Fine tuning
Mizo’s AI-powered automation can be fine-tuned to match your organization’s specific workflows, priorities, and business logic. This guide provides comprehensive instructions for configuring the AI behavior through the Fine-Tuning tab in the Mizo console.
Key Principles:
- Fine-tuning rules are applied as additional instructions to the AI
- Rules are processed in the order they appear
- More specific rules should be placed before general rules
- Rules can enable, disable, or modify AI behavior for specific scenarios
How Fine-Tuning Works Rule Structure
Each fine-tuning rule consists of:
- Condition: The criteria that triggers the rule (e.g., ticket title pattern, content keywords, company name)
- Action: What the AI should do when the condition is met (e.g., select a specific value, skip a step, apply a pattern)
- Context: Additional information to help the AI make better decisions (e.g., definitions, exceptions, examples)
Rule Priority
- Rules are evaluated in the order they are written
- Earlier rules take precedence over later ones
- Use this to create fallback logic (specific cases first, general cases last)
Writing Effective Rules
Good Rule:
If the ticket title contains “Printer” or “Print”, categorize it as “Hardware / Printers”.
Poor Rule:
Sometimes tickets about printers should go to hardware. Why? The good rule is specific, actionable, and unambiguous. The poor rule is vague and leaves too much to interpretation.
Fine-Tuning Categories
Mizo supports fine-tuning across several AI processes
| Category | Purpose | Ai Process |
|---|---|---|
| Ticket Title Rules | Control ticket renaming behavior | Triage |
| Ticket Categorisation Rules | Guide ticket type classification | Triage |
| Type Definitions | Define what each ticket type means | Triage |
| Urgency & Impact Rules | Set priority levels | Triage |
| Category Rules | Guide ticket categorization | Triage |
| Attribution Rules | Control ticket-to-contact assignment | Attribution |
| Dispatch Rules | Guide ticket assignment to teams/users | Dispatch |
| Team Definitions | Define team expertise and responsibilities | Dispatch |
Triage Fine-Tuning
1.Ticket Title Rules
Controls how Mizo renames tickets for clarity and consistency.
Disabling Renaming
Use Case: Preserve original titles for template tickets or specific patterns • If the ticket title starts with “Order #”, do not rename the ticket. • If the ticket title contains “URGENT:” at the beginning, do not rename the ticket. • If the ticket title is “Monthly Report Delivery”, set the updated title to “undefined”.
Custom Renaming Patterns
Use Case: Apply consistent naming conventions
- Rename tickets using this pattern: [{ClientName}] {UpdatedTicketTitle}
- For VIP clients (ACME Corp, Global Industries), use this pattern: [VIP] {UpdatedTicketTitle}
- If the ticket is from an end user (not an internal email), prepend the title with the end user’s name: {EndUserName} - {UpdatedTicketTitle}
Conditional Renaming
Use Case: Different rules for different ticket types
- If the ticket is about password reset, use the pattern: “Password Reset - {EndUserName} - {SystemName}”
- If the ticket mentions “down” or “outage”, prepend “OUTAGE:” to the title.
- For tickets about software requests, use: “Software Request: {SoftwareName} for {EndUserName}”
Advanced Example:
Ticket Renaming Rules:
1. Template Tickets (Preserve Original):
- If the title starts with “Send report -”, do not rename
- If the title starts with “Order #”, do not rename
- If the title is exactly “User Departure: {Name}”, do not rename
2. Emergency/Outage Tickets:
- If the description contains “down”, “outage”, “offline”, or “not working” and affects multiple users,
use pattern: “[OUTAGE] {SystemName} - {Problem}”
3. User Request Tickets:
- Use pattern: {EndUserName} - {RequestType} - {Summary}
- Example: “John Smith - Software Install - Adobe Acrobat”
4. Default Renaming:
- Maximum 80 characters
2. Ticket Urgency & Impact Rules
Controls how Mizo sets priority levels (urgency and impact).
Disabling Priority Setting
Use Case: Preserve manually set priorities or skip for certain templates
- If the ticket title contains “Monthly Report”, do not set urgency or impact.
- If the ticket is created by the “Automation User”, do not modify urgency or impact.
For tickets with “Template:” in the title, do not set urgency or impact.
Setting Specific Priorities
Use Case: Enforce priority levels for critical scenarios
- If the ticket mentions “server down”, “complete outage”, or “network down”, set urgency to High and impact to Company-Wide.
- If the ticket is about password reset, set urgency to Medium and impact to Single User.
- If the client is “VIP Corp” and the ticket contains “urgent”, set urgency to High.
Advanced Example:
1. Priority Rules:
. VIP Client Rules:
- Clients: ACME Corp, Global Industries, Premium Partners LLC
- Minimum urgency: Medium
- If they use “urgent” or “ASAP”, set urgency to High
2. Template Tickets (Skip Priority):
- If title starts with “Scheduled -” or “Report -”, do not set priority
- These tickets have predefined SLAs
3. Ticket Category Rules
Controls how Mizo categorizes tickets into your service desk’s category structure.
Category Selection Rules
Use Case: Guide the AI to the correct category
- If the ticket is about “Novipro”, categorize as “ERP”.
Never select the “Archived” category or any of its subcategories. \
Dispatch Fine-Tuning
4. Team Definitions
Defines what each team handles and their expertise areas.
Use Case: Help AI understand team responsibilities
Team Definitions:
Service Desk (Tier 1):
- Handles: Password resets, basic how-to questions, initial triage
- Skills: Windows basics, Office 365, Active Directory basics
- Escalates to: Infrastructure Team (network/server issues), Desktop Team (hardware)
- SLA: 15-minute response time
Desktop Support Team:
- Handles: Hardware issues, laptop/desktop problems, printer issues, mobile devices
- Skills: Hardware repair, Windows troubleshooting, driver installation
- Escalates to: Service Desk (after hours), Vendor Support (warranty issues)
Infrastructure Team:
- Handles: Server issues, network problems, storage, backups, virtualization
- Skills: Windows Server, VMware, networking, SAN/NAS, backup systems
- On-call: 24/7 for critical issues
- Escalates to: Vendor Support (hardware failures)
Application Team:
- Handles: Line-of-business applications, database issues, custom software
- Skills: SQL Server, .NET applications, web applications
- Escalates to: Software Vendor (application bugs)
Security Team:
- Handles: Security incidents, malware, suspicious activity, access control
- Skills: Firewall, antivirus, SIEM, security policies
- Priority: All security tickets are high priority
- Available: 24/7 for security incidents
7. Update Satus Rules
In each category — Triage, Dispatch, and Resolution — you can add rules.
Example:
When a ticket is dispatched, change the status to “Dispatched.”
6. Dispatch Rules
Controls how tickets are assigned to teams and technicians.
Team Assignment Rules
- Always assign password reset tickets to the Service Desk team.
- Assign all tickets from VIP Corp to the “Premium Support” team regardless of ticket type.
- If the ticket is about servers or network infrastructure, assign to the Infrastructure Team.
- Do not assign tickets to the “Training” team - this team is for internal use only.
Conditional Dispatch Logic
1. Team Assignment Rules:
1. Issue Type Routing:
- Security-related keywords (“phishing”, “malware”, “breach”) → Security Team
- Hardware issues → Desktop Support Team
- Server/network issues → Infrastructure Team
- Application-specific issues → Application Team
- Everything else → Service Desk Team
User Assignment Rules
- If the ticket is for “ACME Corp” and involves SQL Server, assign to John Smith (SQL expert).
- If the ticket is about the “Legacy CRM” application, assign to the Legacy App team or Sarah Johnson.
- Do not assign tickets to users in the “Disabled Users” list: Bob Wilson, Jane Doe.
- For tickets about Office 365 admin tasks, assign to users with “O365 Admin” in their team description.
Attribution Fine-Tuning
7. Ticket Attribution Rules
Controls how Mizo identifies the correct contact and company for catch-all tickets.
Use Case: Improve accuracy for tickets sent to generic mailboxes
Attribution Rules:
1. Internal Tickets:
- If sender email is from @ourcompany.com, it’s an internal ticket
- Do not update attribution for internal tickets
Best Practices
1. Start Simple, Iterate
Begin with a few high-impact rules and expand over time:
- Start with 3-5 rules addressing your most common scenarios
- Monitor results for 1-2 weeks
- Refine rules based on AI behavior
- Add new rules gradually
2. Be Specific and Clear
Good:
- If the ticket title contains “printer” or “printing” (case-insensitive), categorize as “Hardware / Printers”.
Poor:
- Printer tickets should probably go to hardware.
3. Use Examples in Rules
Help the AI understand by providing examples:
- For password reset tickets (examples: “forgot password”, “locked out”, “cannot login”), assign to Service Desk and set urgency to Medium.
4. Order Matters
Place specific rules before general rules:
Good Order:
1. If ticket is from VIP Corp about SQL Server, assign to John Smith
2. If ticket is about SQL Server, assign to Database Team
3. If ticket is from VIP Corp, assign to Premium Support
Poor Order:
1. If ticket is from VIP Corp, assign to Premium Support
2. If ticket is about SQL Server, assign to Database Team
3. If ticket is from VIP Corp about SQL Server, assign to John Smith
(This rule will never be reached!)
5. Document Your Intent
Add comments to explain complex rules:
Team Assignment Rules:
# VIP Client Handling
# VIP clients pay for premium support with faster response times
# They should always go to the Premium Support team regardless of issue type
If the client is “VIP Corp” or “Global Industries”, assign to Premium Support Team.
# Security Incidents
# Security issues need immediate attention 24/7
# Route directly to Security Team
If the ticket contains security-related keywords (list below), assign to Security Team immediately:
- “phishing”, “malware”, “virus”, “ransomware”
- “hacked”, “breach”, “unauthorized access”
- “suspicious email”, “spam with attachment”
6. Avoid Over-Specification
Don’t create rules for every single scenario. Let the AI use its intelligence:
Too Many Rules (Avoid):
- If ticket includes “Space X”, categorize as “Space X —> {Matching Category}”
- If ticket includes “Tesla”, categorize as ” Tesla —> {Matching Category}”
- If ticket includes “Mizo”, categorize as “Mizo —> {Matching Category}”
… (50 more Customer variations)
Better Approach:
- If ticket is from a customer, categorize as “{Customer Name} / {Matching Category}”
- The AI will understand variations naturally
7. Regular Review and Maintenance
Schedule periodic reviews:
- Weekly: Check for tickets that were miscategorized/assigned
- Monthly: Review overall rule effectiveness
- Quarterly: Major rule updates based on changing business needs
Getting Help
If you continue to experience issues with fine-tuning:
- Check the Logs: Review processed tickets to see how rules were applied
- Contact Support: Email [email protected] with:
- Description of the issue
- The rule you’re trying to implement
- Example tickets where it’s not working
- What you expected vs. what happened
- Request a Review: Your Mizo contact can review your fine-tuning configuration and suggest improvements
Appendix: Quick Reference
Triage Quick Reference
| Area | Disable Behavior | Custom Behavior |
|---|---|---|
| Title | Set to “undefined” or “do not rename” | Use pattern: {Field} - {Field} |
| Urgency/Impact | “Do not set urgency or impact” | “Set urgency to [High/Medium/Low]” |
| Category | “Never select category [X]” | “Categorize as [Category / Subcategory]” |
| Type | N/A | Define types, then “Classify as [Type]” |
Dispatch Quick Reference
| Area | Purpose | Example |
|---|---|---|
| Team Definitions | Describe team expertise | “Team X handles: [skills], Hours: [schedule]” |
| Team Assignment | Route to correct team | “Assign [ticket type] to [Team Name]” |
| User Assignment | Assign to specific person | “Assign [client/issue] to [User Name]” |
Attribution Quick Reference
| Area | Purpose | Example |
|---|---|---|
| Company Mapping | Map domains to companies | “Email from @acme.com → ACME Corp” |
| Promotional Detection | Filter spam/marketing | “Keywords: ‘unsubscribe’ → promotional” |
Conclusion
Fine-tuning Mizo’s AI is an iterative process that improves over time. Start with your most important use cases, validate results, and expand gradually. Well-crafted fine-tuning rules will help Mizo automate more of your ticket workflow accurately and consistently.
For questions or assistance, contact [email protected] or your Mizo account representative.