Very few organisations leave their IT provider because of one dramatic failure. They leave because of an accumulation: the same issue for the fourth time, another update that never arrived, a project that's been "on the plan" since last year. Each item is small enough to let go. Together they add up to a team that has quietly stopped expecting things to improve.
Worth saying up front: switching isn't always the answer. Some of these signals are fixable with one honest conversation, and we'd rather you had that conversation than moved for the sake of moving. But if several of them look familiar, it's reasonable to start looking.
1. The same problems keep coming back
Tickets get closed, then reopened a fortnight later with the same symptom. That's the clearest sign you're getting workarounds rather than fixes. A provider looking after you properly should be able to name the three issues that generate the most tickets and tell you what they're doing about the underlying cause.
2. Nothing happens unless you chase
If your team has learned that a ticket needs a follow-up email to move, the service has effectively become self-service with extra steps. Good support chases itself. You shouldn't be the project manager for your own support requests.
3. You only hear from them when something breaks
No review meetings, no report on what's been happening, no view on what's coming. IT support that's purely reactive can still be competent, but it means nobody's thinking a year ahead on your behalf. If your provider hasn't raised your ageing servers, your renewal dates or your security gaps, it's usually because nobody's looking.
4. Every answer starts with a quote
Advice that always arrives attached to something to buy is hard to trust. A good partner will sometimes tell you the honest answer is to do nothing yet, or to fix a process rather than buy a product. If you can't remember the last time you were told something wasn't worth doing, that's worth noticing.
5. You can't get a straight answer about security or backups
- Ask when a restore was last tested and what it showed.
- Ask which accounts don't have multi-factor authentication and why.
- Ask what happens to an account when someone leaves, and who checks it happened.
Three fair questions. If the answers are vague, reassuring or arrive a week later, you're carrying more risk than you think you are - and it's your risk, not the provider's.
6. Your team has stopped reporting things
This one's easy to miss because ticket numbers go down. People decide it's faster to work around the problem than to raise it. Printing from a different machine, keeping a personal copy of the file, using their phone instead. That's not a quieter IT estate. It's a team that's given up, and it hides real problems until they get expensive.
7. Nobody knows how anything is set up
If documentation lives in one engineer's memory, you're one resignation from an outage. Ask to see what's documented about your environment. A provider who can share an up-to-date picture of your systems, licences and admin accounts is looking after you. A provider who can't is holding you in place through inconvenience rather than value.
Before you move: have the conversation
Providers can't fix problems they haven't been told about, and plenty of relationships recover from one clear discussion. Go in with specifics rather than a general sense of frustration: the three issues that recur, the response times you actually experienced last month, the two things you need on the roadmap this year. Ask for a written plan with dates. Then give it a quarter.
If the plan arrives and things improve, you've saved yourself a migration. If nothing changes after a quarter of clear expectations, you have your answer without having to argue about it.
What a sensible handover looks like
The fear that keeps organisations in place is the switch itself. Done properly, it's an organised piece of work over four to eight weeks, and your team should barely feel it.
- Discovery first: systems, licences, admin accounts, contracts and renewal dates, written down.
- Access transferred deliberately, with the old provider's administrator accounts removed at a known point.
- The support route changed once, with clear communication, rather than two desks running in parallel indefinitely.
- The known problems picked up in the first month, so the change is visible to your team quickly.
- A review at ninety days against what was promised.
Check your notice period and what happens to your data and documentation on exit before you commit to anything. Any provider worth moving to will talk you through their own exit terms without flinching.
Where to start
You don't need to decide today. Write down the last five things that frustrated your team and how long each took to resolve. That short list is a better basis for a conversation - with your current provider or a new one - than any comparison of contracts.