Preventive IT Maintenance Checklist: What to Do Monthly, Quarterly and Yearly

The short version: Preventive IT maintenance means doing small, scheduled checks so faults are found before users report them. Split the work by frequency (weekly, monthly, quarterly, yearly) and by area (servers, network, endpoints, backup, accounts), give every task an owner, and record the result. The thresholds below are starting points to tune for your environment.

This is for the person who looks after IT in a small or mid-sized organisation, or who needs to check that an outsourced provider is doing the routine work. You will get a checklist you can copy into a shared document or ticketing tool, with tasks grouped so each visit or each month has a clear list.

Use it as a template. Delete tasks that do not apply, add your own, and keep a dated record of each run.

How to use this checklist

1. Make an inventory first. List servers, network devices, endpoints, cloud services and backup jobs. Each task below applies to items on that list.
2. Give every task an owner and a due date. "IT" is not an owner.
3. Record the result, not just "done". Write the number you saw (disk space left, backup age, firmware version) so you can compare next month.
4. Turn repeated checks into alerts. If you check the same thing every week and it is always fine, monitor it and only look when it alerts. See what to monitor and which alert thresholds to start with.
5. Schedule changes in a window. Anything that restarts a service or a device goes in an agreed maintenance window with a way back (see the patch management process).

Weekly (about 30 minutes, mostly review)

1. Backup: read the job report for every backup. Any failed or skipped job becomes a ticket the same day.
2. Alerts: review open monitoring alerts and close, fix or tune each one.
3. Storage: check disk space on servers and shared storage. Starting point: investigate any volume with less than 20% space left or less than 50 GB left, whichever comes first (on very large volumes a percentage alone alerts too late).
4. Security console: review new alerts and any devices reported as unprotected or out of date.
5. Accounts: review failed sign-in spikes, locked-out users and new admin role assignments.
6. Tickets: look for the same fault reported by three or more users. That is a problem to fix, not three tickets to close.

Monthly

Servers

1. Apply the month's operating system and application updates through your patch cycle, and check each server after its restart.
2. Review the system and application event logs for repeated errors and warnings.
3. Check disk health, RAID or storage-pool status, and hardware alerts (fans, power supplies, temperatures).
4. Check CPU and memory trends against last month. Rising trends need a plan before they become outages.
5. Confirm time sync, DNS and DHCP services are healthy and DHCP scopes have spare addresses.
6. Remove stale accounts, scheduled tasks and test services.

Network

1. Check device status, uptime and link errors on switches, firewalls and access points.
2. Confirm the internet link and any backup link work, and note latency and loss against your baseline.
3. Review firewall logs for blocked patterns and for rules that are never hit.
4. Check wireless coverage complaints and busy access points.
5. Confirm configuration backups exist for firewalls, switches and routers, and that the latest one is recent.
6. Check support and licence expiry dates for network hardware and security subscriptions.

Endpoints

1. Check patch status and pending restarts across laptops and desktops.
2. Review devices that have not checked in to management for 14 days or more (starting point).
3. Confirm disk encryption and security protection are on for every device.
4. Check batteries and obvious hardware faults reported by users.
5. Update the asset list for joiners, leavers and moved devices.

Backup

1. Restore one file or mailbox item from backup to prove the data is readable.
2. Check retention matches what you agreed, and that offsite or cloud copies are up to date.
3. Check backup storage capacity and growth.

Accounts and security

1. Disable accounts of leavers and review accounts with no recent sign-in.
2. Review who holds administrator roles and remove those who no longer need them.
3. Check certificate and domain expiry dates (30 days notice is a starting point).
4. Review mail protection reports for rising spam or phishing trends.

Quarterly

1. Restore test: restore a full server or critical system to a test location, time the restore, and compare it with the recovery targets you set. For the terms, see RTO vs RPO.
2. Firmware and drivers: plan updates for switches, firewalls, access points, storage and server firmware in a maintenance window.
3. Firewall rule review: remove unused, duplicate or overly broad rules, and confirm every rule has a named owner.
4. Capacity review: compare storage, licence counts, bandwidth and server load with the last three months and flag what will run out within six months.
5. Access review: ask managers to confirm who should have access to shared folders and key applications.
6. Physical check: inspect the comms room or cabinet for dust, heat, cable damage, loose labels and UPS battery status.
7. Documentation: update the network diagram, IP list and contact list for anything that changed.
8. Metrics review: look at ticket trends, recurring faults, patch coverage and backup success rate, and pick the top two things to fix next quarter.

Yearly

1. Disaster recovery exercise: run a scenario (loss of the main office, loss of a key server, ransomware) and write down what took longer than planned.
2. Hardware lifecycle review: list devices past their planned age or out of vendor support, and budget their replacement. Plan secure data wipe and disposal for retired items; see IT asset disposal.
3. Contracts and renewals: review support contracts, licences, domain names, certificates and internet lines.
4. Security assessment: check your configuration against a recognised baseline, or arrange an independent review; see the IT security assessment.
5. Policy review: update acceptable use, password, remote access and joiner/leaver procedures.
6. User training: run a refresher on phishing and safe handling of data.
7. Provider and SLA review: compare what you receive against what was agreed; our guide to what an SLA means in IT support lists the points to check.

A quick health check you can run on a Windows server

This script lists physical disk health, the space left on each volume and critical or error events from the last 24 hours. Run it in PowerShell as administrator. The cmdlets are documented on Microsoft Learn (checked 28 October 2026).

Get-PhysicalDisk | Select-Object FriendlyName, HealthStatus, OperationalStatus

Get-Volume | Where-Object DriveLetter |
  Select-Object DriveLetter,
    @{n='SizeGB';e={[math]::Round($_.Size/1GB,1)}},
    @{n='PercentLeft';e={[math]::Round(100*$_.SizeRemaining/$_.Size,1)}}

$since = (Get-Date).AddHours(-24)
Get-WinEvent -FilterHashtable @{LogName='System','Application'; Level=1,2; StartTime=$since} -MaxEvents 50 |
  Select-Object TimeCreated, ProviderName, Id, LevelDisplayName

Common mistakes

1. Backups that are never restored. A green job report does not prove the data is usable. Restore something every month.
2. No baseline. Without last month's numbers you cannot see a trend. Record values, not ticks.
3. Firmware ignored. Network and storage firmware gets skipped because it feels risky. Plan it quarterly, with a rollback.
4. One person holds it all in their head. Write the checklist and the documentation so someone else can run it.
5. Skipping the physical check. Heat, dust and failing UPS batteries cause outages that no software alert predicts early.
6. Doing everything at once. Spread the work across the weeks so it actually happens.

How to check it worked

1. Every task on this list has an owner and a dated record for the last cycle.
2. The count of unplanned outages and repeat tickets is lower than last quarter. If it is not, change the checklist.
3. The latest restore test met the recovery targets you set.
4. No device in your inventory is missing from monitoring, patching or backup.
5. Expiry dates for certificates, domains and support contracts are all more than 30 days away, or have a renewal ticket.

FAQ

What are examples of preventive IT maintenance measures?

Patching on a schedule, reviewing backup reports and testing restores, checking disk and hardware health, updating firmware, reviewing accounts and admin roles, and checking certificate and contract expiry dates.

What is the difference between preventative and preventive?

In everyday use they mean the same thing: done in advance to stop a problem. "Preventive" is the shorter form and the one we use here.

What is covered under preventive IT maintenance?

Servers, network equipment, endpoints, backups and the accounts and security settings around them. The checklist above groups the tasks by area and by how often to do them. Your own scope list should name exactly which systems are in and out.

How often should preventive maintenance be done?

Use a mix: weekly reviews, a monthly set of checks, a quarterly deeper set (restore tests, firmware, rule reviews), and a yearly review. Systems that carry more risk or change more often deserve a shorter cycle.

Next step

For help with the network and hardware side of this list, see our network and hardware maintenance service.

Sources checked 28 October 2026: Microsoft Learn (Get-PhysicalDisk, Get-Volume, Get-WinEvent). Thresholds are starting points to tune.

Brocent publishes its managed IT support plans with per-user monthly prices for Hong Kong, Singapore, mainland China and the United States, so you can compare before you talk to anyone: see the plans and pricing.