umbraco Docs)
Yes. Azure Backup’s cross‑region restore lets you run drills in the paired region without an active outage. Plan for replication lag before restore points appear in the secondary region. (Source: Microsoft Azure Updates) (Source: Microsoft Learn)
Start time, restore duration, validation checklist, swap time, issues found, owner, next action. Compare against your RTO and fix gaps.