What Are Managed Services (MSP)? A Model That Complements Your IT Team
Nanda Teknoloji Engineering Team · · 8 min read
In brief
A managed service provider (MSP) takes on, under contract, the operational work of an organization's IT infrastructure, such as monitoring, maintenance, patching, backup, and incident management. Rather than replacing the in-house team, it is a model that complements its capacity and provides continuity.
A managed service provider (MSP) operates, under contract, certain parts of an organization’s IT infrastructure through ongoing work such as monitoring, maintenance, patching, backup, security operations, and incident management. The important distinction is this: from an MSP you buy not people or hours, but a defined service and its measurable outcome. This model does not replace the in-house IT team; it complements the team’s capacity, secures on-call coverage and continuity, and lets the team spend time on strategic work. This article explains what an MSP covers, how responsibilities are shared, in which situations it makes sense, and what to look at when choosing a provider.
What does the MSP model cover?
Scope varies from provider to provider and contract to contract; so saying “we use an MSP” says nothing on its own. Typical service areas are:
- Monitoring and alert management: Continuous monitoring of server, network, storage, and application health; classification of alerts.
- Patch and update management: Planned updates of the operating system, hypervisor, and middleware software.
- Backup and recovery operations: Monitoring backup jobs, restore tests, and applying retention policies.
- Incident and change management: Opening and resolving fault tickets, root cause analysis, and making changes through an approved process.
- Security operations: Vulnerability scanning, access reviews, log collection; incident response depending on scope.
- Capacity and performance management: Monitoring growth trends and making recommendations.
In areas such as backup and recovery, thinking about the details of the service together under managed services and backup solutions keeps responsibilities clear.
Shared responsibility: who is responsible for what?
The place where MSP contracts most often cause problems is the boundaries of responsibility. The sentence “they manage the infrastructure” does not say who is responsible if the application crashes. Put it in writing with a RACI table:
| Area | Organization (in-house team) | MSP | Note |
|---|---|---|---|
| Business requirements and priorities | Responsible | Consulted | The business unit sets the priority |
| Infrastructure monitoring and alerting | Informed | Responsible | Thresholds are set together |
| Patch application | Approver | Responsible | The maintenance window is agreed in advance |
| Application code and version | Responsible | Supporting | Generally out of scope |
| Operation of backup jobs | Approver | Responsible | Restore testing is included |
| Access authorization | Approver | Implementer | The access request comes from the internal system |
| Incident communication (internal and external) | Responsible | Provides technical input | Communication authority stays with the organization |
Adapt the table to your own environment and attach it to the contract. Gray areas turn into arguments at a moment of crisis.
When does an MSP make sense?
The model is not right for every organization. The following situations are strong signs to evaluate an MSP:
- The in-house IT team is small and keeping 24/7 on-call coverage is not realistic.
- The team spends most of its time on patching, alerts, and recurring failures; project work is constantly postponed.
- A specific expertise (storage, virtualization, security operations) depends on a single person, and the knowledge disappears if that person leaves.
- The infrastructure has become complex quickly because of growth or a merger.
- Audit and reporting obligations require regular evidence production.
Conversely, if the infrastructure is small and stable, and the team already finds time, a full-scope MSP can be an unnecessary cost; in that case, handing over a specific area (for example, only monitoring or backup operations) is more balanced.
What to look at in the contract and SLA
A service level agreement (SLA) makes the promise measurable. When talking to a provider, ask for the following explicitly:
- Scope and out-of-scope: Which systems are included, and which are excluded? What is the process for adding a new system?
- Priority classes: The definition of critical, high, medium, and low incidents and the response and resolution targets for each.
- Measurement method: When does the clock start, and when does it stop? With which tool is it measured? Who receives the reports?
- Working hours: Is the service 24/7, or within business hours? How do holidays and weekends work?
- Remedies: What happens when a target is missed? Without remedies, an SLA is only a statement of intent.
- Exit conditions: When the contract ends, how are documentation, configuration, and access handed over?
To compare providers systematically, you can use the question set in the article on the IT vendor evaluation form.
Security and access management
Privileged access given to an MSP is an attack surface in its own right. That is why technical controls should be defined along with the contract:
- An individual account for every technician; no shared administrator account.
- Multi-factor authentication and, where possible, time-limited, request-based privilege elevation.
- Recording of privileged sessions and keeping logs in a place the organization can also access.
- Quick closure of access when personnel change; regular (for example, quarterly) access reviews.
- Data processing agreement: if access to personal data is involved, obligations as a processor must be in writing. As a general security framework, the NIST Cybersecurity Framework can be used to establish a common language with the provider.
How is the transition done?
The move to an MSP does not end when the contract is signed. A healthy transition follows these steps:
- Discovery: Inventory, existing monitoring, backup status, known risks, and technical debt are identified.
- Baseline: The current state is measured before the service starts, so improvement can be proven.
- Gradual takeover: Monitoring and reporting are handed over first, then patching and backup, and incident management last.
- Process integration: The ticketing system, communication channels, and escalation matrix are built together.
- First review: At the end of the first quarter, scope, alert noise, and SLA performance are evaluated.
When handing over critical work such as disaster recovery, regular testing should be done to verify the service actually works; the article on the disaster recovery test checklist can be used for this.
Example scenario: a three-person IT team
A logistics company’s three-person IT team manages about forty servers, several warehouse networks, and an ERP system. Night on-call duty is held in rotation, and when one team member goes on leave, capacity drops sharply. Patch and backup checks are done “when there is time.” The company decides to hand over not the entire infrastructure but only three areas: 24/7 monitoring and alerting, monthly patch management, and the operation of backup jobs plus a quarterly restore test.
After three months, the in-house team is relieved of the on-call load and devotes its time to the ERP integration project. Alert noise stays high in the first weeks; thresholds are readjusted together. Architectural decisions and relationships with the business unit remain with the in-house team. The important point here is that the MSP does not “take over” the work, but carries out, at a defined standard, the continuous work the in-house team could not.
Pricing models: what do you pay for?
MSP services are usually priced with one of the following models or a mix of them. Which one you choose determines the incentive structure:
- Per managed asset: A fixed fee per server, device, or user. It is predictable; when the asset count changes, the fee changes too.
- Tiered packages: Tiers such as monitoring, monitoring plus patching, and full management. Easy to compare when the scope is clear.
- Fixed fee plus usage: The base service is fixed; extra projects and out-of-scope work are charged by the hour or day.
- Outcome-based: A fee tied to certain targets (for example, an availability level); the measurement method must be very clear.
Whichever model it is, see in the quote the unit price for out-of-scope work and the process for adding a new system. Unexpected line items are the most common source of lost trust in MSP relationships.
Common mistakes
- Leaving scope vague: The expectation of “everything included” leads to disappointment in the first crisis.
- An SLA without targets: Promises that cannot be measured or have no remedies.
- Handing over knowledge completely: If no infrastructure knowledge remains in the in-house team, the organization becomes dependent on the provider. Require that the documentation belong to the organization.
- Leaving security for later: If privileged access and logging are not in the contract, adding them later is difficult.
- Not managing the transition period: Alert noise and uncertainty are high in the first weeks; a project owner should be assigned to manage this.
- Not reading the reports: If the monthly report only goes into the archive, the SLA has no value.
Decision summary
- Write down what consumes the in-house team’s time and which risks worry you most.
- Decide whether you want a full-scope or area-by-area handover.
- Clarify responsibilities with a RACI table.
- Evaluate the SLA together with its measurement method, remedies, and exit conditions.
- Include privileged access controls in the contract and in the technical configuration.
- Make the transition gradual and review it at the end of the first three months.
Moving forward with Nanda Teknoloji
On projects like this, Nanda Teknoloji carries out the discovery, design, migration, and operations steps: assessing the current environment, defining the scope together, gradual takeover, and regular reporting. If you are looking for an operating model that complements your in-house team, you can start from the managed services page and narrow or widen the scope according to your needs.