Auto Escalation question

[Adrian Simpkins]

Hi All

So I have a test service setup with a test SLA with auto escalation rules setup currently only for priority 4 and priority 5 requests (has priorities set from 1 to 5 but I have only inserted auto escalation rules into the bottom priorities at the moment).

I have raised some test requests against this service as incoming at level 5 priority, and have accepted the request then I have been monitoring for the auto escalation to trigger. I can see the movement from Priority 5 to Priority 4 along with all the associated trigger events taking place (notification to owner, manager, add to board etc). However, I was expecting the request to move up to level 3 as once it moved to priority 4 and the next set of triggers are reached, and I have just been monitoring for the next escalation. 

However, the requests are still at priority 4 and have passed the timing trigger point but have not moved up another level - am I correct in assuming it would move up automatically to priority 3 ?

Many thanks

[David Hall]

Hi @Adrian Simpkins

Sounds like it should be moving as you would expect, but without seeing the configuration etc it's sometimes hard to confirm whether the timer has not reached a firing time due to a mis-configuration of the escalation time or if there has been an error when it has fired. 

Perhaps as a test you could try adding a notification/email notification to yourself for each one to see whether you get an email even if the priority doesn't change, that would help to confirm if the event fired or not.  If it does fire the event and send the email then there could be an issue specifically with the escalation of priority etc which could be investigated via the log files.

Kind Regards,

Dave

[Adrian Simpkins]

Hi David, thanks for the above - I had an epiphany moment and realised we have set our timings incorrectly as we have not considered the expiration of the overall SLA timer, and have treated each priority level as if they have a fresh timer reset (which of course would not happen). So I will revisit the timings and test again.

Out of interest I would be interested in how this is configured / used by others, as I can see issues with how to configure it as we want to raise a requests priority level if the request starts to breach, but of course as the priority level rises, the timers for response / resolution become shorter. So in effect how to raise a priority 5 up levels but taking into consideration that the next priority up has a shorter Response / Resolution in place. 

Many thanks as always

[James Ainsworth]

31 minutes ago, Adrian Simpkins said:

we want to raise a requests priority level if the request starts to breach, but of course as the priority level rises, the timers for response / resolution become shorter.

Hi Adrian, 

We provided a separation between Priority and Service Level to accommodate for this scenario that you mention.  I agree with your approach to increase the priority of a request as the target comes closer.  The Service Level Targets will only become shorter if you are using Priority as one of the rules to assess which Service Level to use.  If you are using Priority as one of your criteria for the Service Level Rules, then you can consider other criteria to set the initial Service Level, this way allowing the priority to be changed without it influencing the Service Level being applied.  

Regards,

James

[Adrian Simpkins]

Hi James, Thank you for the comments - now you specify the separation of Priority and Service Level here, I can see how both would work in tandem.

I am already working on inserting some prioritisation questions into the catalogue items to auto set the Priority, so I will make this part of this rollout as well. Currently we only have 2 catalogue items where we ask some questions to set the priority in the BPM dependant on the responses, and I can see this is key to implementing this new feature for the teams as correct prioritisation at the outset of raising a request negates some of the need to set this in the SLA escalation logic.

Many thanks as always