Here’s the honest boundary: if the machine turns on and can reach the internet, a remote technician can fix nearly anything wrong with the software on it. If it doesn’t turn on, or the screen is cracked, or the drive has physically died, no amount of remote access helps — there’s nothing to connect to, and nobody can replace a part through a network cable.

That line is worth understanding before you call anyone, because it tells you whether you’re looking at a 40-minute session this afternoon or a trip to a shop with a two-day turnaround. Most small business computer problems fall on the fixable-remotely side. People are often surprised by how far that side extends.

Briefly, on mechanics: you get a link, you run a small program, the technician connects, you watch the entire session on your own screen, and you can end it whenever you like. We cover the step-by-step walkthrough in more depth separately — this article is about scope.

What can and can’t be handled over a remote connection

The pattern is simple once you see it. Everything on the left is information — files, settings, registry keys, configuration. Everything on the right is atoms.

Why the “can’t” list is genuinely can’t

Two distinct reasons, and it’s worth separating them.

No connection exists. A remote session runs over software on a booted operating system. If the machine won’t power on, won’t POST, or can’t reach the network, there is no channel. This is why “my computer is completely dead” is the one problem remote support definitively cannot touch — not because of skill, but because there’s nothing to talk to.

The problem is physical. Software can’t reseat a RAM stick, can’t clean dust out of a heatsink, and can’t swap a swollen battery. A technician can identify all three from thousands of miles away and tell you precisely what needs doing — but the doing requires a screwdriver and a person.

Beware anyone who claims otherwise. If a service tells you they can fix a cracked screen or a dead power supply through a remote connection, you’ve found a scam. That claim is technically impossible and it’s a reliable tell.

The grey area: diagnosable remotely, fixable locally

A large middle category gets misjudged in both directions. These are hardware problems where the diagnosis — the hard, expensive part — happens remotely, and only the final physical step needs someone present.

A failing drive. If a machine still boots, SMART data tells the whole story over a remote session. Reallocated sector counts, pending sectors, wear leveling on an SSD, power-on hours — all readable remotely, along with the Windows event log entries for disk errors. The outcome is a specific answer: your drive is failing, here is the exact model number and capacity to order, and here’s what a replacement costs. Then either you swap it yourself with guidance, or a local shop does a fifteen-minute part swap instead of a full diagnostic they’d bill an hour for.

Critically, that diagnosis usually arrives while the drive still works — which means the data can be backed up before it doesn’t.

Overheating and thermal throttling. A machine that’s slow under load and fine at idle shows a clear pattern in temperature and clock speed monitoring. Remote diagnosis identifies dust or thermal paste as the cause. The cleaning is physical, but you now know it needs cleaning rather than replacing.

Memory faults. Random blue screens with varying stop codes usually point at RAM. Windows Memory Diagnostic and minidump analysis can be run and reviewed remotely; if a stick is bad, you know which slot and what to buy.

Battery health. On Windows, generating a battery report remotely shows design capacity versus current full-charge capacity. A battery down to 60% of original is a replacement decision, made with real numbers rather than a guess.

The practical benefit here is avoiding the worst outcome in computer repair: paying a shop to spend two hours discovering something a remote session could have told you in twenty minutes, then paying again for the actual work.

Is it safe to let someone connect to your computer?

Reasonable question, and the reason it’s reasonable is that tech support scams use exactly this mechanism. The FBI’s Internet Crime Complaint Center logs tens of thousands of tech support fraud complaints a year, with losses in the hundreds of millions. The tool is the same one legitimate technicians use.

So the tool isn’t the signal. The circumstances are.

The single most reliable test: who started it?

Legitimate remote support is something you initiate. You noticed a problem, you called a company you chose, you agreed to a session.

Scams start the other way. A phone call out of nowhere claiming to be Microsoft. A full-screen browser warning with a siren sound and a number to call. A popup saying your computer is infected and support is standing by.

Microsoft does not call you. Neither does Apple, Google, or your bank’s IT department. No legitimate company monitors your personal computer for viruses and phones you about it. If contact came to you unsolicited, end it — regardless of how convincing the caller ID or the warning screen looks.

The gift card one deserves emphasis because it’s so consistent. No legitimate business in the United States accepts Apple gift cards, Google Play cards, or Target cards as payment for IT services. Not one. The instant that request appears, the conversation is fraudulent and there’s no ambiguity about it.

What a legitimate provider looks like

• A real business name, a US address, and a phone number that’s published on their own website
• Clear pricing stated before any work begins — not “we’ll see what we find”
• Payment by card or invoice, through a normal payment processor
• Willingness to explain what they’re doing and why, as they do it
• No objection to you watching, asking questions, or stopping the session

Your controls during the session

You keep three things throughout, and they matter more than any promise a provider makes.

You see everything. The screen the technician is working on is your screen. Every click, every window, every command is visible to you in real time. Stay at the desk. If something looks wrong — a banking site opening, a file being copied somewhere unfamiliar — ask about it immediately.

You can end it instantly. Closing the remote support application terminates the connection. You don’t need permission and you don’t need to explain. Pulling the network cable or turning off Wi-Fi works too. There’s no negotiation involved.

Access stops when the session stops. Attended support tools generate a one-time session code that expires. Once the session ends, that code is dead and nobody can reconnect with it. If a provider wants persistent access for ongoing monitoring, that’s a separate agreement you should explicitly agree to and be able to revoke.

Sensible precaution: before a session, close anything you’d rather not have on screen — online banking, payroll, HR files. Not because you distrust the technician, but because it costs you nothing.

Remote, on-site, or drop-off?

The economics favor remote for a specific reason: no travel time. An on-site provider has to charge for the drive whether they bill it as a callout fee or bake it into an hourly minimum. A remote technician handling a printer driver issue spends twenty-five minutes on the actual problem and zero minutes in a car.

How long things typically take

Rough figures from common small business requests:

• Printer not printing, driver reinstall — 15 to 30 minutes
• Outlook won’t send or receive — 20 to 45 minutes
• Malware removal on a bootable machine — 45 to 90 minutes
• General slowness cleanup — 30 to 60 minutes
• Windows Update failure — 30 minutes to 2 hours, depending on whether DISM has to run
• New user setup on a fresh laptop — 1 to 2 hours
• Hardware diagnosis with a parts recommendation — 20 to 40 minutes

Some of that runs unattended. A DISM repair or a full malware scan doesn’t need you watching — you go back to work and get a message when it’s done.

When a small business should default to remote

For an office of five to fifty people, most of what breaks is software, and software problems recur. The same printer stops working, the same person forgets the same password, the same laptop needs the same update forced through.

Handling those remotely means help arrives in minutes rather than days, and it means the fix isn’t gated on somebody’s route for the afternoon. If your team is spread across locations or partly remote, it’s the only model that covers everyone equally — a laptop in someone’s home office is exactly as reachable as one in the building.

Keep a local repair shop’s number for the physical work. Hardware failure is the minority of what goes wrong, but it does happen, and the two approaches complement each other rather than compete.

Common questions

Can a remote technician fix my computer if it won’t turn on?

No, and anyone claiming they can is not being straight with you. Remote support needs the machine booted and connected to the internet. A computer that won’t power on has no way to establish a connection. What can be done remotely is helping you narrow down the cause over the phone — power supply, battery, display, or motherboard — so you know what to tell the shop.

Can they see my files and passwords?

During an active session a technician sees whatever is on your screen and can browse the file system, the same as if they were sitting at your desk. That’s what makes the work possible. You see everything they do in real time, and you can end the session instantly. Passwords saved in your browser are not readable without your explicit action, and a legitimate technician will ask you to type any password yourself rather than requesting it.

Do I have to stay at the computer the whole time?

You don’t have to, but the first time you work with a provider it’s a good idea. Once you know them, most people step away during long scans or updates. Do stay present for anything involving credentials, financial software, or files you’d consider sensitive.

What happens if the fix doesn’t work?

Ask about this before you engage anyone. A reasonable provider either doesn’t charge for a problem they can’t resolve, or tells you clearly at the outset what happens if the diagnosis turns out to be hardware. Flat-rate monthly arrangements sidestep this entirely — you’re not paying per incident, so an inconclusive session doesn’t cost extra.

Is remote support secure enough for a business handling client data?

Yes, when the provider is legitimate. Reputable remote tools use end-to-end encryption, one-time session codes, and full session logging. The genuine risks aren’t technical — they’re granting access to the wrong party, or leaving standing access enabled with no oversight. If you handle regulated data, ask for a business associate agreement or equivalent in writing and confirm sessions are logged.