Choose the function you need
| Function | Primary work and team | Buy when the gap is | Targets to define |
|---|---|---|---|
| NOC | Infrastructure operators monitor systems, triage alerts and perform permitted fixes | Alerts, patch follow-up or infrastructure escalation outstrip your coverage | Alert intake, action, restore and escalation |
| Help desk | Support technicians handle users, applications and service requests | Client contacts, common requests or overflow need a supported queue | Contact acknowledgement, useful action, restore and request completion |
| SOC | Security analysts investigate suspicious activity and coordinate approved containment | Security detection, investigation or response needs dedicated coverage | Security triage, escalation and authorized containment |
The service schedule decides the permitted actions. For example, receiving an endpoint alert and isolating that endpoint are separate responsibilities. Keep change approval, client communications and the authority to contain a suspected compromise explicit.
Route an incident across the teams
| Starting signal | First question | Next owner to agree |
|---|---|---|
| Server health alert | Is an approved infrastructure fix available? | NOC, with an MSP approver for restricted changes |
| User cannot log in | Has identity been checked and is the request supported? | Help desk, with application or identity escalation |
| Suspicious endpoint behavior | Is security investigation or containment required? | Security owner, with the agreed incident commander |
| Backup job failure | Is this a routine job error or a recovery problem? | Monitoring owner, then the authorized recovery team |
| Application unavailable | Is the cause infrastructure, vendor, identity or security related? | One coordinating incident owner across the relevant teams |
Use a parent incident when several teams are involved. Each team can finish its assigned action while the coordinating owner keeps the customer impact, next decision and update deadline visible. Close the parent only when its agreed recovery or closure condition is met.
Make the handoff usable
The receiving team needs the tenant and asset, business impact, observed signal, actions taken, approval limits and a specific next action. Name the accepting owner and the fallback if that person cannot be reached. The after-hours handoff example shows those fields in one incident.
See the example below for an application outage accompanied by a suspicious endpoint signal.
The joint CISA advisory for MSPs and their customers (external website) emphasizes a shared commitment to security. Use the supplier conversation to assign that responsibility in your own service arrangement.
Example: an outage with a suspicious endpoint signal
The following is a proposed tabletop scenario, not an incident observed at a listed provider. An application is unavailable while a security tool flags an endpoint used by an affected user. The reports may be related or separate. Ask each supplier what it will do, who can authorize the next action and how the parent incident stays visible across the queues.
MSP incident lead owns the impact summary, decisions, next update and accepting handoff. The relationship between the suspicious endpoint signal and the outage is still being investigated.
Help desk: user impact
Document affected users and applications, link related contacts to the parent and give the approved status update. Keep identity-sensitive requests on their authorized verification route.
NOC: infrastructure condition
Supply the affected tenant, asset, signal and diagnostics. If a proposed restart is outside the agreed authority or conflicts with the security response, ask the incident lead for a decision before acting.
SOC / security owner: investigation and containment
Assess the suspicious signal and state the proposed containment, required authority and operational effect. Document the decision and communicate it to the NOC and desk.
A completed diagnostic task can close its own work item. The parent stays open until the agreed recovery and closure criteria are met; investigation, containment, restored service and final closure are separate states.
Agree this division with the actual suppliers before use. The scenario does not authorize a restart, isolation or recovery action.
Agree the decision and the closure condition
Before any containment or restoration step, identify the approver, the evidence needed, the effect on users and the team that will execute the decision. An unavailable approver needs an agreed fallback. A ticket must not lose its coordinating owner while that fallback is contacted.
For this example, the lead asks the NOC and security owner to reconcile the proposed infrastructure action with the investigation. The desk continues the approved customer update. After an authorized recovery, the system owner confirms the application works; the security owner documents the security outcome and remaining actions. The lead applies the agreed closure criteria and keeps any follow-up work assigned.
This coordination approach follows the distinction in NIST SP 800-61 Rev. 3 (external website) between a designated incident lead, coordinated response and recovery criteria. Its role guidance includes contracted third parties, while its recovery guidance calls for confirming restored services and completing the incident documentation. The example here is an editorial application, not a NIST-prescribed supplier workflow.
Prepare the next action, decision owner and update deadline in the incident handoff worksheet. Keep detailed evidence in approved systems and use sanitized references in the draft. In the RFP, ask whether cross-team escalation, out-of-hours approval or containment needs a separate service or fee.
Compare the same obligations and units
Request separate scope and fee lines for each function. A NOC endpoint fee, help desk user fee and security log-volume fee measure different things. Include the MSP time needed to coordinate teams and approve restricted work. Ask whether a cross-team escalation creates a second charge or waits for a different service window.
Use the NOC guide and help desk guide to define the operational purchase. Put measurable timers in the SLA worksheet and test the shared handoff during the NOC pilot or help desk pilot.