Here we aim to make sense of the terms used throughout Analytics and answer any and all questions that you may have. If you have a question that is not covered in this section, feel free to reach out to us and we'll get it updated. 


TABLE OF CONTENTS


Active Sessions

Active Sessions have a 60-second filter applied in an attempt to provide insight into valid and engaged listeners. This is especially important for customers who monetise their online audience as this is the metric advertisers are interested in and rate 'valid audience' at. 


Sessions Started 

Sessions Started have a 1-second filter applied to minimise the data-skewing effects of snooper bots or traffic manipulation. This means that a listener has to remain connected to the stream for at least 1 second before they will be counted as a valid session start.


Total Listening Hours

Total Listening Hours is the total time listeners remained connected and is based on sessions of at least 60 seconds in duration (Active Sessions).


Time Spent Listening

Time Spent Listening is the total time listeners remained connected for at least 1 second (Sessions Started).


Bounce Rate

The Bounce Rate refers to the percentage of listeners not converted from Session Started to Active Sessions. Said more plainly, the Bounce Rate represents listeners who listened for less than 60 seconds but more than 1 second. A high Bounce Rate can be due to a number of causes, including:

  • Listener interest;
  • Website accessibility and reliability;
  • A listener's internet connectivity and stability;
  • Stream availability, etc. 


What percentage of Bounce Rate is considered to be a 'good' Bounce Rate is subjective to the person viewing the data. There are certain scenarios where a high Bounce Rate might be expected and others where it is not. 


Simple actions you can take to reduce your Bounce Rate percentage include:

Reviewing the most popular platforms used by your listeners to access your stream to ensure the listener experience is smooth and without unexpected errors; and
Ensuring you are using the correct load-balanced listen-links provided by the Support Team to distribute your stream.



Session Peak

Session Peak refers to the maximum number of listeners connected at the same time. When reviewing this data over a longer time period, for example, a week, the Session Peak will refer to a single Session Peak observed over that week. No filters are applied to Session Peaks. 


Note: Groupings of stations will reflect the highest Session Peak in the group (Session Peak will never be accumulated). 


Average Time Spent Listening

The Average Time Spent Listening is based on the Total Listening Hours and the number of Active Sessions observed. 


Unique Users

Unique Users refer to listeners accessing the stream from a unique IP address and user agent combination. In this context, the user agent refers to how the platform used for streaming is identified in the system and in the logs alongside the device used to access the platform. 


For example, 5 people in the office listening on 5 different platforms would count as 5 unique users. 10 people listening on the same platform (for example a smart speaker) will count as a single unique user.


Considerations when reviewing unique users

A listener who remains connected for an extended period on the same IP address and user agent combination will still only count as one unique user. This can become a little murky when changing reporting periods as the example below illustrates. 


John starts listening on his mobile phone on Monday and remains connected until Sunday. 

Reporting Period: From Monday to Sunday
    Unique User = 1

Reporting Period: Monday only
    Unique User = 1

Reporting Period: Tuesday only
    Unique User = 1

Reporting Period: Wednesday only
    Unique User = 1

Reporting Period: Thursday only
    Unique User = 1


Note that John is a unique user on each individual time period being considered - this is intended to illustrate that you cannot summarise unique users over limited time periods and expect the summary to reflect unique users over an extended time period as the user would be counted only once during the extended time period while that same user may be counted multiple times in the collection of shorter time periods.