Plan requirement
| Subscription | Suite Professional or higher, Explore Professional or higher |
| Access | Agent |
Metrics are what you count, attributes are how you group. And a metric's name rarely tells you the whole of what it counts.
The distinction
A metric produces a number: Tickets, Solved tickets, average first reply time. An attribute produces a category: agent, month, channel, priority. Numbers go in the middle of the report; categories go along the edges.
The test is whether you can add it up. Tickets can be added; agents cannot. If adding it up is meaningless, it is an attribute.
Read what the metric counts
Every metric in the Tickets dataset has a published definition, and several of them are narrower than their name suggests.
| Tickets | The total number of tickets. Deleted tickets are not in the dataset at all |
| Solved tickets | Solved or closed, so it keeps counting after a ticket closes |
| Reopened tickets | Reopened after being solved, but not one solved and reopened within the same update |
| Unreplied tickets | Tickets with no agent response at all |
| Reassigned tickets | Tickets that have been assigned to more than one agent |
| Agent submitted tickets | Submitted by an agent or an admin, which is not the same as "internal" |
Those definitions are the difference between two people quoting the same metric and meaning different things. Read the one you are about to put in front of a manager.
Where it gets confusing
Priority looks numeric and is a category. Ticket ID is a number and is an identifier. Neither should be summed, and both are offered in ways that let you try.
Custom fields appear as both
A custom field arrives as an attribute you group by, and a numeric one can also serve as a metric. Check which you picked, because the report builds either way and means different things.
One metric, one or two attributes
"Solved tickets" by "agent" and by "month" is readable. Adding channel and priority produces a table with hundreds of cells that answers nothing.
Time attributes have levels
Day, week, month, quarter, year. The wrong one is the difference between a chart showing a trend and one showing noise: daily volume over a year is unreadable, monthly is clear.
If a metric is missing
It is either in another dataset or it is a calculation you need to define. Check the dataset first, because the second is the more expensive of the two, and the published formula for a standard metric is often the best starting point for your own.
Comments
0 comments
Article is closed for comments.