andy.carter | 2019-10-08 10:34:31 UTC | #1
Following the definitions from here (https://help.mypurecloud.com/articles/metric-definitions/) I was expecting to find the time spent in the 'Interacting' routing status to be the sum of the tHandle time from the conversation aggregate stats. This is just for voice agents initially.
However, for any given day there seems to be a discrepancy of about 10 - 20s between the two numbers, with INTERACTING status always working out higher. I can't see any one category that simply accounts for the 'missing' time. (see attached image).
EDIT:
Using the user details analytics end point, I've pulled the start and end timestamps for each routing status change to 'INTERACTING' (left) and compared to the app total handle times for each interaction. I think there are some small differences that add up to to the total discrepancy This is probably getting a bit too pedantic, but can anyone explain the difference?
Is it possible to reconcile the two numbers? I would like to account for the time spent 'not interacting' into hold & ACW, but until I can get back to the 'interacting' stat from individual conversations that's not possible.
tim.smith | 2019-10-08 14:36:24 UTC | #2
The time a user was in a particular routing status and the time an agent was a participant on an interaction are measuring two separate things, each with their own unique timestamps marking their unique events (timestamps when the user's presence changed vs. timestamps of a participant in the conversation). I'd expect them to be similar since they are closely related, but because they are measuring two different sets of data there's not a direct connection between the two that would imply they should be identical.
Customer data cannot be investigated via the forum. If you need a more in-depth investigation into the specific measurements, please open a case with PureCloud Care to investigate.
andy.carter | 2019-10-08 15:48:15 UTC | #3
Thanks Tim, I wasn't sure if both statistics were being derived from the same basis timestamps. If they are from different timestamps that would explain any discrepancy. I'm happy that I'm close enough for it to make no appreciable difference to the analysis.
system | 2019-11-07 15:48:17 UTC | #4
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
This post was migrated from the old Developer Forum.
ref: 6192