Journey Analytics
Airkit's out of the box reporting and analytics features provide a way to gather data on user interactions and gain insights on how users are utilizing an application. When leveraging reporting metrics, you can track any type of user interaction, including button clicks using the counter metric or specific statistics such as CSAT scores using the statistic metric. There are also automatic metrics that are tracked and stored on a per journey basis and can be surfaced through a data export. The metrics and journey data can then be pushed to other reporting tools such as Power BI, SnowFlake, and more for further reporting.
Metric Types
There are two different types of metrics that can be recorded within a journey -- automatic metrics and custom metrics. These metrics can be used in conjunction to build out reports and tie together data for more wholistic insights.
Automatic Metrics
Automatic metrics are built in metrics that are captured per journey and can be surfaced through an export in the reporting console. The automatic metrics include data such as SessionId, ActionRuns, UserAgent.DeviceClass and much more. For example, this data is great for creating reports to surface how many total journeys the application has had per deployment, or insights into what type of device the application is accessed on.
Custom Metrics
Custom metrics are used to gather data on how an application is being used or any other data that needs to be captured in a report. For example, custom metrics can be used to capture a CSAT score per session or can be used to see how long a user is on a specific web page for. The custom metrics are surfaced in the export along with the automatic metrics. The following custom metrics actions are available to use within an application:
Exporting
In the reporting console, there is the ability to export all of the metrics to a .csv to use with other tools. To export, go to an application at console.airkit.com > Click on Reports.
Then click on the Export button on the top right hand corner. This will generate an export link that will download a .CSV file with the app metric data.
Leveraging Analytics in your Application
Both custom and automatic metrics provide a way to create data driven decisions on how to improve the application workflow or get visibility into KPIs that are trackable throughout a journey. Below are some examples on how to leverage Journey Analytics in Airkit:
Capturing a CSAT score using the statistic metric and tying it back to a user session.
Track how long it takes for a user to get through a journey or a specific web flow using the start timer metric and stop timer metric.
Get the average amount of interactions a user has with a chat bot in a conversation using the count metric.
Reference: Automatic Metric Properties
SessionId
OrganizationId
AppId
BranchId
DeployId
CxrVersion
SessionTime
SessionEndTime
LatestEventTime
SessionCount
MessagesSent
MessagesReceived
MessagesDelivered
MessageDeliveryFailures
VoiceSent
VoiceReceived
VoiceDuration
CanvasLinkOpen
SocketConnections
SocketDisconnections
CallCompletions
CallDuration
BridgeCallCompletions
BridgeCallDuration
TotalSessionDuration
EventsOccurred
ErrorsThrown
ConnectionsRun
TriggersRegistered
TriggersExpired
TriggersInvoked
ActionErrors
ActionRuns
RuntimeErrors
MinScreenSizeWidth
MinScreenSizeHeight
MaxScreenSizeWidth
MaxScreenSizeHeight
HttpSource
IsDebug
UserAgent.DeviceClass
UserAgent.DeviceBrand
UserAgent.DeviceName
UserAgent.OperatingSystemClass
UserAgent.OperatingSystemName
UserAgent.OperatingSystemVersion
UserAgent.OperatingSystemVersionMajor
UserAgent.OperatingSystemNameVersion
UserAgent.OperatingSystemNameVersionMajor
UserAgent.LayoutEngineClass
UserAgent.LayoutEngineName
UserAgent.LayoutEngineVersion
UserAgent.LayoutEngineVersionMajor
UserAgent.AgentClass
UserAgent.AgentName
UserAgent.AgentVersion
UserAgent.AgentVersionMajor
UserAgent.AgentNameVersion
UserAgent.AgentNameVersionMajor
Last updated