- 06 Jun 2023
- 4 Minutes to read
Dialpad BI Datasets and Data Concepts
- Updated on 06 Jun 2023
- 4 Minutes to read
The beauty of Dialpad BI is that it provides additional fields and metrics than what's found elsewhere in your Dialpad Analytics. This allows you to really get granular with your data and analyze your company's health.
Let's talk about BI Datasets and concepts, what they are, and how to use them.
Datasets are the foundation of your BI reports and analyses and consist of records of your Dialpad data that are thematically similar.
BI offers 4 datasets:
- Call Records
- Call Records (Live)
- Agent Status
- Agent Status by Call center
BI reports can only be created from one dataset at a time.
Hover over a field or column name when viewing or customizing a report — you'll then see a description/definition of that field.
Don't forget to reference Dialpad's Analytics Glossary article for detailed definitions.
Call Records Dataset
The Call Records dataset is a complete call log detailing each call in the Dialpad system, and its corresponding metadata.
Call Records data is available after 4 hours and can be searched for as long as the data is stored.
There are 195 total metrics available, grouped as follows:
- CSAT Response
- Entry Point
- Master Call
- Notification Settings
The Call Records dataset is useful if you want to create/view a report that contains core calling metrics like the call volume, duration, CSAT scores, etc for a set of users or groups in your organization.
This dataset is similar to the information that is available in the Dialpad Call Logs Export, however, it includes many additional fields.
Call Records (Real-time) Dataset
This dataset is experimental, and may not yet be available for all accounts.
The Call Records (Real-time) dataset provides call records uniquely from the past 24-hour period, in real-time.
Data records are effectively up-to-date as of the present moment and uses the same metrics as the Call Records dataset, however, there are several known limitations:
- CSAT data is not available
- User/Group names may not immediately reflect changes made within Dialpad, and may display a "null" value for newly created groups (with a 4 hour lag until they are properly reflected)
Agent Status Records Dataset
The Agent Status Records dataset monitors each change in status an Agent makes throughout their shift across all call centers — including the corresponding metadata.
Only global statuses are available in this dataset (time the agent is available, on call, in wrap-up, unavailable, or on break). Local “Do not disturb” (DND) for individual call centers do not affect these metrics.
Agent Status metrics include:
- Agent Status ID
- End Time
- Office ID
- Operator ID
- Operator Name
- Setter ID
- Setter Kind
- Start Time
- Status Duration
- Total Durations
- Total Duration
This dataset is useful for productivity monitoring when you want to report on how much time an agent spent in each status.
Agent Status Records by Call Center Dataset
The Agent Status Records by Call Center data is almost identical to the Agent Status Records Dataset, however, it's broken down by Agent AND the individual call centers they are working in.
This dataset is useful for reporting on an agent's time while working simultaneously in multiple call centers as you can break down what activities are happening in each call center and independently.
Datasets Not Yet Available
The following datasets are not yet available within Dialpad BI.
- Dialpad Meetings
- Dialpad Messages & Channels
- Call Keywords, Moments, and Transcriptions
- Dialpad Digital Engagements Sessions
- Dialpad Self Service Interactions
Entry Point Calls
Entry Point calls are one of the most common data concepts when reporting on Call Records metrics.
Entry Point categorizations exist because what seems like a single "call" to an external caller can actually have multiple call records.
For example, if a call was made to a call center that rung out to 5 agents, (4 missed the call and one picked up), there will be a total of 6 records in the call records dataset to account for the initial call into the call center, and then for the 5 separate fanouts (aka ring outs) to the agents.
Similarly, if a call comes into a department and is routed to an operator who picks it up, then transfers this call to someone else, this will be recorded as 3 separate call records.
When you want to generate reports for a group like a department or a call center, you'll often want to have the "Is Entry Point Call" field filtered to "Yes", to include this call just once
If you are reporting on individual operators, you'd likely want each of their 5 calls legs to be separated and accounted for, at which point you may want "Is Entry Point Call" to be filtered for "is any value"
When in doubt if you should be including/excluding entry point calls, follow the guidance of one of the pre-made template reports which has a description similar to the type of reporting that you are hoping to do.
This data concept is currently an experimental set of fields. Additional documentation is coming soon
When should I use which dataset?
Use the Call Records datasets to report on call metrics and the Agent Status to report on agent availability.
Can I mix different data sets?
Not currently. However, we have a new feature called “Merge” that can help link together 2 reports created with 2 different data sets (see the next question for more details).
What does the “MERGE” feature do?
Merge allows you to tie together 2 different reports that have 2 different datasets (such as reports with Call Records and Agent Status). This means that you would start the 1st record using one dataset (e.g. Call Records), then click “merge” and run a 2nd report using a 2nd dataset (e.g. Agent Status).
There must be 1 dimension that is common across both datasets in order to apply the “merge” function in order for the 2 reports to display in a common report view.
Merged reports can not be scheduled, reused, or saved. It is not recommended to use in production but can be used on an Ad-hoc basis to check something.