A website backup strategy is a documented plan that defines what parts of a site get copied, how often each copy runs, where copies are stored, how long they are kept, and how restores are tested. The standard framework is the 3-2-1 rule: three copies of your data, on two storage types, with one copy offsite.

Most site owners already have backups. Far fewer can restore from them under pressure. This guide covers the full plan, including schedules by site type, recovery targets, retention rules, and a restore test you can run this week.
Key Takeaways
- Follow the 3-2-1 rule: 3 copies, 2 storage types, 1 offsite copy in an account your web server cannot access.
- Back up files, the database, email, DNS records, and configuration as one recovery set.
- Set backup frequency from your recovery point objective (RPO), not from a default setting.
- Take a manual snapshot before every core, plugin, theme, or PHP update.
- Keep daily, weekly, and monthly versions so you can restore to a point before a problem started.
- Test a full restore at least once per quarter and time it.
What Is a Website Backup Strategy?
Quick Answer: A website backup strategy is a written plan covering five decisions: what to copy, how often, where copies live, how long they are kept, and how restores are verified. Backup tools create the files. The strategy decides whether those files can actually bring the site back after a hack, crash, or bad update.
A website backup strategy covers files, databases, email, and configuration, and it assigns a schedule, a storage location, and an owner to each one. A single backup file is a snapshot. A strategy is the system that keeps snapshots current, protects them from attackers, and proves they restore.
The table below shows the five decisions every backup plan must answer.
| Decision | Question It Answers | Example Answer |
| Scope | What gets copied? | Files, database, email, DNS zone, server configuration |
| Frequency | How often does each part get copied? | Database every hour, files once per day |
| Location | Where do copies live? | Hosting server, separate cloud storage account, encrypted local drive |
| Retention | How long are copies kept? | 7 daily, 4 weekly, 12 monthly versions |
| Verification | How do you prove copies restore? | Quarterly test restore in a staging environment |
Backups vs. a Backup Plan
A backup is a single copy of website data, while a backup plan is the process that creates, protects, and tests those copies on a schedule. A backup answers “do I have a copy?” A backup plan answers “can I get the site back, how fast, and with how much data lost?”
The difference matters during an incident. A site owner with backups but no plan has to work out restore steps while the site is down. A site owner with a plan follows a written runbook.
Where Host Backups Fit in a Website Backup Strategy
Host backups fit into a website backup strategy as one copy among several, not as the entire plan. The official WordPress documentation notes that most hosts back up the entire server, but requesting a copy of your site from those backups takes time, so site owners should learn to back up and restore their own files (WordPress.org backup documentation).
Host backups usually live inside the host’s environment. A suspended account, a lapsed payment, or a compromised hosting login can block access to the live site and its backups at the same moment. That is why the 3-2-1 rule requires at least one copy under separate credentials.
Responsibility is shared. The host protects the infrastructure. The site owner decides what data matters, how much loss is acceptable, and where the independent copy lives.
At Prestige Technologies, our SMB Hosting plans include automated weekly website backups, and we treat them as the baseline copy in your 3-2-1 plan. We encourage every client to keep at least one more copy in a storage account they control. Our 24/7 chat support team can answer questions while you set up that second copy.
Why Do Website Backups Fail When You Need Them?
Quick Answer: Website backups usually fail for predictable reasons: copies sit on the same server as the site, scheduled jobs stop without warning, the backup misses the database or uploads folder, or nobody has ever tested a restore. Attackers also target backups directly, which makes storage location as important as backup frequency.
Website backups fail when they are needed because most failures sit in the setup, not in the backup software. Insurance claims data shows the gap. Cyber insurer At-Bay analyzed 186 ransomware claims involving backups and reported in October 2023 that 92% of businesses had backups, yet 31% could not restore data from them (At-Bay backup report). Businesses that did restore had claims that were 41% less severe and were 3 times less likely to pay a ransom.
Attackers treat backups as a target. In a March 2024 report covering 2,974 organizations hit by ransomware, Sophos found that 94% said attackers tried to compromise their backups (Sophos research on compromised backups). Median recovery costs reached $3 million for organizations with compromised backups, compared with $375,000 for organizations whose backups stayed intact. Only 26% of organizations with compromised backups fully recovered within a week, compared with 46% of organizations with intact backups.
The table below lists the most common failure modes and the control that prevents each one.
| Failure Mode | What Happens | Prevention |
| All copies on one server | A server failure or compromise destroys the site and its backups together | Keep one copy in a separate account and location |
| Silent job failures | A scheduled backup stops and nobody notices for weeks | Turn on failure alerts and review backup logs weekly |
| Incomplete scope | The database restores but the uploads folder is missing, so images break | Back up files and database as one set |
| Mismatched files and database | Plugin versions do not match the database schema, causing errors | Capture both in the same window, database first |
| Archives in the public web root | A predictable filename exposes the full database | Store archives outside the document root |
| Infected backups | The restored site carries the same malware | Keep older versions and scan before restoring |
| Untested restores | A corrupt archive is discovered during an outage | Schedule test restores |
| Plugin timeouts | A PHP time limit cuts off a large archive mid-run | Use server-level backups or chunked exports |
Free malware scanning on our hosting plans helps you spot infections sooner. The sooner you spot an infection, the more clean backup versions you still have to choose from when you restore. Our web application firewall and DDoS protection add filtering in front of the site, which reduces the number of incidents that end in a restore at all.
What Should a Website Backup Include?
Quick Answer: A complete website backup includes the site files, the full database, email data if mail is hosted on the same account, DNS and server configuration, and the credentials needed to rebuild. Missing any one of these turns a restore into a partial rebuild, which is where most recovery time gets lost.
A website backup should include every component needed to rebuild the site on a clean server without guesswork. The table below maps each component to what it contains and what breaks without it.
| Component | What It Contains | How Often It Changes | What Breaks Without It |
| Website files | CMS core, themes, plugins, custom code, uploads | Weekly to monthly, uploads more often | Design, functionality, and all media |
| Database | Posts, pages, settings, users, orders, form entries | Hourly to daily on active sites | All content and transactions |
| Email data | Mailboxes, folders, filters, forwarders | Continuously | Message history and mail routing |
| DNS zone | A, CNAME, MX, and TXT records | Rarely | Site resolution and email authentication |
| Server configuration | Redirect rules, cron jobs, PHP settings | Rarely | Redirects, scheduled tasks, compatibility |
| Credentials and keys | Logins, API keys, licenses, encryption keys | When rotated | The ability to rebuild or decrypt |
Website Files
Website files include the CMS core, themes, plugins, custom code, and every uploaded image, PDF, and media file. In WordPress, the wp-content folder holds themes, plugins, and uploads, and the wp-config.php file holds the database connection settings and security keys.
The uploads folder is usually the largest part of the file backup. It is also the hardest to replace, because product photos, blog images, and downloadable files often exist nowhere else.
The Database
The database stores posts, pages, settings, user accounts, comments, orders, and form entries. On active sites, the database changes more often than any other component.
WordPress documentation points out that the database lives outside the WordPress directory, so downloading the site folder does not capture it. A database backup is an export file, often called a dump, created with tools such as phpMyAdmin, mysqldump, or WP-CLI. The same documentation recommends backing up the database first, then the files, and treating both as one backup set created around the same time.
Email Accounts and Mail Data
Email data belongs in a website backup when mailboxes are hosted on the same account as the website. Mail data includes message folders, filters, forwarders, autoresponders, and the mailbox list itself.
Mailboxes accessed by POP may already store messages only on local devices. Mailboxes accessed by IMAP keep the primary copy on the server, so a server-side backup or a scheduled export protects them. If your team uses our business email hosting or the mailboxes included with our hosting plans, add each mailbox to your backup inventory and export important folders on a set schedule.
Our standalone Business Mail plans include 1 GB, 2 GB, or 5 GB of mailbox storage depending on the tier. Regular exports of older mail protect your message history and keep mailboxes within that storage allocation. SMB Hosting plans include free, unlimited mailboxes, so the same export routine applies to every address you create.
DNS Records, SSL Certificates, and Server Configuration
DNS records, SSL certificates, and server configuration are the settings that connect a domain to the site and keep it behaving correctly. Export your DNS zone file and store it with your backups. Losing TXT records breaks SPF, DKIM, and DMARC, which causes legitimate email to land in spam or bounce.
Server configuration includes redirect rules, rewrite files such as .htaccess, cron jobs, PHP version settings, and any custom headers. These settings are small, but rebuilding them from memory during an outage adds hours.
Every one of our SMB Hosting plans includes free SSL, so a site rebuilt on our platform gets a certificate again without a separate purchase. You still need your exported DNS zone to restore mail and verification records.
Credentials, Keys, and Licenses
Credentials, keys, and licenses are the access details required to rebuild a site from its backups. Store them in a password manager or secrets vault, not inside the backup archive.
The list should cover the hosting account, domain registrar, DNS provider, CDN, payment gateway API keys, premium plugin and theme licenses, and the backup encryption key. A backup encrypted with a key you cannot find cannot be restored.
The 3-2-1 Backup Rule for Websites
The 3-2-1 backup rule for websites means keeping three copies of site data, on two different storage types, with one copy stored offsite. Photographer Peter Krogh popularized the rule for digital asset management, and it is now a standard baseline in IT data protection.
- 3 copies: The live site plus two independent backups.
- 2 storage types: Two separate systems, so one platform failure cannot take out every copy.
- 1 offsite copy: A copy stored outside the hosting environment, under credentials the web server does not hold.
How to Apply 3-2-1 to a Website
Applying 3-2-1 to a website means mapping each copy to a specific system and a specific set of credentials. For websites, “two storage types” rarely means tape. It means two independent systems that do not share a failure point.
| Copy | Where It Lives | What It Protects Against |
| Copy 1: Live site | Hosting server | Nothing on its own; this is production data |
| Copy 2: Host-level backup | Host backup storage | Bad updates, accidental deletions, file corruption |
| Copy 3: Offsite backup | Separate cloud storage account or encrypted local drive | Account lockouts, host outages, compromised hosting credentials |
One rule makes this work: if every copy can be reached with one login, the copies count as one copy during a breach. Use a separate account, a separate password, and multi-factor authentication for the offsite destination.
On a Prestige Technologies hosting account, your live site and our automated weekly backups cover copies one and two. Your offsite copy completes the rule.
The 3-2-1-1-0 Rule for Ransomware Resilience
The 3-2-1-1-0 rule extends 3-2-1 with one offline or immutable copy and zero errors in restore verification. The extra “1” protects against attackers who reach every connected copy. The “0” means each backup is checked and test-restored until it produces no errors.
This extension follows government guidance. The #StopRansomware Guide from CISA, the FBI, the NSA, and MS-ISAC tells organizations to maintain offline, encrypted backups of critical data and to regularly test their availability and integrity, because many ransomware variants search for and then delete or encrypt accessible backups (CISA #StopRansomware Guide).
Immutable and Offline Backup Copies
Immutable backups are copies that cannot be changed or deleted until a set retention period ends, and offline backups are copies with no live network connection. Both prevent an attacker with stolen credentials from erasing your recovery path.
Practical ways to add an immutable or offline copy:
- Turn on object versioning and object lock on your cloud storage bucket.
- Give the web server write-only credentials, so it can upload backups but cannot delete them.
- Keep a monthly encrypted copy on a drive that is disconnected after each run.
- Protect the backup account with multi-factor authentication and a password stored outside the hosting account.
Full, Incremental, and Differential Backups
Full, incremental, and differential backups are the three main backup types, and they differ in what each run copies, how much storage they use, and how many steps a restore takes.
| Backup Type | What Each Run Copies | Storage Use | Restore Steps |
| Full | Every file and table, every time | Highest | One archive |
| Incremental | Only changes since the last backup of any type | Lowest | Last full backup plus every incremental since |
| Differential | All changes since the last full backup | Medium, grows until the next full | Last full backup plus the latest differential |
Many website plans combine a periodic full backup with incremental runs in between. This approach keeps storage costs low and still allows frequent restore points. The tradeoff is that an incremental chain depends on every link, so a single corrupted incremental can block restores after it.
Plugin, Control Panel, and Server-Level Backups
Plugin, control panel, and server-level backups differ in where the backup process runs and what it can capture. Each method fits a different part of the plan.
| Method | Where It Runs | What It Captures | Strength | Limitation |
| CMS backup plugin | Inside the website’s PHP process | Files and database | Easy setup from the dashboard | Can time out on large sites; restore often depends on the plugin working |
| Control panel account backup | Hosting control panel | Files, databases, email, DNS zone | Captures the full account | Requires control panel access to restore |
| Server-level snapshot | Host infrastructure, outside the site | Whole account or server | No PHP limits; fast rollback | Restore is usually requested through the host |
| Command-line scripts | Cron jobs using mysqldump, rsync, WP-CLI, or restic | Whatever you script | Full control and easy offsite upload | Requires SSH access and technical skill |
| Git repository | Version control system | Code only | Full history of every code change | Does not cover uploads or the database |
Our SMB Hosting plans include Git integration, which gives your theme and custom code a version history that sits alongside your file and database backups.
How Often Should You Back Up a Website?
Quick Answer: Back up a website as often as its data changes and as often as you can afford to lose that data. A brochure site that changes monthly needs far fewer backups than a store taking orders every hour. Two numbers, RPO and RTO, turn that judgment into a concrete schedule.
How often you should back up a website depends on two numbers you set before choosing any tool. The sections below define both numbers and turn them into schedules by site type.
Set Your RPO and RTO First
Setting RPO and RTO first means deciding how much data you can afford to lose and how long the site can stay down before you pick a backup schedule.
| Term | Definition | Question to Ask | Drives |
| Recovery Point Objective (RPO) | Maximum age of data you can afford to lose | “How many hours of changes can we redo or lose?” | Backup frequency |
| Recovery Time Objective (RTO) | Maximum time the site can stay offline | “How long can we be down before it costs real money?” | Storage location and restore method |
A worked example shows the math. A store taking 20 orders per hour with a $60 average order value has $1,200 in order data at risk for every hour of RPO. With a 24-hour RPO, a restore could erase up to 480 orders worth $28,800 from the store database, even though customers were already charged. With a 1-hour RPO, the exposure drops to 20 orders.
RTO depends on storage speed and archive size. Pulling a large archive from cold storage and importing it can take hours. Rolling back a recent snapshot on the hosting server can take minutes. Most plans use both: fast local snapshots for routine rollbacks and slower offsite copies for disasters.
Our SMB Hosting plans carry a 99.99% uptime SLA, which covers the availability of the hosting infrastructure. An uptime SLA lowers how often you face an outage, but it does not replace your own RPO and RTO. Hacks, bad updates, and accidental deletions happen on fully available servers, and only your backups undo them.
Backup Frequency by Site Type
Backup frequency by site type follows how quickly each kind of site creates data that cannot be recreated. The table below gives starting points. Adjust each row to your own RPO.
| Site Type | How Data Changes | Database Backups | File Backups | Event Trigger |
| Brochure or portfolio site | Edits monthly or less | Weekly | Weekly | Before every update or edit session |
| Blog or news site | Posts and comments several times per week | Daily | Weekly full, daily uploads | Before every update |
| Lead generation or booking site | Form entries and bookings daily | Daily or more often | Weekly | Before every update |
| WooCommerce store | Orders and inventory hourly | Hourly or continuous | Daily | Before every update or sale event |
| Membership or e-learning site | Payments and course progress continuously | Hourly | Daily | Before every update or course launch |
| SaaS application | Customer data continuously | Continuous or replicated | Per deployment through Git | Before every deployment |
Host-level backups give you a full-account restore point on the host’s schedule. When your RPO is shorter than that schedule, add your own database exports between those points. This layering keeps the host copy as your baseline and your own exports as the fine-grained layer.
Event-Triggered Backups Before Every Change
Event-triggered backups are manual snapshots taken immediately before any change that could break the site. WordPress documentation advises backing up the database regularly and always before an upgrade.
Take a snapshot before these events:
- WordPress core, plugin, or theme updates
- PHP version changes
- Migrations to a new server or domain
- Bulk product, user, or content imports
- DNS or email routing changes
- Design overhauls or page builder changes
- Major sales or launches that will generate heavy order volume
Our hosting includes One-Click Staging, so you can test core, plugin, and theme updates on a copy of your site before they reach production. A staging test plus a pre-update snapshot gives you two separate ways back from a bad update. PHP Version Control on our plans also lets you change PHP versions on your schedule, which is one more change worth a snapshot first.
How Long Should You Keep Website Backups?
Quick Answer: Keep website backups long enough to restore from a point before any problem began, not just from yesterday. Malware and data corruption often go unnoticed for weeks, so a rotation of daily, weekly, and monthly copies protects a site far better than one file that gets overwritten each night.
How long you should keep website backups depends on how quickly you would notice a problem and on any legal or contractual retention rules. A common rotation is called grandfather-father-son (GFS), which keeps short-term daily copies, medium-term weekly copies, and long-term monthly copies.
| Backup Tier | Frequency | Typical Retention | What It Covers |
| Daily (son) | Every day | 7 to 14 days | Recent mistakes and bad updates |
| Weekly (father) | Every week | 4 to 8 weeks | Problems noticed after a delay |
| Monthly (grandfather) | Every month | 6 to 12 months | Slow compromises and long lookbacks |
| Pre-change snapshot | Before each change | Until the change is verified, plus 30 days | Rollback from a specific update |
Three factors adjust these numbers:
- Detection time: If a compromise could stay hidden for 60 days, a 14-day retention window leaves you with only infected copies.
- Storage cost: Longer retention costs more, so incremental backups and exclusions keep the archive set small.
- Privacy law: Data protection laws such as the GDPR give people a right to have personal data erased. Individual records cannot easily be deleted from inside a backup archive, so document how your team re-applies deletion requests after any restore.
How to Build a Website Backup Plan in 8 Steps
Building a website backup plan takes eight steps: inventory the site, set recovery targets, pick methods, choose storage, automate, secure, test, and document. Each step below builds on the one before it.
Step 1: Inventory Every Component of the Site
Inventorying every component of the site means listing each part that would need to be rebuilt after a total loss. Include files, databases, mailboxes, DNS zones, server configuration, third-party integrations, and the credentials for each. Note where each component lives today and who has access to it.
Step 2: Define RPO and RTO for Each Component
Defining RPO and RTO for each component means assigning a maximum acceptable data loss and a maximum acceptable downtime to every item in the inventory. A store database might get a 1-hour RPO, while theme files might get a 7-day RPO. Different targets for different components keep costs reasonable.
Step 3: Choose Backup Methods for Each Component
Choosing backup methods for each component means matching a tool to each RPO. Use host-level backups for the full account, scheduled database exports for fast-changing data, Git for custom code, and mailbox exports for email. Avoid relying on a single method for everything.
Step 4: Pick Storage Destinations That Satisfy 3-2-1
Picking storage destinations that satisfy 3-2-1 means placing at least one copy outside the hosting environment under separate credentials. Common offsite choices are S3-compatible object storage in a separate account, an encrypted local drive, or a network storage device at another location. Turn on versioning and object lock where the destination supports them.
Step 5: Automate Schedules and Turn On Failure Alerts
Automating schedules and turning on failure alerts removes the human memory problem from the plan. Set each job to run on its own schedule, send success and failure notices to a monitored inbox, and review the backup log every week. A job that fails silently is worse than no job, because it creates false confidence.
Step 6: Encrypt and Lock Down Backup Access
Encrypting and locking down backup access protects the copies from theft and tampering. Use AES-256 encryption for archives at rest and TLS for transfers. Store the encryption key in a password manager, give the web server write-only access to the offsite destination, require multi-factor authentication on backup accounts, and keep archives outside the public web root.
Step 7: Test a Full Restore on a Schedule
Testing a full restore on a schedule proves the backups work before an emergency forces the test. Restore the latest backup set into a staging or local environment at least once per quarter, and monthly for stores. Verify the archive with a checksum such as SHA-256, then work through the restore checklist in the next section.
Step 8: Document the Recovery Runbook
Documenting the recovery runbook means writing the restore steps so anyone on the team can follow them under pressure. Include where each backup lives, how to access it, the restore order, expected restore times, and who to contact at the host, registrar, and payment provider. Make sure at least two people can run the restore on their own, and keep a printed or offline copy of the runbook.
How Do You Test a Website Backup Restore?
Quick Answer: Test a website backup restore by rebuilding the site from an archive in a staging or local environment, then checking pages, media, forms, logins, and checkout. Time the full process. That measured time is your real RTO, and on a first test it is often longer than the estimate.
Testing a website backup restore means proving that the archive can rebuild a working site, not just that the archive file exists. Use the checklist below during every test.
| Check | How to Verify | Pass Criteria |
| Archive integrity | Compare checksums before and after transfer | Checksums match |
| Database import | Import the dump into a fresh database | No errors; table count matches production |
| Home page and key pages | Load the top 10 pages by traffic | Pages render with correct content |
| Media | Open images, PDFs, and downloads | No broken files |
| Forms | Submit each contact and lead form | Entries save and notifications send |
| Logins | Sign in as an admin and as a customer | Both succeed |
| Checkout | Complete a test order in sandbox mode | Order records correctly |
| Scheduled tasks | Confirm cron jobs run | Tasks fire on schedule |
| Restore time | Record start and finish times | Total time is within your RTO |
Run a test restore at least quarterly for most sites, monthly for stores, and immediately after any change to backup tools or storage destinations. Record the results in the runbook. If a test fails, fix the cause and retest before the next scheduled backup cycle.
If a restore test turns up a hosting-side problem you cannot solve alone, such as a PHP version mismatch or a missing database permission, our 24/7 chat support team can help you work through it. Our PHP Version Control lets you match the restored site to the PHP version it ran on before.
Backup Planning for WordPress and WooCommerce Sites
Backup planning for WordPress and WooCommerce sites needs extra attention because these platforms change through plugins, and stores generate transaction data every hour. Patchstack’s State of WordPress Security in 2026 report counted 11,334 new vulnerabilities in the WordPress ecosystem in 2025, a 42% increase over 2024, with 91% found in plugins (Patchstack 2026 report). The same report found that 46% of vulnerabilities had no patch available at the time of public disclosure. When a fix does not exist yet, a clean backup is often the fastest way back after an exploit.
Our hosting plans include managed WordPress updates, WordPress security updates, and a Smart WordPress Plugin Manager to help reduce how often you need a restore. Free malware scanning, a web application firewall, and DDoS protection add more layers in front of the site. If WordPress runs your business, start with our managed WordPress hosting plans.
WordPress Backup Plugin Pitfalls
WordPress backup plugin pitfalls come from the fact that plugins run inside the same PHP process and file system they are trying to protect. Watch for these issues:
- Timeouts: PHP execution limits can stop a large backup partway through and leave a truncated archive that looks complete.
- Default storage location: Many plugins save archives inside wp-content, which keeps backups on the same server and sometimes inside the public web root.
- Restore dependency: Restoring often requires installing the same plugin first, which fails if WordPress itself will not load.
- Attack surface: A backup plugin is still a plugin, so it needs updates like any other plugin.
Use a plugin as one layer, send its archives to an offsite destination, and pair it with host-level or command-line backups.
Backing Up WooCommerce Without Losing Orders
Backing up WooCommerce without losing orders requires frequent database backups and a plan for reconciling transactions after a restore. Rolling a store database back 12 hours erases every order placed in those 12 hours, even though the payment gateway already captured the payments.
Follow these rules for store restores:
- Export recent orders as CSV before restoring the database.
- Put the store into maintenance mode during the restore.
- Restore files alone for code problems and the database alone for data problems, instead of rolling back both by default.
- Reconcile orders after the restore using the payment gateway dashboard and order confirmation emails.
- Schedule restore tests during low-traffic hours.
If you are launching or moving a store, our WooCommerce hosting gives you a WordPress-ready starting point, with One-Click Staging for testing updates before they reach checkout. Free CDN and DDoS protection on our plans help the store handle sale-day traffic, which is exactly when you want to avoid an emergency restore.
Exclude Cache, Logs, and Generated Files
Excluding cache, logs, and generated files makes backups smaller, faster, and cheaper to store without reducing recovery value. Skip cache directories, log files, generated image thumbnails that the CMS can rebuild, and old backup archives stored inside the site folder. Smaller archives also restore faster, which shortens your real RTO.
Website Backup Mistakes to Avoid
Website backup mistakes to avoid include keeping every copy on one server, not knowing your host’s backup schedule, backing up only the database, and never restoring a test copy. These mistakes come up repeatedly in site owner and developer discussions on forums such as Reddit and Quora.
- Not knowing your host’s schedule. Check how often host backups run, how many versions are kept, and how restores are requested. Fill any gap between that schedule and your RPO with your own copies.
- Keeping backups only on the live server. A server failure or account compromise removes both the site and the backups.
- Backing up the database but not the uploads folder. The restored site loads with every image missing.
- Storing archives in the public web root. Predictable filenames let anyone download a full database copy, including password hashes.
- Never testing a restore. A corrupted or incomplete archive stays hidden until the day you need it.
- Using one login for everything. If the hosting account and the backup account share credentials, one breach exposes both.
- Ignoring failed job alerts. Failure notices sent to an unmonitored inbox give no protection.
- Keeping only one version. Overwriting the same archive each night means an infected or corrupted site replaces the only clean copy.
- Forgetting email and DNS. A restored website with missing MX and TXT records cannot send or receive mail reliably.
- Storing the encryption key inside the backup. An encrypted archive whose key was lost with the server cannot be opened.
Moving hosts is a common moment for these mistakes. If you move to Prestige Technologies, our free site migration moves your existing site for you. Take a full backup before the move, and keep that backup as an offsite copy until the migrated site passes every check in your restore test.
Sample Website Backup Policy Template
A sample website backup policy template lists each component, its backup method, its schedule, its storage locations, its retention period, its owner, and its test cadence in one table. The example below fits a small business WordPress site with an online store.
| Component | Method | Schedule | Storage Locations | Retention | Owner | Test Cadence |
| Store database | Scheduled database export | Hourly | Host server plus offsite object storage | 14 daily, 8 weekly, 12 monthly | Site administrator | Monthly |
| Website files | Host-level backup plus offsite file sync | Daily file sync, host backups on host schedule | Host backup storage plus offsite object storage | 4 weekly, 12 monthly | Site administrator | Quarterly |
| Custom theme and code | Git repository | Every change | Git remote plus local clone | Full history | Developer | Every deployment |
| Business email | Mailbox export | Weekly | Encrypted local drive | 12 monthly | Office manager | Quarterly |
| DNS zone | Zone file export | After every DNS change | Password manager attachment plus offsite storage | All versions | Site administrator | Annually |
| Credentials and keys | Password manager | When rotated | Password manager with emergency access | Current plus previous | Business owner | Annually |
Copy this table into your own documentation and replace each value with your actual tools, schedules, and owners. Review the policy every 6 months and after any change to hosting, storage, or team roles.
On our SMB Hosting plans, the “host backups on host schedule” entry in this table is our automated weekly backup. Your hourly database exports and offsite sync sit on top of that baseline, and Git integration on our plans covers the “Custom theme and code” row.
How Prestige Technologies Supports Your Backup Plan
Prestige Technologies supports your backup plan with hosting features that cover the baseline copy and reduce how often a restore is needed. We handle the infrastructure layer of the plan, so your team can focus on the copies only you can own. Here is how our plans map to the steps in this guide:
- Automated weekly backups: Your baseline host-level copy in the 3-2-1 rule.
- One-Click Staging: Test updates safely before they go live.
- Git integration: Version history for themes and custom code.
- Free malware scanning, web application firewall, and DDoS protection: Security layers that reduce restore events.
- Managed WordPress updates and WordPress security updates: Updates handled on your behalf.
- 99.99% uptime SLA and global data centers: Listed on every SMB Hosting plan.
- Free site migration: Bring an existing site, then set up your offsite copy from day one.
- 24/7 chat support: Help from our team when you have questions.
Compare our small business hosting plans to see what each tier includes.
Build a Backup Plan That Restores
A backup plan that restores is one with complete scope, frequency set by RPO, at least one offsite copy under separate credentials, versioned retention, and a tested runbook. Start with the inventory in Step 1 this week. Then set your RPO, add an offsite destination, and schedule your first test restore.
The goal is not to own backups. The goal is to recover in a known amount of time with a known amount of data lost. A written website backup strategy gives you both numbers before an incident instead of after it. If you want a hosting foundation that handles the baseline copy, staging, and security layers while you own the offsite copy, explore our web hosting plans or talk to our team through 24/7 chat.
Frequently Asked Questions
What is the 3-2-1 backup rule for websites?
The 3-2-1 backup rule for websites means keeping three copies of your site data, on two different storage systems, with one copy stored offsite. For most sites, that is the live site, a host-level backup, and an offsite copy in a separate cloud or local account. The offsite copy should use credentials your web server does not hold.
How often should I back up my website?
Back up your website as often as its data changes and as often as you can afford to lose that data. Static brochure sites can run weekly backups. Blogs usually need daily database backups. Online stores and membership sites need hourly or continuous database backups. Always take an extra snapshot before updates, migrations, or bulk imports.
Is my web host’s backup enough on its own?
A web host’s backup is a strong baseline copy, but it should not be your only copy. Host backups usually sit inside the same account as the live site, so a lockout, billing issue, or compromised login can block both. Keep at least one additional copy in a separate account you control to satisfy the 3-2-1 rule.
What is the best way to back up a website?
The best way to back up a website is to combine methods instead of relying on one tool. Use host-level backups for the full account, scheduled database exports for fast-changing data, version control for custom code, and an offsite copy under separate credentials. Automate every job, encrypt the archives, and test a full restore every quarter.
What files do I need to back up for a WordPress site?
A WordPress backup needs the full database plus the site files. The files include WordPress core, the wp-content folder with themes, plugins, and uploads, and the wp-config.php file. The database holds posts, pages, settings, users, and orders. Export the database first, then copy the files, so both reflect the same point in time.
Should I back up my website to the cloud or a local drive?
Use both when possible, because each covers a different risk. Cloud storage in a separate account provides an offsite copy that survives server failures and account lockouts. An encrypted local drive that stays disconnected between runs provides an offline copy that ransomware cannot reach. Together they satisfy the two-media and offsite parts of the 3-2-1 rule.
How long should I keep website backups?
Keep website backups long enough to restore from a point before any problem started. A common rotation keeps 7 to 14 daily copies, 4 to 8 weekly copies, and 6 to 12 monthly copies. Longer retention helps when malware or data corruption goes unnoticed for weeks. Adjust retention for storage cost and any legal retention requirements.
What is the difference between RPO and RTO?
RPO, or recovery point objective, is the maximum amount of recent data you can afford to lose, measured in time. RTO, or recovery time objective, is the maximum time your site can stay offline during a restore. RPO sets how often backups run. RTO sets where backups are stored and how quickly they must restore.
What is the difference between incremental and differential backups?
An incremental backup copies only data that changed since the last backup of any type. A differential backup copies all data that changed since the last full backup. Incremental backups use less storage but need every link in the chain to restore. Differential backups use more storage but restore with only the full backup and the latest differential.
How do I know if my website backup actually works?
You know a website backup works only after restoring it into a staging or local environment. Verify the archive checksum, import the database, then check key pages, images, forms, logins, and checkout. Time the full process to confirm it fits your recovery time objective. Repeat the test every quarter and after changing backup tools.
Can ransomware infect website backups?
Yes, ransomware and other attackers often search for connected backups and then encrypt or delete them. Any backup reachable with the same credentials as the live site is at risk. Protect at least one copy by keeping it offline or immutable, storing it under separate credentials with multi-factor authentication, and giving the web server write-only access to it.
What is an immutable backup?
An immutable backup is a copy that cannot be changed or deleted until a set retention period ends. Cloud object storage provides immutability through versioning and object lock settings. Immutable copies protect against attackers and mistakes, because even a user with valid credentials cannot erase them early. They form the extra protected copy in the 3-2-1-1-0 rule.
Should website backups be encrypted?
Yes, website backups should be encrypted, because they often contain customer data, password hashes, and API keys. Use AES-256 encryption for archives at rest and TLS for transfers to storage. Keep the encryption key in a password manager outside the backup itself. A lost key makes the backup impossible to restore, so give two people secure access.
Do I need to back up my website before updating plugins?
Yes, take a backup before every plugin, theme, or core update. Updates can conflict with other plugins, change database tables, or break layouts. A snapshot taken minutes before the update lets you roll back quickly if something fails. Testing the update on a staging copy first adds a second layer of protection for the live site.
How do I back up a WooCommerce store without losing orders?
Back up the store database hourly or continuously, because orders change faster than any other data. Before any restore, export recent orders and put the store into maintenance mode. After restoring, reconcile orders against the payment gateway dashboard and confirmation emails. Restore only files for code problems, so order data stays untouched whenever possible.
Are backup plugins safe to use on WordPress?
Backup plugins are safe when kept updated and configured correctly, but they should be one layer, not the whole plan. Plugins can time out on large sites, often store archives inside the web root by default, and may need to be reinstalled before a restore. Send plugin archives offsite and pair them with server-level backups.
Should I back up business email with my website?
Back up business email with your website when mailboxes are hosted on the same account. Include message folders, filters, forwarders, and the mailbox list. Mail accessed through IMAP keeps its primary copy on the server, so schedule exports or server-side backups. Also export your DNS zone, because MX and TXT records control mail delivery and authentication.
How much storage do website backups need?
Website backup storage depends on site size, backup type, and retention. Full backups multiply site size by the number of versions kept, so a 5 GB site with 30 full versions needs about 150 GB. Incremental backups store only changes, which often cuts that total sharply. Excluding cache folders and logs reduces storage further.
What is a website disaster recovery plan?
A website disaster recovery plan is a written procedure for restoring a site after a major failure, such as a hack, server loss, or account lockout. It lists where backups live, the restore order, recovery time and data loss targets, team responsibilities, and vendor contacts. Backups are one part of the plan, not the whole plan.
Can I restore a website backup to a different host?
Yes, a complete backup can usually be restored to a different host if it includes the files, the database, and the configuration. Import the database, upload the files, update the database connection settings, and then point DNS to the new server. Backups that depend on one host’s proprietary format may need converting before they can restore elsewhere.