2470809 - How are the totals calculated in the Customer Usage Data Report?

SAP Knowledge Base Article - Public

2470809 - How are the totals calculated in the Customer Usage Data Report?

Symptom

How are the total calculated in the Customer Usage Data Report?

Environment

SAP SuccessFactors Learning Management System (LMS)

Reproducing the Issue

  1. Login to Learning Admin
  2. Navigate to Reports (button in upper right)
  3. Search for Customer Usage Data
  4. Required filter: From Date
  5. Run the Report
  6. Note the totals:
    1. CustomerUsageData.png

Resolution

Here is a breakdown of each of the Calculations provided in the Customer Usage Data report:

Total Users

  • Sum of Total Active & Total Inactive Users (below)
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Total Active Users

  • Sum of users that have Active set to "Yes"
  • User record has been updated within date range specified in report filter:
    • The SQL adds a filter that checks that the Student ID has had some activity in the Student History table (ph_student), or Last Updated in the Student table (pa_student) falls within the specified date range

Total Inactive Users

  • Sum of users that have Active set to "No"
  • User record has been updated within date range specified in report filter:
    • The SQL adds a filter that checks that the Student ID has had some activity in the Student History table (ph_student), or Last Updated in the Student table (pa_student) falls within the specified date rang

Total User Logins

  • Sum of User Logins which are tracked in the pa_login_session_info table
  • The SQL filters on the login time falling between dates specified in report filter

Total Unique User Logins

  • Sum of distinct/unique logins from the pa_login_session_info table
  • If a date time frame is specified, it will filter on the login time falling between those date
  • If the User logged in multiple times within the time frame specified in the report filter, then they will only be counted once
  • For example:
    • UserA logged in 5 times between May 1st and May 6th
    • For Unique Logins, UserA will count as 1
    • For Total Logins, UserA will count as 5

Average Daily User Logins

  • Average number of logins based on the sum of Total User Logins divided by the total number of days in the time frame specified in the report filter:
    • [Total User Logins]/[Day Count]
    • Where Day Count = total number of days within the date range selected in the report filter

Average Unique Daily User Logins

  • Average based on Total Unique User Logins divided by total number of days determined by the dates specified in the report filter
    • [Total Unique User Logins]/[Day Count]
    • Where Day Count = total number of days within the date range selected in the report filter

Average Concurrent Users

  • Queries the pa_login_session_info table and checks User Login Times
  • Counts the number of times there were concurrent logins (Users logged in at the same time)
  • Sums the total number of concurrent logins that occurred
  • Averages that number by taking the sum of concurrent logins and divides by the count of total concurrent logins
    • [Sum of Concurrent User Logins]/[Total Count of Concurrent User Logins]
  • Example:
    • 5 Users logged in April 1st at noon
    • 3 Users logged in April 1st at 3pm
    • 8 would be the sum of those counts
    • 2 would be the number of times there were concurrent logins
    • The Average would be 8/2 = 4

Max Concurrent Users

  • Queries the pa_login_session_info
  • Finds the times there were concurrent logins (Users logged in at the same time)
  • Performs a max function on those times to show the highest number of concurrent logins
  • Example:
    • 5 Users logged in April 1st at noon
    • 3 Users logged in April 1st at 3pm
    • Max concurrently logins would be 5

Total Admin Logins

  • Sum of Admin Logins which are tracked in the pa_login_session_info table
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Total Unique Admin Logins

  • Sum of distinct/unique logins from the pa_login_session_info table
  • If a date time frame is specified, it will filter on the login time falling between those date
  • If the Admin logged in multiple times within the time frame specified in the report filter, then they will only be counted once
  • For example:
    • Admin1 logged in 5 times between May 1st and May 6th
    • For Unique Logins, Admin1 will count as 1
    • For Total Logins, Admin1 will count as 5
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Average Daily Admin Logins

  • Average number of logins based on the sum of Total Admin Logins divided by the total number of days in the time frame specified in the report filter:
    • [Total Admin Logins]/[Day Count]
    • Where Day Count = total number of days within the date range selected in the report filter
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Average Unique Daily Admin Logins

  • Average based on Total Unique Admin Logins divided by total number of days determined by the dates specified in the report filter
    • [Total Unique Admin Logins]/[Day Count]
    • Where Day Count = total number of days within the date range selected in the report filter
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Average Concurrent Admins

  • Queries the pa_login_session_info table and checks Admin Login Times
  • Counts the number of times there were concurrent logins (Admins logged in at the same time)
  • Sums the total number of concurrent logins that occurred
  • Averages that number by taking the sum of concurrent logins and divides by the count of total concurrent logins
    • [Sum of Concurrent Admin Logins]/[Total Count of Concurrent Admin Logins]
  • Example:
    • 5 Admins logged in April 1st at noon
    • 3 Admins logged in April 1st at 3pm
    • 8 would be the sum of those counts
    • 2 would be the number of times there were concurrent logins
    • The Average would be 8/2 = 4
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Max Concurrent Admins

  • Queries the pa_login_session_info
  • Finds the times there were concurrent logins (Admins logged in at the same time)
  • Performs a max function on those times to show the highest number of concurrent logins
  • Example:
    • 5 Admins logged in April 1st at noon
    • 3 Admins logged in April 1st at 3pm
    • Max concurrently logins would be 5
  • Query determines if the login is admin or user by checking login_type which shows 'A' for admin and 'U' for user

Total Item Completions

  • Queries the pa_cpnt_evthist (Learning Event History table)
  • Determines Total Items to be internal + external completions
  • Filters that the completion date is within the time frame specified in the report filter

Average Item Completions

  • Takes the Total Item Completions and divides it by the Total number of Users
    • [Total Item Completions]/[Total Users]

See Also

This WIKI Page covers the breakdown of the SQL used for each calculation.  If you are familiar with SQL, you can analyze the queries to get a better understanding of how the calculations are performed:

https://wiki.scn.sap.com/wiki/x/koERG

Keywords

SF SuccessFactors LMS Learning Management System Report CustomerUsageData Customer Usage Data total calculate sum average login RPT REP BIRT PRD , KBA , LOD-SF-LMS-REP , Reporting Data , How To

Product

SAP SuccessFactors Learning all versions