Why managed IT fails in SMEs

When managed IT fails in an SME, the cause is rarely technical: it almost always comes down to a vague contractual scope, a provider stuck in reactive mode, service commitments that were never formalised, or a relationship that nobody steers anymore. The good news: these causes can be detected early, and most can be prevented at signing time. Here are the five most frequent failure mechanisms in French-speaking Switzerland, and how to neutralise them.

1. The vague scope: failure programmed at signature

"Management of the company's IT infrastructure": this sentence, or a variant, appears in countless managed IT contracts. It means nothing. Which workstations? Which servers? Is the business software included? The printers? The receptionist's phone? The meeting-room Wi-Fi?

Every grey area becomes a future dispute: the client believes it is covered, the provider bills extra or refuses to intervene. After a few episodes, trust is dead — and without trust, managed IT does not work, whatever the technical level. The solution is well known: a precise inventory annexed to the contract, and an explicit list of exclusions. Our managed IT contract comparison guide details the clauses to demand.

2. The firefighter provider: paid to be proactive, working reactively

This is the most widespread cause of failure. The point of managed IT is anticipation: continuous monitoring, updates applied, backups tested, problems fixed before they become visible. Yet many providers sell a monthly fee and then operate like repair shops: they wait for the ticket.

The symptom is measurable: if the number of incidents has not dropped after six months, if the same problems come back every month, if no unprompted recommendation ever arrives, you are paying a subscription for break-fix. The monthly report is the best detector — a proactive provider shows its actual response times, the state of the backups and what it fixed before you noticed.

3. Missing SLAs: expectations never formalised

Without quantified response times, everyone has their own definition of "fast". The client expects an intervention within the hour, the provider considers 48 hours reasonable: both are in good faith, and the relationship deteriorates with every incident. A written SLA — for example, a critical incident handled within 4 business hours — is not bureaucracy: it is what aligns expectations on both sides before the first problem.

4. Chronic under-investment

Failure also comes from the client side. A provider can monitor and maintain, but it cannot make a ten-year-old server, a never-replaced firewall or workstations on an end-of-life operating system reliable. When every investment recommendation is refused year after year, incidents multiply — and managed IT gets the blame.

The same mechanics apply to contracts negotiated to the bone: remove off-site backup "to save money", then security, then extended support, and you get a contract that no longer covers anything and will fail by construction. A realistic IT budget for an SME — our article on managed IT services gives the ranges in CHF — always costs less than the accumulation of incidents it prevents.

5. Nobody steers the relationship

Managed IT is not a subscription you sign and forget. It needs a contact person on the client side — not an IT specialist, but someone who reads the monthly report, arbitrates recommendations and reviews progress with the provider two to four times a year. Without that minimal steering, decisions pile up unanswered, the provider loses motivation, and the value of the contract silently erodes. If your IT involves several providers, our guide on managing several IT providers completes this point.

How to restart on solid ground

If your current managed IT is failing, prepare the exit: check the reversibility clause (return of data, passwords, documentation), choose the successor before terminating, and require a documented initial assessment from them. Taking over an infrastructure managed by another provider is routine for a serious professional — at AlpenData it is even the most common starting point for AlpenCare: audit of the existing environment, documentation, then progressive takeover, with an initial commitment of only 6 months and monthly renewal afterwards.

Frequently asked questions

What is the first sign that managed IT is failing?

Silence. A provider that no longer sends monthly reports, no longer proposes improvements and is only visible when something breaks has slipped into reactive mode — exactly what managed IT was supposed to prevent. Recurring incidents that are never fixed at the root are the second signal.

How long before you can judge a managed IT contract?

After 6 months. The first three months are for taking over the existing environment, documenting and stabilising; from the fourth month, proactive mode must be visible: fewer incidents, regular reports, unprompted recommendations. If nothing has changed after 6 months, the problem is structural.

Is failure always the provider's fault?

No. Managed IT also fails on the client side: expectations never formalised, no internal contact person, systematic refusal of recommended investments (obsolete hardware replacement, security), or a scope negotiated down so far that it excludes half the infrastructure. The cheapest contract that covers nothing always fails.

How do you exit a managed IT contract that isn't working?

First check your exit clause: return of data, administrator passwords and documentation, with a defined timeframe. Select the new provider before terminating, and plan a coordinated transition period. A serious provider knows how to take over an existing infrastructure — it is a common situation.

Can remote-only managed IT work for a French-speaking Swiss SME?

Partially. Monitoring, updates and much of the support work remotely. But hardware failures, deployments and the relationship of trust require a real local presence. A provider without technicians able to travel in French-speaking Switzerland ends up handling physical problems with unacceptable delays.