Custom Fields for Insights
Custom Fields are where MiaRec stores AI insight outputs (scores, categories, extracted values). Once a field is populated, it becomes usable in:
- Conversation details (Analytics view)
- Dashboards (averages, distributions, trends, drilldowns)
- Search and filters (e.g., CSAT < 3)
- Reports/exports (where enabled)
Menu path
Administration > Customization > Custom Fields
When to create a new Custom Field
Create a Custom Field when you want to:
- capture a new business-specific metric (e.g., “Reservation Start Date”)
- store a score (e.g., CSAT, lead score)
- store a classification (e.g., churn risk: Low/Medium/High)
- store extracted entities or text (e.g., “Next actions”)
If a prebuilt AI Task already maps to a prebuilt field, you may not need to create anything—just enable the task.
Field type selection (practical guidance)
Choose the field type based on how you plan to use the value:
- Number (integer/decimal) — scoring, averages, thresholds, range filters
Examples: CSAT (1–5), Lead Score (0–100) - Dropdown / choice — consistent categorization and easy dashboards
Examples: Churn Risk (Low/Medium/High), Call Outcome - Date — extracted dates for timeline analysis
Examples: Reservation Start Date, Follow-up date - Text — summaries, reasons, notes, explanations
Examples: Call summary, “CSAT explanation”, “Top objection detail”
Recommendation: prefer Dropdown over free text for categories you want to chart consistently.
Step-by-step: create a field for AI Insights
- Go to
Administration > Customization > Custom Fields. - Click Add (or equivalent).
- Configure:
- Name (human-readable)
- Computer name (stable identifier; avoid spaces; don’t change after rollout)
- Description (what the field represents)
- Type (Number / Dropdown / Date / Text)
- Enable AI Insights integration:
- Check the AI Insights checkbox to make the field available to AI Tasks
- Configure display:
- Choose a Display group (e.g., "CX Metrics", "Sales Metrics") so the value appears in Conversation Details → Analytics.
- Save the field.
Figure: Custom Field configuration showing AI Insights checkbox and Dashboard option.
Thresholds and labels (scores like CSAT)
For numeric scoring fields, thresholds help you:
- show labels (e.g., “Dissatisfied”)
- apply colors for quick scanning
- create clickable buckets in dashboards
Example CSAT thresholds (1–5): - 1 = Very Dissatisfied - 2 = Dissatisfied - 3 = Neutral - 4 = Satisfied - 5 = Very Satisfied
Keep thresholds consistent across teams to avoid KPI confusion.
Creating dashboards from fields (recommended for key metrics)
Enable the Dashboard checkbox when creating the field to automatically generate a dashboard. This creates:
- average score display
- distribution buckets (based on configured thresholds)
- trend over time
- clickable drilldown to matching conversations
Figure: Threshold configuration for numeric fields. Define buckets with labels and colors for dashboard visualization.
Storing "value + explanation"
Many AI insights are best represented as: - a structured value (score/category/date) - plus an explanation for a human reviewer
MiaRec stores explanations automatically alongside the insight value. When you configure an AI Task to return both a value and an explanation, the explanation is displayed in the Conversation Details view next to the metric value without requiring a separate custom field.

