When creating a new Real-Time dashboard, you may want to group your metrics into custom buckets. Brightmetrics supports two grouping types: Column Grouping and Count Grouping. This article covers when to use each and how to set them up.
Why Use Groupings?
A common reporting challenge in contact centers is that queue-based charts can double-count agents. Because an agent may be a member of multiple queues, queue data receives a count for each queue individually and adds them together. Agent-based data, by contrast, deduplicates — each agent is counted only once, regardless of how many queues they belong to.
You filter a queue chart to 7 queues, and 6 agents are members of all 7. The queue chart may show 42 agents On-Queue (7 × 6 = 42), while an agent-based chart correctly shows 6. This is expected behavior, not a bug. Count Groupings solves this by building counts from agent-based (Realtime Agents) data, which deduplicates automatically.
Choosing the Right Grouping Type
| Column Grouping | Count Grouping | |
|---|---|---|
| What it shows | Individual agents organized into named groups by status | A single numeric count per group (e.g., total agents in each state) |
| Best for | Seeing which agents are in each state | A wallboard count: 12 Available, 4 Unavailable, 6 Interacting |
| Avoids queue double-counting? | Yes — uses agent data | Yes — uses agent data |
| Data source | Realtime Agents | Realtime Agents |
| Typical use case | Manager reviewing a list of agents grouped by Available / Unavailable | Real-time wallboard showing accurate headcounts by state |
Setting Up a Grouping Chart
Both grouping types follow the same initial setup. The steps below walk you through the full process, with notes where Column and Count Grouping differ.
-
1
Start a New Real-Time Chart: While in edit mode on your dashboard, click Add Chart.
-
2
Choose Your Data Source: Select a Real-Time data source from the list. -
3
Select the Data to Display: For grouping charts, select Realtime Agents as your data set. This is important — using queue data here will result in duplicate agent counts if agents belong to multiple queues.
-
4
Choose Your Grouping Type: In the Chart Layout dropdown, select either Column Grouping or Count Grouping.
-
5
Set Up Grouping Criteria: Under the Grouped By section, select the attribute to group your data by. For this example, we'll select Primary Status.
-
6
Add Groupings:
Click Add to open the grouping setup, then click Add Group to create your first group.
Define each group by clicking the blue link to select values.
Select the statuses that belong in this group, then click Apply. In this example, we select statuses indicating that an agent is unavailable or off the phone.
In the box to the right of the grouping, name your group (e.g., "Unavailable"), then click Okay.
Repeat these steps to create additional groupings (e.g., an "Available" group).
-
7
Finalize Your Groupings: Once you've defined all groupings, click Okay. -
8
Choose Chart Values: In the Chart Value(s) section, select the metrics you want to display. For this example, we'll select Primary Status.
-
9
Apply Filters: Click Filters at the bottom to specify which data to include in the chart. Once applied, your chart will display your selected groupings.
Example Views
Column Grouping View
Displays individual agents organized into named columns by their grouped status. Use this to see who is in each state. For example, a supervisor needs to identify specific agents who are unavailable.
Count Grouping View
Displays a single numeric count for each group rather than listing individual agents. Use this when you want a high-level, at-a-glance summary. For example, a wallboard showing how many agents are currently available, interacting, or unavailable.
Note on Count Grouping and queue data: Count Grouping is particularly valuable as a replacement for queue-based agent counts. If you've noticed your queue chart showing a much higher agent count than expected (e.g., 42 instead of 6), this is likely because agents are members of multiple queues and are being counted once per queue. Switching to a Count Grouping chart using Realtime Agents data will give you an accurate, deduplicated count.
Additional Customization
You can reorder the columns or counts by dragging them up or down in the grouping setup menu. This lets you prioritize or organize the display as you prefer.
By following these steps, you can easily create meaningful groupings for your Real-Time dashboard to better visualize and analyze your data. If you have any questions, feel free to reach out to support@brightmetrics.com.