Help with a report on INCIDENT State Changes

tobrien
Kilo Guru

Hi,

The incident's ACTIVITY FORMATTER on the NOTES tab is tracking the state changes of an incident.

E.g. an entry might show the "field" State was NEW and is now ACTIVE with a timestamp.

We have 7 possible state changes and I need a report to roll-up, say, the AVERAGE time between each state as incidents progressed. 

Other references to the Activity Formatter in this community suggest that the data is in sys_journal_entry or sys_history but I do not see it there.

Can anyone tell me where his data is saved? Then I can figure out whether or not the SNow Reports UI can do this averaging or if I have to write a script.

 

Thanks,

tony

 

 

3 REPLIES 3

Linda S
ServiceNow Employee
ServiceNow Employee

This is out of the box on a Metric see this documentation:  https://docs.servicenow.com/bundle/newyork-it-service-management/page/product/incident-management/co...

Navigate to Metrics -> Definition -> Incident State Duration

Run report on this table - metric_instance with the criteria of Definition = Incident State Duration

tobrien
Kilo Guru

Thank you, Ms. Linda

I looked through the REPORTs and METRIC areas you suggested; and I am left with these followup questions:

1. The report using that DURATION metric returns all qualifying rows in [metric_instance]. I can limit them by selecting, say, the VALUE = New, to return only incident-durations whilst in the STATE = NEW. But what I seemingly cannot do is limit the output to specific customers or any other field associated with the [incident] table. How do I further limit the number of rows returned using Incident Table limiters?

2. I found the "definition" of the Incident State Duration metric but the script is blank. Where is the actual working algorithm/script?

Thanks, tony

 

tobrien
Kilo Guru

I've discovered that you can use a DATABASE VIEW to 'join' the incident table to both the metric instance and definition tables and gain access to all 3 table's fields.

There are OOB Dbase Views; one of which is called incident_metric !

 

tony