I am trying to create a metric that shows me a list of my agents/admins, and the number of late first replies for each agent (in a bar graph so I can see who is not meeting our custom SLA). I have defined a late first reply as one that is greater than 1 business day from when the ticket was created (480 minutes).
So far I have a Report in GoodData where "What" is equal to:
SELECT ( SELECT # Tickets WHERE (SELECT First reply time in minutes within business hours By Ticket Id) > 480 ) / # Tickets
This is giving me a percentage of late first replies as a function of the total number of tickets.
This is then sorted under "How" By Ticket Assignee.
Lastly I added a Filter to filter by only those tickets that have a reply.
All of this is great, but I see 2 flaws. First is that I am using the Ticket Assignee to determine the if he/she "owns" the late reply. What if Agent 1 makes a first reply late (say 2 business days from ticket creation), then the ticket gets assigned to Agent 2, and he/she closes it? Doesn't this result in the late reply appearing as though Agent 2 did it since the Ticket Assignee is currently Agent 2?
The second flaw is the percentage is a percentage of the total number of tickets. I would prefer that this is a percentage of each agents own tickets, as some agents deal with a lot of tickets, and some deal with very few.
Any input would be appreciated. Thanks all!