You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We are looking for input on the Zendesk agent replies metric. Currently we believe that our model accounts for agent replies in most situations (see our latest PR).
But we did find discrepancies between the Zendesk Explore numbers and ticket reply numbers on our data occasionally. Here were two edge cases we encountered:
If an agent creates a ticket with an internal comment, and then replies to that ticket, is that internal comment the ticket being created and the external reply logged as a legitimate reply? We were seeing an undercount of replies by 1 in our own modelling compared to the number of replies in a ticket in the Zendesk Explore.
How replies are logged to tickets that are follow up to tickets. Does a new ticket log a reply from an old ticket as a reply, or the first comment created? We were seeing an overcount of replies by 1 in our own modelling in the Zendesk explore, because our model can bring in those agent comments from merged tickets.
Have you encountered these issues as well? If you have, or have seen any differences between the Zendesk replies on your internal data vs our own, please let us know in the comments!
The text was updated successfully, but these errors were encountered:
also if an employee leaves the company and their role might in zenedsk. we specifically look at the role (agent or admin) for calculating reply time metrics, but these former employees might now be end-user. we may want to introduce a variable where people can input the company's email domain or some other criteria for determining former employees
also if an employee leaves the company and their role might in zenedsk. we specifically look at the role (agent or admin) for calculating reply time metrics, but these former employees might now be end-user. we may want to introduce a variable where people can input the company's email domain or some other criteria for determining former employees
@jackiexsun@fivetran-jamie There appears to be an audit log for tracking changes, perhaps this is where we can find information on whether users used to be admin?
We are looking for input on the Zendesk agent replies metric. Currently we believe that our model accounts for agent replies in most situations (see our latest PR).
But we did find discrepancies between the Zendesk Explore numbers and ticket reply numbers on our data occasionally. Here were two edge cases we encountered:
If an agent creates a ticket with an internal comment, and then replies to that ticket, is that internal comment the ticket being created and the external reply logged as a legitimate reply? We were seeing an undercount of replies by 1 in our own modelling compared to the number of replies in a ticket in the Zendesk Explore.
How replies are logged to tickets that are follow up to tickets. Does a new ticket log a reply from an old ticket as a reply, or the first comment created? We were seeing an overcount of replies by 1 in our own modelling in the Zendesk explore, because our model can bring in those agent comments from merged tickets.
Have you encountered these issues as well? If you have, or have seen any differences between the Zendesk replies on your internal data vs our own, please let us know in the comments!
The text was updated successfully, but these errors were encountered: