Insights metrics reference (Professional and Enterprise)

Return to top
Have more questions? Submit a request


  • Billy Macken

    Hi Alex!

    Functionally, there isn't any difference between the two metrics. The longer definition in the first metric allows solved tickets to be sliced by Event date attributes in addition to Ticket Solved date attributes. This is so the results from the # Tickets Solved metric would make sense alongside other metrics that might not make sense using Ticket Solved date attributes.

    I hope this helps-- Happy reporting!

  • Graeme Carmichael
    Community Moderator


    The first reply time belongs to the ticket not any support group. So transferring a ticket between support groups has no effect on the first reply time metric.

    Even if the final support group replies instantly on being assigned the ticket, the first reply time will reflect all the time through all the groups that the customer has waited for a response.

  • Rob Stack
    Zendesk Documentation Team

    Hi Emile Cohen, I'm not an API expert, but on a quick look at the docs, it does seem like there might be different options for each API. I'd suggest starting with this article - - and asking the question in there if it doesn't help.

  • Nhia Lor
    Zendesk team member

    Hi Daniel,

    Based on what you're wanting you should definitely be able to use the attribute "Ticket Group (Historic)".  As mentioned here, the historic attribute shows the value at the start of an update, not the end.

    So if your tickets are moved from say Group 1 to Group 2.  Then back to Group 1.  Group 2 should be one of the attribute values on that ticket under the "Ticket Group (Historic)".

    Therefore if you were to slice your Satisfaction metric results by the "Ticket Group (Historic)" attribute, you should in theory be seeing the satisfaction results compared and broken down by the groups which the tickets were assigned from one group to another at some point in the tickets history.

    Would you be able to confirm if in your report you are slicing your metric results by this attribute?

    Another alternative method to reporting on this data I can recommend is to build some custom metrics to report on the ticket satisfaction based on movement between your respective groups.  This recipe here should help further if you want to explore this option.

    If so and you're still not receiving the results as expected please let us know so that we can take this one offline here and dig a little further into this for you!

  • Leah
    Zendesk Voice of Customer Team

    Hi Monica!

    For this, you’ll need to create a custom metric. If I understand you correctly, you’ll need to create two different metrics: one for % Tickets Created and one for % Tickets Solved. The metrics you want are:

    SELECT # Tickets Solved/ (SELECT # Tickets Solved BY ALL OTHER)


    SELECT # Tickets Created/ (SELECT # Tickets Created BY ALL OTHER)

    For your report, you’ll select these two metrics and then select Ticket Assignee under the How section. The report will show a distribution of how many tickets are created and solved for each of your assignees by percentage.

  • Julia Kornilenko

    Hi there!


    I'm wondering if anyone can help on the case:

    I need to create report with users who created at least one ticket (monthly):

    1) during the selected month

    2) cumulatively from the very beginning till the end of the selected month

    I can easily calculate the first one ( select count(User, Ticket Id) where User Active = true and Ticket Status <> Deleted)


    But the second one would not work (select count(User, Ticket Id) where User Active = true and Ticket Status <> Deleted and Date (Ticket Created) < (select min(Date (Timeline)) by all other))



    P.S The reason I need cumulative data is to have number of unique users from the very beginning till the end of the month.


    Any advice would be appreciated.

  • Jamie Davies


    I am having trouble creating averaged measures, do you have any advice?




  • Tom Kippenberger

    From the descriptions, it looks like the three Ticket Age metrics all keeping counting even when a ticket is solved of closed. They do a comparison to 'Present' rather than to 'Closed Date'. Is this correct?
    Ticket Age (days) [AVG]
    Ticket age (days) [MAX]

    Ticket Age (Open Tickets)

    I would be more interested in:
    Average age of Unresolved tickets
    Average age of Tickets when Solved
    Average age of all tickets (combination of the above)

    Does anyone else have an example of these?

  • Daniel Petersson

    Hello Nhia!

    Thanks for the reply.

    I am slicing the report by Ticket Group (Historic).
    Currently I'm trying this as simple as possible by:
    What: % Satisfaction Score
    How: Ticket Group (Historic)
    Filter: Ticket Group (Historic) is Group1, Group2 and Group3 (actual group names are Integration, Konvertering and Techsupport)

    Here I get the results that satisfaction is 80% for G1, 75% for G2 and 30% for G3 which I suppose could be true even if 30% seems very low.
    (No filter applied regarding time period at this time)

    When I try to slice this into actual ticket id's however (in order to see what the rating is actually based on in data) I get more or less no results, perhaps around 80 NKI ratings total for all of the groups since 2013. Seeing as we have around 20-25% answer rate on our Satisfaction-ratings and just one of these groups (Group 3) sends back around 200 tickets per week back to the base-support-group this seems extremely low and makes we wonder if this is correct?

    Another question regarding this is if I want to split this into weeks what metric should I be using, Week (mon-sun) (Event)? Or which metric would be best?

    Thanks in advance!

  • Daniel Petersson

    Perfect, just what I was looking for - couldn't manage to get # Replies to go along with Updater but that explains it. Thanks again Nhia!

  • Jean Camalionti

    Hello guys!


    With the new chat metrics in GoodData many new possibilities have appeared and I have felt great difficulty in just one of them ... we have the "Agent activity metrics" in specific "Online Duration" and "Online Chatting Duration", is valuable information but I was not able to separate per agent from the chat, only the sum of all, which makes the metric a little pointless for me ... has anyone been able to separate these specific metrics by chat agent?


    Many Thanks,


  • Stephen Fusco
    Zendesk team member

    Hey Gurunn, 

    When you create a report for Group Stations by Group Name the number you're seeing is the # of tickets that have been changed to that group. 

    Thanks for your question! 

  • Billy Macken

    Hi Jill!


    I would recommend checking out our article on Building custom metrics for the events model-- you're going to want to specify the groups you're interested in using the [Text Field] New Value and [Text Field] Previous Value attributes to pull tickets that had an update submitted that changed groups in the way you've specified.


    I hope this helps! Happy reporting!

  • Daniel Petersson

    Hi Jessie!

    I might have been a bit unclear in my message, I just want to measure the satisfaction separate for each group which touched the ticket. So for instance I have my first line support group: "Support", my first second-line supportgroup "Techsupport", my second second-line supportgroup "Integration support" and my third second-line supportgroup "Conversions".

    90-95% of all tickets come to first line supportgroup "Support" first but in 5% of cases they have to be escalated to one of the three second-line supportgroups.

    I want a report to measure all the tickets that has touched each second-line separately, i.e:
    Group                       Satisfaction
    Techsupport              85%
    Integration support    75%
    Conversions              80%

    The purpose is to measure how (and if) satisfaction% is changed from a normal first-line support ticket in comparison to tickets escalated to a second-line support. Since second-line support-tickets usually takes longer time I would guess the satisfaction drops when being escalated.

  • Brett Bowser
    Zendesk Community Team

    Hey Daniela,

    Have you had a chance to look at our Explore Recipes Reference to see if there's any already created reports that would get you what you're looking for?

    It may also be worth posting your question in our Explore topic to help provide visibility to some of our Explore experts.

    Hope this helps!

  • Graeme Carmichael
    Community Moderator


    Yes, that is correct. Think of it as the ticket failing, rather than the last group.

  • Nhia Lor
    Zendesk team member

    Hi Daniel,

    In this situation the "Ticket Assignee" attribute isn't going to be suitable for what you're needing as this attribute will always only contain the value of the current agent assigned to the ticket.  It doesn't record and store the value of the agents/admins who performed the update on the ticket on the event.

    You can approach this in two ways.  The first is that you would need to break this into seperate reports.  One to report on the tickets solved by ticket assignee.  And the second is to report on the public comments by updater.

    For this second report you would want to use the metric "# Public Comments".  And the attributes you'd want to use are "Updater" and "Week (Mon-Sun)/Year (Event)".  You can then filter your report results to just "Updater Role" is "Agent" and "Admin".

    However if you'd like to combine both within the same report it won't be possible to use the default "# Tickets Solved" metric and may require creating a custom metric.  This helpful recipe HERE should offer a good starting point.  You'll need to modify the metric formula so that it looks for the ticket solve event.  This should be represented by the following:

    • Text Field = Status
    • [Text Field] New Value = [Status] solved

    This metric you should then be able to use in the same report as the one above reporting on the comments by updater name.  The metric should then return the number of tickets that were solved based on the updater who solved the ticket.

    Hopefully this helps you further on this one!

  • Daniela López Aguilar

    Hello! Can anyone help me? I'm trying to build an agent occupancy report for email and chat using zendesk explore, does anyone know which metrics I would need to use and how to go about that? Thank you!

  • Brent

    Hi Tom,

    You are correct - Ticket Age (days) [AVG] and Ticket age (days) [MAX] will measure the age of a ticket from the day it was created to today, regardless of its status.

    Making custom metrics for your own needs based on the default metrics should be pretty straightforward:

    • Average age of Unresolved tickets: Take the "Ticket Age (Open Tickets)" metric, make a new copy of it, and replace "MAX" with "AVG" in its MAQL.

    • Average age of Tickets when Solved: I duplicated the "Ticket Age (days) [AVG]" metric and modified it slightly to calculate the number of days between the date a ticket was created and the date a ticket was solved:

    SELECT AVG(SELECT Date (Ticket Solved) - (SELECT Date (Ticket Created) BY Ticket Id)) WHERE Ticket Status IN (Closed, Solved)

    • Average age of all tickets - The default metric "Ticket Age (days) [AVG]" might be what you're looking for here. It measures the time between when tickets were created and today.

    Otherwise, if you mean you want to combine the average age of unsolved tickets (today minus Date (Ticket Created)) and the average age of the tickets when they were solved (Date (Ticket Solved) minus Date (Ticket Created)), I don't think that is possible with Insights. My best recommendation in that case is to use the two metrics I described above side-by-side on a dashboard and compare them.

    Hope that helps! Please let us know if we can be of further assistance or if you'd like more detail on how to make the metrics.

  • Thomas

    Hi There, 

    I am wondering if it's possible to have a report that displays, for example, the 25 tickets in the past 30 days, with a status of either solved or closed that had the longest resolution time and that had highest first reply time? 




  • Kent Carlo Jimunzala

    Hi Graeme,


    So I want to confirm if the first reply time is already a failure to the first group. Will it reflect towards the final group?


Please sign in to leave a comment.

Powered by Zendesk