Standard calculated metrics in Explore are dynamic. They can recalculate based on the attributes used, date periods specified, and filters applied (see Standard calculated metrics). However, if you are comparing across categories or benchmarking, you might want the calculation to remain constant. To achieve this, you can use fixed calculated metrics.
Creating fixed calculated metrics
At the most simple level fixed calculated metrics can be used to produce unchanging metric results. For an example of a more complex use, see Creating a constant average using fixed calculated metrics.
Advanced fixed calculated metrics enable you to construct a formula in the same way as a standard calculated metric (see Standard calculated metrics and attributes). Simple metrics do not require you to enter any formulas, but the range of metrics from which you can choose are more limited.
This topic contains the following sections:
Creating a simple fixed calculated metric
In this example, you'll create a simple fixed metric containing tickets that were reopened, that contain the tag 1st_floor.
To create a simple fixed calculated metrics
- In Explore, open a new report using the Support: Tickets dataset.
- In the report builder, open the calculations menu ().
- Select Fixed calculated metric.
- Enter a name for the metric, in this example, Tickets reopened with 1st_floor tag.
- Select the Simple option for creating the metric.
- From the Will always return the value of this metric drop-down list, select Reopens.
- Choose the computing options you want to apply. There are two available:
- Compute according to the filters in the report restricts results to your applied filters. If you do not select this, the fixed calculated metric will not be affected by your result manipulations.
- Restricted for the values of an attribute restricts results to a specific attribute's values. If you do not select this, the fixed metric will not be divided by any attribute.
- When you are finished, click Save.
Creating an advanced fixed calculated metric
In this example, you'll create an advanced metric that uses a formula to display only the tickets tagged support.
To create an advanced calculated metric
- In Explore, open a new report using the Support: Tickets dataset.
- In the report builder, open the calculations menu ().
- Select Fixed calculated metric.
- Enter a name for the metric, in this example, Tickets with support tag.
- In the formula window, enter If ([Ticket tags]="support") THEN [Ticket ID] ENDIF".
- When you are finished, click Save.
Creating a constant average using fixed calculated metrics
Creating your fixed calculated metrics
You will need to create two fixed calculated metrics to use in your average calculation. The first calculated metric will return a fixed number of replies and the second will return the fixed number of month values.
To create a fixed replies calculation
- In Explore, open a new report using the Support: Tickets dataset.
- In the calculations menu (), click Fixed calculated metric.
- On the Fixed calculated metric panel, enter a name for your metric, for example, Fixed agent replies.
- Select the metric you want in your fixed average from the Will always return the value of this metric drop-down list. This example uses the Agent replies metric.
- Click Save.
To create your fixed number of months metric
- In Explore, open a new report using the Support: Tickets dataset.
- In the calculations menu (), click Fixed calculated metric.
- On the Fixed calculated metric panel, click Advanced.
- On the Advanced fixed metric panel, enter the name for your second fixed metric, for example Fixed month.
- In the Formula box enter DCOUNT_VALUES([Ticket created - Month]).
- Click Save.
Creating your standard calculated metric
After you have created your fixed calculated metric, you can create your average metric. This example uses a standard calculated metric.
To create your constant average comparison
- In Explore, open a new report using the Support: Tickets dataset.
- In the calculations menu (), click Standard calculated metric.
- On the Standard calculated metric panel, enter a name for your average. This example uses Monthly average agent replies.
- In the formula field, type the following formula (replace the example metrics here
with your own metrics):
SUM(Fixed agent replies)/SUM(Fixed month)
Your metric should look like the following example:
- Click Save.
- Add your standard calculated metric to your report as a trend line (see Adding metrics as trend lines).
You will now be able to see how your monthly replies compare against the average. Your report should resemble the image below:
16 Comments
How can I create a chart which compares the average weekly ticket solved by an individual agent vs. the average of the all team. This should be basically a time series chart includes two lines: individual agent solved ticket vs. average of all agents solved tickets.
Hello @...,
The best way to create an average like this is by using the ATTRIBUTE_ADD function that is part of the Explore aggregate functions EAP.
Here are some links about the EAP:
This one is not available for me. I have signed up for the EAP. How long does it take to be included?
Hello Gokhan,
I've gone ahead and pulled this into a ticket so one of our experts can look into this for you. You should receive an email followed by a response relatively soon.
Best regards.
Hi There!
We have our Organizations divided into groups in Zendesk. If we wanted to show the total number of Organizations per group shown in Queries in a fixed way (so that the number doesn't change when we select date ranges, for example), would that be possible?
Hello. I am trying to create a result metric calculation which multiplies Full Resolution Time by Solved Tickets.
I enter the name, the formula "SUM(MED(Full resolution time (min))*SUM(COUNT(Solved tickets)", but the option to +ADD stays greyed out. I've also tried "SUM(MED(Full resolution time (min))SUM(COUNT(Solved tickets)"
What's the proper syntax for multiplying?
Hi Leopoldo,
I remember you reached out via a ticket previously and I was able to answer you from there. Hope I cleared this one out. Cheers! :)
Hi Nicholas,
I think the article Using result metric calculations will better guide you regarding your concern.
I tried to do a raw formula based on what you mentioned and this is how it looked like on my end:
I'm not totally sure though if that's what you're trying to achieve.
You can also check Explore functions reference which lists all available functions or syntax in Explore and how to use them. Hope I was of any help. :)
Thank you! This has pointed me in the right direction.
Hi, I try to create "fixed calculated metric" which will be also "date range metric".
Scenario: I would like to track % of One-Touch tickets and compare the score to the % of One-Touch tickets for 2021:
I would like to:
1. The top value to change depending on the period of time I am selecting in tab filters AND
2. The bottom value to be based on the fixed result for 2021
My problem now is that 1. works correctly but 2. does not. The base for Bottom value also change when filter change.
Scenario 2: I have chosen a time filter to show the result for Today:
1. Top result is correct.
2. Bottom result is incorrect. Instead to compare Today result to fixed 2021 result, it change a base forr the bottom metric according to the filter. So of course there is 0% difference between Top and the bottom score.
Is there a way to disable time filter for bottom metric?
Hi Elzbieta,
I'll need to take a closer look into your query where you're trying to create a fixed calculated metric so I can properly provide you a resolution for the scenarios you have mentioned.
I've gone ahead and created a ticket on your behalf. Kindly lookout for an email notification. Cheers! :)
Hi Elaine,
Good day!
Is it possible to use queries to create a new calculated metric?
ex.
Query 1 + Query 2 / 2
2 department has merge and we are trying to get the overall performance of the business unit using the above formula.
Thanks in advance
Hello! Maybe you can help me with the issue I faced by using the Fixed Metrics.
Long story short: I need to make the average amount of closed non-merged tickets by dividing the total amount of those tickets by the total amount of agents. With Agents, it is working fine giving me the total result but the issue appears with the number of tickets. Here you can see the formula I use with the Standart Calculated Metric:
It gave the precise result so I basically copy it in the Fixed Metric, but the results there are differ:
The correct result if on the Standart Metric and it is 176. For some reason, it doesn`t count the "NOT Include any" part and shows me the total amount of tickets (203) in each field. Still, that should be 176 as well.
Can you please help me with that?
Just in case, how it look in the Fixed metric as well: https://flmst.net/s/fppHggAq2tKeDpe/preview
Yes, it does look like INCLUDES_ANY does not work well with an advanced fixed metric formula. As a workaround, I suggest that instead of filtering tags at the metric level, you may create a standard calculated attribute (Filtering a specific tag from all metrics on a report) and apply that as filter to the whole the query with the native Tickets metric and an advanced fixed metric for ticket count.
Hi Team,
i want to create metric about "unsolved tickets time since update-business hrs" but i couldnt do. This metrics doesnt include business hours, and this is the screenshow what i am trying to add.
can you help me ?
Thanks
I'm afraid there's no option to calculate that metric in business hours, even through a custom metric. At this time, the only ones that are measured in business hours are the default metrics for first reply time, resolution time, agent/requester wait time and on-hold time.
If you have time, I recommend that you start a thread in the Product Feedback topic in our Community to engage with other users who have similar needs and discuss possible workarounds. Conversations with a high level of engagement ultimately get flagged for product managers to review when they go through roadmap planning.
Thanks Emre!
Please sign in to leave a comment.