Start with what you actually need
Before you ring anyone, be honest about the scale of the problem. A five-person studio that mostly needs someone reachable when the internet drops has very different needs from a thirty-seat office running a line-of-business application. Write down what breaks, how often, and what it costs you per hour when it does. That single page tells a good provider more than any brief, and it stops you buying a heavyweight managed contract when you needed a reliable phone number, or the reverse.
Be clear too about where the work happens. Some issues genuinely need someone on site — cabling, a dead machine, a new office fit-out. Plenty of others are solved faster over a screen-share. The best arrangements blend both, so ask how a prospective provider decides which is which rather than assuming everything means a call-out.
The questions that separate good from average
Anyone can say they are “responsive” and “proactive”. The point of these questions is to make them prove it:
- Who actually answers when I call? A named engineer or a triage queue in another time zone? Ask what happens at 9am on a Monday when three clients have problems at once.
- What are your response times, in writing? Not marketing copy — the figure in the agreement. “We aim to respond quickly” is not a commitment.
- Do I speak to the same people each time? Continuity matters. An engineer who knows your setup fixes things faster than one reading your history cold.
- How do you handle security and backups? If they can’t explain their approach to patching, multi-factor authentication and tested restores in plain English, that tells you something.
- What happens when I leave? Ask before you sign. Who owns the passwords, the domain, the documentation? A confident provider answers this cheerfully.
Ask for two or three references from businesses of similar size, then actually call them. Ask those clients the dull, telling question: when something went wrong, what did they do?
Understanding response times honestly
Response time and resolution time are not the same thing, and providers sometimes blur the two. A one-hour response target means someone acknowledges your ticket within an hour — not that your problem is fixed. That is still valuable, because knowing a competent person is on it changes how you plan your morning. But read the agreement for how issues are prioritised. A server outage and a jammed printer should not sit in the same queue.
Be realistic about geography. A provider promising a rapid on-site arrival across a wide area is either exaggerating or very local to you. For quick physical fixes, proximity is a genuine advantage — a workshop a few miles away can often get hands on your hardware far faster than one two counties over.
Local versus national
National providers offer scale, out-of-hours desks and standardised processes, which suits multi-site organisations. The trade-off is that you become a ticket number, and on-site visits can mean a subcontractor you have never met.
A local provider tends to offer continuity, faster physical response, and someone who will remember that your accounts machine is the temperamental one in the corner. The trade-off is depth of bench — a small team can be stretched. There is no universally right answer. What matters is matching the model to how you work: if you rarely need anyone in the room, strong remote IT support may serve you better than a nearby office. If you value a known face and prompt hands-on help, local onsite IT support is worth paying for.
Contracts versus pay-as-you-go
Both models are legitimate; the trick is matching them to your usage.
- Pay-as-you-go suits businesses with occasional, unpredictable needs. You pay only when you call, with no monthly commitment. The downside is that nobody is watching your systems between incidents, so problems tend to be found after they bite.
- Monthly contracts (often called managed support) bundle monitoring, patching, backups and a set amount of help for a predictable fee. They suit businesses that depend on their systems daily and want issues caught early. The risk is paying for a package richer than you use — so scrutinise what is genuinely included versus billed as “extra”.
Whichever you choose, insist on transparent pricing. Ask exactly what an out-of-hours call, a project, or an additional user costs, and get it in writing.
Red flags worth walking away from
- Vague or evasive pricing. If you can’t get a straight answer on costs before signing, it won’t improve after.
- Lock-in on your own systems. A provider who keeps sole control of your passwords, domain or backups is protecting themselves, not you.
- No mention of security. In 2026, a support provider that treats cybersecurity as an afterthought is a liability. Backups, patching and MFA should come up unprompted.
- Over-promising. Anyone guaranteeing “everything fixed same day, every time” is selling, not supporting. Real work has exceptions.
- No documentation. If they can’t show how they record your setup, you are one departed engineer away from starting over.
A sensible way to decide
Shortlist two or three, give each the same honest brief, and compare how they respond rather than how they present. The provider who asks you good questions — about your workflow, your risks, your busiest hours — is usually the one who will look after you well. Substance in the conversation predicts substance in the service.
If you would like a straightforward chat about the right level of support for your business, whether that is occasional help or ongoing onsite IT support, call PC Macgicians on 020 7610 0500, drop into the Putney workshop on Lower Richmond Road, or use the contact form. We are happy to tell you honestly what you do and don’t need.
