Genesys Cloud - Main

Hi all,

We have recently gone through a Genesys uplift and as part of this are redesigning our call stats.  Traditionally somebody copied the total calls from the queue performance page to a separate spreadsheet however now we are pulling a report of all interactions which is being fed into powerBI. 

The issue we are facing is the figures are slightly different.  On a queue that has roughly 350 calls, it is often out by ~10 calls.  I have tried pulling the raw data and applying all kinds of filters (inbound, outbound, abandoned, callbacks etc etc) to try and get the figures to match with no luck.  I know other genesys users have reported where interactions are started one day and finalised another that it can skew the figures however we have tried manually removing the 1 or 2 callbacks that were left overnight which still does not fix our numbers.  We do not have emails or messages etc on this queue it is purely inbound, outbound and callbacks.  

Does anyone have any idea why these two figures would be different or have guidance on what filters should be applied to make sure our numbers match in the genesys report to what is being exported to powerBI?  I know the business will not be happy if we cannot justify why the two figures are different (and its also breaking my brain and I want to understand it!).  

Thanks.

Hi Robert,

From my understanding, queue transfers, flow-outs (callbacks, voicemail, transfers, etc.), and metric timestamp attribution can sometimes result in differences between Queue Performance metrics and interaction exports. Queue Performance metrics are also interval-based, which may lead to differences when compared with conversation detail exports.

I noticed you've already accounted for the overnight callbacks, so I'd be interested to hear if anyone from the community has encountered something similar and how they reconciled the figures.

Hey Robert, if you could give more details and examples (Which metrics exactly are not matching, etc.), maybe someone will be able to see something that you are not seeing.

Also, have you tried investigating by just looking at a smaller sample? Looking at the whole day (350 interactions, assuming that's the number for the whole day) might be a little hard to see which interactions are not matching, etc. But if you try to check for just 1 hour, or maybe a 30 minutes interval, the difference might only be for 1 or 2 interactions, and then you can check 1 by 1 to try to identify which interaction is missing on the source that has a lower number, etc.

Edit: Also, something else to keep in mind.. Are both sources using the same timezone? Make sure that the data exported is using the same one you use on the Genesys interface, otherwise you could see some interactions being counted towards the wrong day without noticing, etc.

------------------------------
Marcello Jabur
------------------------------