When a computer is failing, the two biggest risks are (1) losing data and (2) spending money on the wrong fix. This workflow is designed to reduce both. It is written for Christchurch SMEs, but the same sequence works anywhere.
For background context on who I am and the kind of work these notes come from, see About John Finnerty, the Projects hub, and the concise profile at johnfinnerty.nz.
1. Decide what "success" means before you touch anything
Repairs can be done fast, but only if everyone agrees what "done" looks like. Write this down in one sentence.
- Do you need the machine usable today, or can it be out of action for a day or two?
- Is the priority data recovery, stability, speed, or secure re-setup?
- What apps must work at the end (accounting, email, POS, VPN, printer/scanner)?
- Is this a business-critical machine, or could a spare laptop temporarily take its place?
2. Protect the data first (even if you think it's "just slow")
If a drive is failing, repeated restarts and updates can make recovery harder. Backups turn a scary incident into a routine fix.
If you need help setting up a restore-first backup, migrating data to a new machine, or recovering access after a repair, the Data backup and migration (Christchurch) page explains what a safe handover looks like.
What to back up
Documents, photos, accounting data, customer files, browser passwords, and email archives.
Where the data lives
Confirm whether files are local, OneDrive/Google Drive, an external drive, or a NAS.
Account access
Write down the login email and recovery method so cloud sync can be reconnected after repair.
Encryption note
If BitLocker/FileVault is enabled, confirm recovery keys exist before any reinstall plan.
Restore-first gate: prove the recovery path before repair
A repair workflow should have a stop point before any risky operation. If the data cannot be restored, the next step is recovery planning, not updates, reinstall attempts, or repeated boot cycles.
Backup exists
Confirm where the backup is stored, who can access it, when it last ran, and whether it includes the folders and app data that matter.
Restore works
Open a small sample from the backup on another device or profile. A backup that cannot be opened is not yet a recovery path.
Encryption is understood
Check BitLocker or FileVault before storage, boot, firmware, or reinstall work. Missing keys can turn a repair into a reset.
Old device preserved
Keep the old device or drive untouched until the repair, replacement, or migration has been tested in normal use.
3. Isolate the fault domain (hardware, Windows, Wi-Fi, or upstream internet?)
A lot of "computer problems" are actually Wi-Fi, router, DNS, or Microsoft 365 sign-in issues. The fastest workflow is to separate layers.
Hardware symptoms
Overheating, fan screaming, random shutdowns, clicking drive noise, swollen battery, or storage nearly full.
Windows symptoms
Blue screens, repeated boot loops, updates failing, corrupted profiles, or constant crash reports.
Network/Wi-Fi symptoms
Works on Ethernet but not Wi-Fi, dead zones in specific rooms, or drops when the microwave is on.
Account/access symptoms
Cannot sign in, MFA issues, mail not syncing, or Microsoft 365 apps asking to "reconnect" repeatedly.
If the symptom is "internet is slow", start with Wi-Fi and slow internet troubleshooting and Wi-Fi dead zones and coverage fixes before you pay for a new laptop.
4. Choose repair vs replace with a simple decision rule
Replacing the wrong thing is expensive. A clear rule makes decisions easier when you are stressed.
- Repair-first when the device is less than a few years old, the data is safe, and the fault is likely a clean fix (SSD upgrade, battery, OS repair).
- Replace-first when the device is unreliable, parts are scarce, the business cannot tolerate repeated downtime, or the effort to stabilise it exceeds the cost of replacement.
- Recovery-first when data is at risk or storage is failing. Stabilise and recover before attempting updates or reinstall.
Repair-vs-replace evidence table
The decision should be based on evidence, not sunk cost. Write down which signal is driving the next move before money is spent.
Repair-first signals
Recent device, clear single fault, available parts, safe backup, known software state, and a repair cost that is clearly below replacement/migration cost.
Replace-first signals
Unsupported OS, unreliable battery or charging, repeated storage faults, scarce parts, weak performance for current work, or repeated downtime.
Recovery-first signals
Only copy of important files, clicking drive, failed boot loops, encryption uncertainty, corrupt files, or no recent restore-tested backup.
Warranty or rights check
If the device or repair may be covered by warranty or consumer guarantees, preserve evidence and check the appropriate supplier path before opening or modifying it.
5. Write handover notes that prevent "repair ping-pong"
Good notes stop the problem bouncing between a repair shop, an ISP, and "someone good with computers" while time and money disappear.
Use the Computer repair triage checklist (Christchurch) as a handover template. At minimum, capture:
- What changed, when it started, and any exact error messages (screenshots/photos help).
- What has already been tried (restart, updates, safe mode, antivirus scans, router reboot).
- Backup status: what is backed up, where it is stored, and whether the backup was tested.
- Access details: is it a local account or Microsoft account sign-in, and is MFA/recovery available?
- Any business constraints: after-hours work, must-keep apps, or deadlines.
Official reference points for the workflow
These are reference points for the parts of the workflow where mistakes can create avoidable downtime: backups, encryption, Windows recovery, and repair/replacement expectations.
Backup testing
NCSC's backup guidance supports the restore-first habit: backups should be protected, monitored, and tested before a crisis.
BitLocker keys
Microsoft explains recovery-key locations and why key access matters before recovery or reinstall work.
Startup recovery
Microsoft's Windows Startup Settings page covers Safe Mode and related recovery options for startup problems.
NZ consumer context
Consumer Protection explains repair, replacement, and refund context under the Consumer Guarantees Act. This is background, not legal advice.
6. Escalate to the right kind of help
Not every problem is "computer repair". The best next step depends on the fault domain you isolated above.
Computer repair Christchurch
When the machine is unstable, storage is failing, or the OS needs repair/rebuild work.
IT support Christchurch
When the bigger issue is access, identity, Microsoft 365, printers, email, and day-to-day reliability.
Wi-Fi support Christchurch
When the pain is coverage, dead zones, roaming, or instability that looks like a "computer issue".
Remote IT support NZ
When you are not in Christchurch, or you want quick diagnosis before scheduling onsite work.
A simple takeaway
Most "computer repair" incidents are solvable faster if you (1) back up first, (2) isolate the layer that is failing, and (3) keep notes. That is what keeps the fix focused and prevents rework.