Using the Insights dashboard
To be HIPAA compliant while using Insights, personal participant information needs to be removed. To do so, navigate to your Configuration settings, choose the Features tab, and scroll down to Privacy Features to enable the privacy toggle.
Rooms and sessions
On the Rooms page, you can see a list of all the rooms that currently exist and have been created in your account. They are listed by most to least recently used. You can easily filter the list by searching a room name.
You can also access details of all the sessions that have happened within a specific room by clicking on the room name. A session starts once there are two or more people in a call.
Participants
Within a session, you can find the Session details. The default view is an Overview of the participants that attended the session. A single timeline is created for each participant. Any time your user reconnects to the room, the reconnection displays as a new "segment" within the participant's timeline.
Display name
The display name of the participant is the name they enter the call with. Your users can either enter this name themselves, or you can pass in for them using the displayName parameter.
Some customers choose to use generic names, such as Doctor or Patient for added anonymity. If you would like for names not to appear in your Insights, you can configure this in your account settings by going to "Configure" > "Features", then scrolling down to the Privacy features section and toggling on 'Remove participant display names'.
Display names will then appear as Participant 1, Participant 2, etc. Participants who leave their name blank when joining a session will appear as Guest 1, Guest 2, etc.
Participant Call Ratings
Whereby has a call ratings feature in beta that you can ask us to turn on for your account.
If you have this feature turned on, each participant will be asked at the end of their call to rate the call quality, and you can track these ratings in a dashboard will will provide to you. Contact Whereby to get this feature enabled.
Browser and operating system
You can use this information to determine if there were any compatibility issues, since not all devices or browsers fully support WebRTC.
Check out our list of supported browsers and devices
Reading the charts
All dates and times in the Insights UI and API are in UTC.
We capture values for each chart every 2 seconds. We believe this frequency helps capture the spontaneous nature of call issues.
Activity
The Activity view provides an events timeline of the participants' camera and microphone. You can use this information to diagnose video or audio issues that your users report.
Connection
The Connection view provides an overview of the data being transferred over the course of the session, as well as the connection quality.
Sending & receiving
We've separated packet loss and bitrate out into separate charts for sending and receiving. Each participant is both sending data in the form of audio and video to others on the call and receiving audio and video data in return. Further details can be seen when clicking the toggle next to Sending or Receiving.
Packet loss
Packet loss is a great indicator if a user had poor network. It is a measure of how many data packets sent over a network are lost in transit. Packets are small chunks of data that are used to transmit information over a network.
Packet loss can have a significant impact on the quality of video calls. When packets are lost, the video can become choppy or pixelated. In some cases, the video call may even drop altogether.
Packet loss can occur for a variety of reasons, including:
Network congestion: When there is too much traffic on a network, it can cause packets to be dropped.
Hardware failure: A faulty network device, such as a router or switch, can cause packets to be lost.
Interference: Electromagnetic interference from other devices, such as microwaves or cell phones, can also cause packet loss.
How to interpret packet loss values
The amount of packet loss is measured in percentage. For example, 1% packet loss means that 1 out of every 100 packets were lost. 0% packet loss is ideal for video calls, but some amount of packet loss between 0% - 2.5% is tolerable and expected.
We consider 3% packet loss to be our threshold for quality. Any value at or exceeding 3% for a sustained period of time means that the participant likely experienced reduced call quality. Very short periods (2 to 4s) of high packet loss are normal and expected. They will show up on the charts as sharp spikes. It is unlikely the participant noticed a dip in video quality during this short period.
Bitrate
Bitrate is a measure of the amount of data that is transmitted over a network per unit of time. In the context of video calls, bitrate refers to the amount of data that is used to transmit audio and video.
In general, a higher bitrate will result in a better quality video call. However, a higher bitrate also requires more bandwidth. If your user's internet connection does not have enough bandwidth, they may experience choppy or pixelated video.
How to interpret bitrate values
We measure bitrate in bits per second (bps), and a "good" value can vary depending on whether participants in the call had turned on HD video, if the call was in normal or group mode, how many participants joined the call, and whether someone was screensharing.
Generally, we expect to see at least 500kbps sending bitrate if the participant has their video turned on. If the bitrate is below this value and their video was on, they may have sent blurry or pixelated video, choppy audio, or dropped frames. Our first suspicion would be that the participant has a slow internet connection or a poor quality webcam.
If the participant is using HD video, we expect to see 1 - 1.5Mbps send bitrate.
Group and normal rooms
We calculate the sum of all of the streams for bitrate sending, and we display the max packet loss for any stream in packet loss sending. This is important to be aware of if you are using "roomMode": "normal" rooms.
If you use normal rooms, each participant on the call is sending a video stream to every other participant. This means if you have 4 participants in a normal room, each participant is sending 3 video streams. Then the bitrate sending chart will be calculated as the sum of those 3 streams and packet loss sending chart will display the stream with the maximum packet loss during the call.
Every room created with "roomMode": "group" is using our selective forwarding unit (SFU) mesh for data transfer. This means that every participant is sending only one single video stream to our SFU which then forwards the stream on to every other participant. This is one of many reasons why we recommend using "roomMode": "group" for better call quality.
Known limitations
When sessions exceed 30 participants or 3 hours duration, the system will not fetch participant activity and connection data. In these cases, support teams will see an explicit message directing them to retrieve the full data via API instead. This approach prevents performance issues that would arise from loading all participant details in very large sessions.
Running into issues?
If you are having difficulty troubleshooting or have questions about the content of the insights data provided, please feel free to reach out to our Support team at [email protected]. Diagnosing call quality issues is not the most straightforward endeavor, and we are happy to get your feedback so we can improve these features.
We also have some information and recommendations in our document about improving call quality.
Last updated
Was this helpful?