Performance Analytics for Incidents

Community Alums · ‎10-07-2016

Hello, How are you using Performance Analytics for Incidents? PS. What's the difference (technical view) between Performance Analystics and normal Reporting module available OOTB?

Chuck Tomasi · ‎10-07-2016

Hi Rafael,

Most customers are using PA for incident management with OOB indicators to determine where they can improve the process. Some have gone further and added additional leading and lagging indicators for additional information. For example, how many times has an incident been escalated from L1 support to L2 support over the last month? Is this trend increasing or decreasing? Once you know that, you can start to look at root causes like staff turnover, training, new technologies implemented, etc.

Technically, reporting is a current view of how the data is today based on live data. It represents the live data. In some cases, that's too late to do anything about.

Performance Analytics takes snapshots to save meta data (scores) that you define. Using this information you can start to gain insights to how your data is changing over time and where you can make performance optimizations based on decision points (or indicators.) You define the indicator sources, indicators, of what data you want to see and breakdowns (e.g. by category, assignment group, etc) how you want to see it.

Let me know if you need more detail than that.

https://docs.servicenow.com/bundle/helsinki-performance-analytics-and-reporting

View solution in original post

Zanna · ‎01-17-2018

Hey Robert,

Thank you for such a prompt answer!

But here's a report that shows data over time too:

We can see how the number of tickets changed over a couple of quarters.. Or do I miss something here?

Uncle Rob · ‎01-18-2018

You're graphing by open date, which only takes the tickets you've selected and arrange them by the date they were opened. Consider:

- how many were opened and STILL OPEN at each interval?

- how many tickets opened from previous intervals were still open at each interval of you X axis?

- how many of the "open at that time" (which you can't currently show) had not been updated in 5 / 10 / 30 days (at each interval)

- how did your backlog (opened during interval minus closed during that interval) change over time?

- what's the avg duration of all the tickets you closed, at each interval, over all that time?

Everything in your chart is perceivable because that's what the data looks like now. But many KPI's will depend on comparing data from past to data right now.

Happy to give you a live demo if you wish.

zyatiko · ‎01-18-2018

OK, now I'm starting to see what's different.

And if I make a report on tickets created during July 2017 and September 2017, it won't reflect the whole situation, right?

Whereas with PA, we can compare this number to other relevant data and analyze it all within a particular cotext?

Thank you so much for your help! Hope it wasn't too irritating to explain obvious things

Uncle Rob · ‎01-18-2018

Performance Analytics is one of my favorite aspects of ServiceNow, so definitely no irritation.

And if I make a report on tickets created during July 2017 and September 2017, it won't reflect the whole situation, right?

If you're doing this with OOB / Non-PA the only thing you can determine from tickets created 06-2017 to 09-2017 is what is true *NOW*. You can't determine what might have been true in the past.

Whereas with PA, we can compare this number to other relevant data and analyze it all within a particular context?

PA is about comparing truths across time. Ponder this example.

Take 3 open tickets at random. What's the average amount of time the 3 have been opened. Now ask yourself the same question tomorrow about the same 3 incidents. Oops, one of them closed overnight.... but the remaining two have are now older by a day. Point being... your average is a completely different number today. In PA you've stored the number for yesterday and today. With OOB reporting you can't tell what things were like yesterday.

Howbowdah?

zyatiko · ‎01-18-2018

Wow, okay)

So, the example you gave is clear. But what about this statement:

"If you're doing this with OOB / Non-PA the only thing you can determine from tickets created 06-2017 to 09-2017 is what is true *NOW*."?

So if a ticket was created in August and then cancelled or somehow deleted in a couple of days, we won't see that in a report (made with OOB functionality)?

Would it be correct to say that the whole thing is that the data in the tables can be changed. Reporting relies on that data, which is why one and the same report will show different things today and tomorrow (like in the example). And PA stores the data as it was at some point in time so that the needed 'pieces' are not lost?