Quality Review Agent
The Quality Review Agent (QA Agent) reviews closed tickets against a set of built-in checks plus your own custom instructions, and surfaces what needs attention on the Service Quality dashboard. It doesn’t replace a manager’s review — it makes sure every ticket gets one.
For full setup steps and advice on writing instructions, see Setting Up the QA Agent.
What it does
Five checks, each with its own on/off toggle:
| Check | What it looks at |
|---|---|
| Documentation Gap Detection | Real work was done but left no useful trace in your knowledge base. |
| Minimum Resolution Detail | Closing notes are too thin for the work performed. |
| Professionalism & Tone | Outbound, technician-written replies read professionally. |
| Time Entries Validation | Logged time, work type and billing align with the work described. |
| Customer Satisfaction | The ticket’s messages show signs of a frustrated or at-risk customer. |
What triggers it
A ticket being closed. You choose the schedule:
- Realtime — each ticket is reviewed as soon as it’s closed.
- Daily, Weekly or Biweekly — everything closed since the last run is reviewed in one batch.
Excluded boards, queues or ticket types are skipped by automatic runs but can still be reviewed manually at any time.
Configuration
- In the Mizo console, turn on QA Agent on the Automation (Workflow) page. This is an admin-only, company-wide setting.
- Open the QA tab under Workflow settings to configure:
- which checks are on;
- the schedule, and an optional digest email to owners and managers on the same cadence;
- boards, queues or ticket types to exclude — see Exclude alert boards from quality review;
- Custom instructions — one free-text box of up to 5,000 characters describing your team’s standards.
Custom instructions refine the built-in checks rather than turning them off. State the default and the exception (“Treat Remote Business Hours as included work by default. Do NOT flag included agreement work unless…”), and pair every “flag when” with a “don’t flag when”.
Human-in-the-loop vs autonomous
The review runs automatically. Acting on findings is up to your team: every finding can be marked Resolved, Dismissed (false positive) or Dismissed (not applicable). Dismissals are the best signal for what to tighten in your custom instructions.
The QA settings also list actions that write back to the PSA — changing a ticket’s status and posting a summary note — each with its own Off / Recommendation / Automatic mode. See Setting Up the QA Agent for their current availability.
What’s written
Findings go to the Service Quality dashboard, grouped by ticket, each with a severity (info, low, medium or high) and, where relevant, the evidence and a recommendation. The dashboard summarizes QA runs, clean rate and high-severity count for the selected period, with breakdowns by severity, category and technician.
Supported PSAs
ConnectWise PSA, Autotask and HaloPSA. The QA Agent reviews real tickets pulled from your connected PSA.
Tips
- Treat the first weeks as a baseline. Sort early findings by type and remove the ones you’d never act on — alert tickets without time entries are the usual suspects.
- Write your definition of “good” down, for example “Resolution notes for recurring issues must list the steps taken, not just ‘resolved’.”
- Route high-severity findings (billing errors, missing billable time) to someone the same day; leave low-severity trends for the digest.