Using result metric calculations



image avatar

Rob Stack

Zendesk Documentation Team

Edited Mar 05, 2025


-5

18

18 comments

Is it possible to divide a value obtain from metrics with an int value? For example, I want to make the formula COUNT_D(Update_ID)/40, since I want an average of how many updates per working hour the agents did. But it is not working.

0


is it possible to use the IF formula in result metric calculations ?

0


Hi 1900214814004 !

I went a head & created a ticket on your behalf to look into your Explore data and see why it's not loading anything on your end. Please keep an eye out for our Email.

0


Im adding some calculated metrics but it results  into nothing, 
(COUNT(AT-RD)+COUNT(AM-RI)+COUNT(AM-RD)+COUNT(FMD-CMA))
maybe because some of these metrics has 0 as a value? how can it be computed? 

0


Hi Jack,
 
Have you already tried to refresh your browser and recreate the Result Metric Calculation? Normally this will do the trick if changes are not being saved in Explore.

0


Hi Tom, 
 
Thanks for reaching out in our community as well by the way. 
 
A ticket was also created from this reply, and as I have also already contacted our Explore team, please refer to my reply on your ticket with me. 
 
Regards, 

0


Hi Guys, 

I'm trying to create a query that will measure my team's AVG resolution time for tickets per ticket priority, and struggling with excluding the extreme results (so it will not affect our AVG dramatically). 

I created the following metric to filter out the 'pending's from the calc:

Resolution time without pending (Biz - days)
IF (NOT INCLUDES_ANY([Ticket tags], "followup_ticket_creation", "closed_by_merge", "closed_by_trigger_blocked_email_channel"))
THEN(VALUE(Full resolution time - Business hours (hrs))-VALUE(Agent wait time - Business hours (hrs)))/24
ENDIF

Finally I got the AVG resolution time per priority, but I found it is not reflecting the reality as few extreme cases are influencing the AVG dramatically. 

Therefore, I tried to use the "Results manipulation" -> "Results metric calculation" to filter out values that are too far from the STD_DEV by creating the following metric:
ST_Dev AVG Resulution time
IF (NOT INCLUDES_ANY([Ticket tags], "followup_ticket_creation", "closed_by_merge"))
AND VALUE(Full resolution time (min)) < 3* STD_DEV(Full resolution time (min))
THEN
(VALUE(Full resolution time - Business hours (hrs))-VALUE(Agent wait time - Business hours (hrs)))/24
ENDIF

 

Unfortunately, the Results manipulation metric is not working as expected, and I couldn't find what is wrong, or what is the best way to exclude the tickets with the extreme AVG resolution time from the general AVG. 

 

Please assist me in this one, as it is a crucial report for measuring the team performance. 

 

Best Regards, 
Tom

0


Thanks for the reply 421249272894!

I didn't have much luck with the repeat pattern - it seemed to just lump the previous periods' data into the number shown.

Correct me if I'm wrong, but I think by using the two metrics for different quarters I've achieved the result I was looking for?

0


Hi Matt,

With the use case that you presented, I think the Date range calculated metrics would help you out on this. You can check the article Adding time and date calculated metrics regarding this one.

You can use data for a quarter of a year under the Advanced tab and Add a repeat pattern within the calculated metric to compare the current quarter to another quarter of the same year or a different year.

For more information about adding a repeat pattern, you can check the article Editing dates and date ranges. Hope this helps! :)

0


I think I know the answer to this already, but I'm looking to create a dashboard that shows data quarter by quarter.

At the moment all I can show is data from quarters that are already complete. There doesn't appear to be a way to show data dynamically to-date. For instance, can I compare 20 days into Q3 with 20 days into Q2?

I don't want to show the total amount for Q2 and the amount so far for Q3 as you can't make a useful comparison.

Hope that makes sense?

0


Sign in to leave a comment.