Global Dashboard: read service results
Use Global Dashboard to locate periods and queues that deserve investigation. Its historical indicators summarize selected data; they do not replace live supervision or the detailed call reports.
Define the comparison
Select the date interval and queues before reading totals. Confirm the displayed period and the basis of any previous-period comparison. Use the busiest-hour indicator, hourly volume/answer-rate chart and day/hour heatmap to locate concentrations of demand. Keep the same selection when comparing charts and the queue table.
Use the correct denominators
Here Answer rate is answered ÷ (answered + missed) × 100, without a response-time threshold. Unanswered agent ringing attempts are separate from this denominator. Global AHT averages waiting plus conversation time for answered calls; ringing is already included in waiting. For speed of answer, inspect the dedicated waiting/response measures instead of substituting this answer rate.
Connect charts to detailed evidence
Compare agent call volume with conversation time, then inspect waiting/ringing/talk trends, duration bands and weekly volume. Repeat-call rate is a signal about numbers returning, not an automatic First Call Resolution score. Read missed-call exit reasons separately from agent no-answer events, because one incoming call can generate several ringing attempts.
Choose the panel that answers the question
Set dates and queues first. Read the i next to a panel before comparing its count or duration with another report.
| Indicator or control | How to read and use it |
|---|---|
| Totals and volume trend | Relate incoming volume, answered episodes, missed exits and agent ring attempts. These are different units; attempts are not additional incoming calls. |
| Answer speed, patience and positions | Use waiting-time and position distributions to locate delayed answers or exits; a total alone does not show how long customers waited. |
| Repeat calls and causes | Read repeat-number signals alongside missed-call and ring-attempt causes. Continue in Contacts and Routing to verify distinct calls and their paths. |
| Agents and quality of time | Compare agent call volume with average talk, then waiting/ringing/talk trends. A high-volume, short-talk point is a reason to inspect records, not a performance verdict. |
| Talk duration and weekly load | Use duration buckets, weekday volume and the day/hour heatmap to identify the periods that need a detailed report or coverage scenario. |
Make a reproducible service comparison
Fix the scope
Select the same queues and comparable dates. Record any special filters before reading changes against the previous period.
Separate the two time questions
For responsiveness use answer-speed measures. For total handled time read global AHT as wait plus conversation. For staffing simulation use conversation plus after-call work instead; copying global AHT into the simulator would count customer waiting as service work.
Open the detailed evidence
Use the corresponding Answered, Missed or Ring no answer report with the same period and queues. Validate records and events before changing staffing or interpreting an agent’s results.
Example: more calls, shorter conversations
If one agent handles more calls with shorter conversations, open Answered calls for the same queue and period, then inspect a sample in Recordings. Check call type and outcomes before interpreting the difference as poor quality. Use Quality & AI only for available analyzed recordings. For staffing simulation, use conversation plus after-call work, not Global AHT.
Use the controls, understand what changes
Move a KPI card directly. Use a chart heading to reorder the chart groups; inside supported groups you can also reorder the individual charts. The peak-hour and queue-ranking insight cards have their own saved order. These are separate layout areas, not unrestricted movement between every panel.
The information icon explains the metric. Hovering over a plotted point instead shows the value at that point. For the agent comparison, read both call count and average talk time; the point alone is not an assessment of conversation quality.
Try dragging from the dotted icon beside each name, or use Move earlier / Move later. This example only changes the cards below.
- Answered calls
- Dashboard SLA
- Average handling time
- Distribution and volume
- Response speed, patience and exit position
- Day × hour heatmap
The product’s information button opens on hover or focus. A click keeps the help visible; click again, click outside or press Escape to close it. The examples below reproduce selected product help text; they are teaching controls, not screenshots.
Percentage answered out of answered + missed queue calls. Formula: Answered / (Answered + Missed) × 100. Missed includes caller abandonment, maximum waiting time, no reachable agents and exit by key press, even if the call continues elsewhere. Individual agents’ unanswered attempts are excluded. No time threshold is applied.
Average time from queue entry to the end of the conversation, for answered calls. Formula: (Total wait time + Total talk time) / Answered calls. Ring time is already included in wait time and is not added again.
Complete product screenshots
Illustrative data. Open the screenshot to enlarge it and match the controls to the instructions in this chapter.
Open complete screenshot
Read the workspace section by section
Match the numbers and panels below to the complete workspace. These screenshots use illustrative data; the procedures above explain how to work with your own installation.
Service indicators and trends
Read answered and missed calls, agent attempts and service times with their trend against the previous period.
Outcomes, volume and service level
Connect call outcomes, hourly volume, agent no-answer events and the answer ratio on the complete daily timeline.
Peak hour and queue performance
Identify the busiest hour and compare the three queues with the highest and lowest answer ratios.
See where waiting turns into a missed opportunity.
Answer speed
See how answered calls are distributed across waiting-time bands: up to 15 seconds, 15–30, 30–60 and over 60. Identify whether most callers receive a quick response or a longer wait is becoming common.
Waiting before a missed call
Read how long missed calls waited before leaving the queue: up to 5 seconds, 5–30, 30–60 and over 60. Combine this distribution with the exit reasons to decide where coverage or routing deserves a closer look.
Exit position
The hourly line shows the average queue position at exit for missed calls. Repeated exits near the front of the queue are worth reading alongside waiting times, staffing and routing rules.
Understand what brings the caller back.
Repeat calls
The rate counts calls beyond the first for each identifiable number in the selected period and queues: (calls − unique numbers) ÷ calls × 100. A rise is a useful signal for investigating repeated contact.
Reasons for missed calls
Distinguish caller abandonment, maximum waiting time, exit because no agents are reachable and exit by key press. Each outcome points to a different operational question.
Agent no-answer events
Inspect the recorded causes, such as an agent not answering or the caller cancelling during ringing. These are attempts towards agents: several attempts can belong to the same incoming call.
Turn repeat contacts into better service.
First Call Resolution asks whether the customer’s need was solved in the first call. The repeat-call rate is a starting signal, not an automatic FCR score: the same number may call about a new issue, on behalf of another person or for an agreed follow-up. Read the contact history and the content of the conversation before drawing a conclusion.
Add context. Choose the next step.
Use Contacts & Journeys to review the sequence of interactions. Explore Action Center to organise available follow-ups and their outcomes; combine these views with qualitative call analysis when available.
Agents and service times
You may have agents who take many calls with short conversation times and others who take few calls with long conversation times. Quality and quantity help you identify anomalies in how calls are handled.
Match the time calls need with the people available.
Conversation duration
The duration bands show how answered calls are distributed: up to 1 minute, 1–3, 3–5 and over 5 minutes. Compare the mix with the type of service: a quick-information queue and a technical-support queue can have very different, equally useful conversations.
Weekly volumes
Read answered plus missed calls grouped by day of the week across the selected period. Combine this demand pattern with conversation duration and the day/hour heatmap to review shifts and coverage at the busiest times.
Workload across the week
Find the combinations of weekday and hour with the most calls.
One table. The detail behind every result.
Compare total, answered and missed calls, agent no-answer events, average and maximum waiting, average total time, maximum conversation duration and SLA. Sort the columns to find the queues you want to review first.
Open the calls behind the number
In Obelisk, the Answered, Missed and Agent no-answer counts open their detailed report with the queue and dates already selected. Move from a queue-level change to the records that explain it.
Bring the review to your team
Choose the number of rows to show, read the totals for the visible queues and export CSV or print the current view. Use a shared period and queue selection to make operational comparisons meaningful.