Why procedures
The procedures that decide whether a provider stays secure
Most serious incidents at managed service providers do not start with an exotic exploit. They start with a procedure that did not hold: a technician who left the company still in a delegated admin group, a password reset for a caller nobody verified, a script pushed to every endpoint without a second pair of eyes, a new client whose old provider still had an agent installed.
This part of the MSP security audit tests those procedures the way an auditor would test them for SOC 2 or ISO/IEC 27001: we read what is written, interview the people who do the work, and then sample the records to see whether the procedure was followed every time. A procedure that exists only in a document scores Partial. A procedure that is followed but not written down scores Partial too. Only a documented procedure with consistent records passes.
- Workforce, service delivery, incident response and vendor areas
- Records sampled and redacted with you
- Every gap mapped to a business risk and an owner

What we test
Nine procedures and the evidence behind them
For each procedure we state the standard a provider should meet and the records we use to test it.
| Procedure | What good looks like | Evidence we sample |
|---|---|---|
| Technician offboarding | Same-day removal of every access path: identity provider, RMM, PSA, documentation platform, delegated admin groups, client firewalls and backup portals, plus rotation of secrets the person knew. | Offboarding tickets for recent leavers matched against tool user lists and delegated admin group membership. |
| Technician onboarding and role changes | Access granted by role through a documented request and approval, and recertified quarterly so people only reach the clients they support. | Access requests, approvals and the latest recertification record. |
| Help-desk identity verification | Password resets, MFA resets and access changes only after a callback to a known number, a verification code or manager approval; never from an unverified email or chat. | A sample of reset tickets checked for the verification step and who performed it. |
| Client onboarding | Asset discovery, takeover and rotation of every administrative credential, removal of the previous provider’s access and agents, and baseline applied. | The onboarding checklist and the records for the most recent new clients. |
| Client offboarding | Hand-over of credentials and documentation, removal of all MSP access, agents and delegated relationships, and return or deletion of client data. | Records for recently departed clients compared with tool and tenant access lists. |
| Change management | Request, risk assessment, approval for high-risk changes, an emergency path, client communication and a record for every change in client environments and networks. | Change records sampled against firewall and configuration history. |
| Script and mass-action control | Review and approval of RMM scripts and automation, a second approval before mass deployment and dual control for credential exports and backup deletion. | Script library history, approvals and recent mass runs. |
| Incident response and client notification | A written plan with an offline copy, playbooks for MSP-compromise scenarios, client notification timelines and an annual multi-client tabletop exercise. | The plan, the last tabletop report and the tracking of its lessons learned. |
| Risk acceptance and exceptions | Declined recommendations recorded, signed by the client and re-presented periodically, so the minimum baseline is either applied or formally declined. | The exception register and the signed acceptances behind it. |
The help desk
Help-desk identity verification is the procedure attackers test first
Social engineering of the help desk is one of the most direct routes into a client: a caller claims to be a manager, asks for an MFA reset, and the attacker owns the account minutes later. In our sample MSP report the missing help-desk verification control was the single finding with the widest reach, because it applies to every client the provider supports.
We look at the written verification rule, at how technicians reach the caller (a callback to a number already on file, not the one the caller gives), at manager approval for privileged changes, and at a sample of real reset tickets to see whether the verification step is recorded.
What we expect to see
- A written rule that no reset is done from an unverified email, chat or caller ID
- Callback to a known number or a verification code for every reset
- Manager approval before privileged or MFA changes
- The verification step recorded in the ticket
- Training and periodic testing of the service desk
How we test a procedure
From the document to the record
Read
We read the written procedure, checklist or SOP and note what it requires and who owns it.
Interview
We ask the people who perform it how they actually do it, including the exceptions.
Sample
We pick recent records, such as offboarding or reset tickets, and agree the sample with you.
Compare
We compare each record with the procedure and with tool access lists and logs.
Rate
We rate the control, describe the gap, the business impact and the fix, and link it to a risk.
Related pages
Continue through the MSP audit

Tooling
RMM and privileged access review
Remote management, the credential vault and delegated admin access into every client tenant.
See the review →
Checklist
MSP security checklist
The 18 areas and the controls inside them, in one list your team can work through.
Open the checklist →
Deliverables
Sample report
Every report from a full audit cycle for a fictional provider, page by page.
See the sample →
Who leads the audit
Led by Ali Hassani, CISO
Certified CISO (C|CISO) and CISSP with 25+ years in IT, cybersecurity, compliance and infrastructure, and dozens of Microsoft and Cisco certifications, including MCSE, MCITP, MCSA Security, CCNP and CCNA.
Every MSP security audit is planned, performed and signed by Ali, not handed to a junior analyst. We stay independent: we assess and advise, and your team carries out the fixes, with our help if you want it.
Questions and answers
Process and procedure audit questions
What is an MSP process and procedure audit?
It is the part of the MSP security audit that tests how your team works, not only how tools are configured: offboarding, onboarding, help-desk verification, client onboarding and offboarding, change management, script approval, incident response and risk acceptance. Each procedure is checked against real records.
Why do procedures matter as much as tools?
Most of the critical findings at providers are procedural. A technician who left keeps a delegated admin group membership, a reset is done for a caller nobody verified, or a mass script runs without a second approval. The tools were capable; the procedure did not hold.
What records do you sample?
Typically offboarding tickets for recent leavers, a set of password and MFA reset tickets, change records, client onboarding and offboarding checklists, restore test results and the latest access recertification. We agree the sample with you and redact client details.
Do you write our procedures for us?
The audit tells you which procedures are missing or not followed and what each one must contain. If you want help writing or rolling out the procedures, that can be scoped separately, and your team stays in control of how they are run.
Is this a separate audit with its own price?
No. Procedures are tested in every MSP Security Audit, which starts at $3,500 and covers all 18 areas.
How are findings rated?
Each control is rated Pass, Partial or Fail and each finding Critical, High, Medium or Low. Findings are grouped into business risks with an owner and a target date, and the roadmap puts Critical items into a 7 to 30-day window.

See whether your procedures hold up on the record
We test them against your own tickets and lists, read-only, as part of the MSP Security Audit, starting at $3,500.
Or call 949-777-5567
