Chart Reference
The Analytics dashboard includes 10 charts organized across 5 rows. This page describes each chart in detail — what it measures, how it computes data, and what to look for.
Interactivity All charts support hover tooltips. Move your cursor over any data point, bar, or area to see exact values. Charts also include legends to identify each data series.
Ticket Volume
| Chart type | Stacked area chart |
| Data window | Last 12 weeks |
| Series | Activity (orange/pink), Development (teal) |
Shows the number of new tickets created per week, split by ticket type. Each week starts on Monday. The stacked view lets you see both the total volume trend and the Activity/Development ratio at a glance.
What to look for:
- Spikes may indicate a client issue or a product release
- A growing ratio of Development tickets may suggest increasing feature requests
- Flat or declining volume suggests the support load is stabilizing
Ticket Types
| Chart type | Donut chart |
| Data window | All time |
| Segments | Activity, Development |
Shows the overall split between Activity tickets (questions, general inquiries) and Development tickets (bugs, features, updates). The total ticket count is displayed in the center of the donut.
What to look for:
- A high Activity ratio may indicate clients need more documentation or training
- A high Development ratio may indicate product maturity issues or active feature development
Status Distribution
| Chart type | Horizontal bar chart |
| Data window | Current snapshot |
| Categories | Assigned, Pending Assignment, On Hold, Requires Human, Pending Clarification, Not Started, In Progress, Completed, Blocked |
Shows how many non-spam tickets currently exist in each status. Bars are sorted by count (largest first).
What to look for:
- Large “Pending Assignment” count means tickets need owners
- Large “Requires Human” count means the AI couldn’t classify confidently
- “Blocked” tickets need attention to unblock progress
AI Classification
| Chart type | Bar chart with summary badges |
| Data window | All time |
| Buckets | 0-9, 10-19, 20-29, 30-39, 40-49, 50-59, 60-69, 70-79, 80-89, 90-100 |
Shows the distribution of AI confidence scores across all analyzed tickets. Below the chart, three summary badges show counts for the system’s classification thresholds:
| Badge | Threshold | Meaning |
|---|---|---|
| Auto-assigned (green) | >= 90% | AI was confident enough to create a JIRA ticket and auto-assign |
| Manual Review (amber) | 70-89% | JIRA ticket created but flagged for human review |
| Requires Human (red) | < 70% | No JIRA ticket created; human must review and classify |
What to look for:
- A large “Requires Human” count may indicate the AI needs more training data or the ticket types are inherently ambiguous
- A strong cluster in 90-100 indicates the AI is performing well
- Scores clustered in the middle ranges suggest borderline classification
Team Workload
| Chart type | Grouped horizontal bar chart |
| Data window | Current open tickets |
| Series | Support (orange/pink), Development (teal), Overdue (amber) |
Shows the number of tickets assigned to each active team member, broken down by assignment type (support contact vs developer) and overdue status. Members are sorted by total ticket count (highest first).
What to look for:
- Uneven distribution suggests workload rebalancing is needed
- High overdue counts for a specific member may indicate they need help
- Members with zero tickets may have capacity for reassignment
Team Composition
| Chart type | Two mini donut charts + stat row |
| Data window | Current snapshot |
Provides a structural view of your team:
- Roles donut — Breakdown by Admin, Lead, Member, and Developer-only
- Capabilities donut — Breakdown by Support Only, Dev Only, and Both (support + development capable)
- Status row — Active, Inactive, and Archived member counts
What to look for:
- Too few Leads may create a bottleneck in ticket management
- A high Archived count may indicate past team churn
- Members with “Both” capabilities provide the most staffing flexibility
Projects
| Chart type | Stacked bar chart |
| Data window | All time |
| Series | Activity (orange/pink), Development (teal) |
Shows the total ticket count per project, split by ticket type. The system Triage project is excluded.
What to look for:
- Projects with disproportionately high volume may need dedicated support staff
- The Activity/Development ratio per project reveals whether a project generates more questions or code work
Top Clients
| Chart type | Horizontal bar chart |
| Data window | All time (top 10) |
Shows the 10 clients with the most tickets, ranked by total volume. Tickets without a client (triage/unknown domain) are grouped as “Unknown / Triage.”
What to look for:
- The top client may deserve a dedicated support contact
- Clients with rapidly growing volume may need proactive outreach
- “Unknown / Triage” tickets indicate emails from unrecognized domains
Spam Detection
| Chart type | Stacked area chart + pattern list |
| Data window | Last 12 weeks |
| Series | Legitimate (teal), Spam (amber) |
Shows the volume of legitimate vs spam emails over time. Below the chart, a list of the top 5 spam patterns shows:
| Column | Description |
|---|---|
| Pattern | The email address, domain, subject, or keyword |
| Type | Badge indicating pattern type (email, domain, subject, body_keyword) |
| Count | Number of times this pattern has matched |
The card header also shows the overall spam rate as a percentage (e.g., “Total spam: 2 (0.4%)”).
What to look for:
- Rising spam volume may indicate a new spam campaign targeting your support addresses
- Recurring patterns in the top list suggest adding them to the spam filter
- A very low spam rate means the detection system is working effectively
Resolution Time
| Chart type | Line chart with data points |
| Data window | Last 12 weeks |
| Y-axis | Average days to resolve |
Shows the average number of days from ticket creation to resolution email, plotted weekly. Each data point represents the average for tickets resolved that week.
What to look for:
- An upward trend indicates resolution times are getting worse — investigate staffing or complexity
- A downward trend indicates improving efficiency
- Spikes in specific weeks may correlate with holidays, staff changes, or difficult tickets
- Weeks with no resolutions will show no data point (gap in the line)
Limitation
Resolution time is only calculated for tickets where a resolution email was sent (resolutionEmailSentAt). Tickets resolved through other means (e.g., direct JIRA closure without email) are not included in this metric.
Next Steps
- Dashboard Overview — Return to the analytics overview
- Spam Management — Configure spam detection and review patterns
- Team Workload — View the detailed workload table
- Ticket Types — Understand Activity vs Development classification