Incident Classifications

[BAX]

Hello ITIL folk,

I am trying to come up with some new classifications for our call types that come through our service desk but we are having probelms deciding how to best classify the incidents.

Should we classify them by symptom, or service offered, or service affected... etc...?

Any help here would be much appreciated.
Thanks in advance,
BAX

[jbryde]

We recently had a huge debate on how to classify our incidents and came up with the following principle:

Always log what the customer or user describes as the symptoms, and connect them to the service the customer or user states beeing affected.

Our "inbound" classification should always give answer to the question: What is the customer trying to do (class.), but can't, and where (service) is he or she trying to do it?

We then added an "outbound" classification, to answer the question "what did WE do to resolve the incident?" (to be filled out before closing the ticket).

By doing it like this, we hope to faciliate knowledge management, and re-use solutions by searching on the symptoms, rather than the solution.

Good luck, BAX.

J. Bryde
Advisor
The Norwegian Correctional Services IT-Centre

[BAX]

Many thanks for your reply.

That does make a lot of sense and would solve the problem of trying to fit our current classifications (based on symptoms) in to our intended classifications (based on service).

I will feed this in to our debates and hopefully it will put us on the right tracks.

Thanks again for your time.
Paul.
_________________
Paul R Baxter
IT Support Analyst/DBA
University of Leeds, UK

[BAX]

Hello Again, sorry to dig this one up...

We are looking at defining our incident classifications based on:

Service >> System >> Incident

Where:
Service is the service lifted from our service definitions (e.g. email services)
System is the system affected ( e.g. Outlook Exchange) regardless of server or client issue
and Incindent is the nature of the call - i.e. Login problem, client issue, Quota increase request ....etc....

We feel this structure is good but have a problem...
The problem we are having, is that the list of incidents seems too generic for some call types, but for others, we can go in to great detail.

For example:
For logging Calls relating to our student service system we have:

Student Service Systems >> JoeBloggs Student Info System >> Login Problem, Functional Issue, Connection Issue.

As for Outlook Exchange types:

Email Services >> Outlook Exchange >> Login Problems, Outlook client issues, GAL Issues, etc.. the list goes on...

So the difference between these two is that one of them shows all types of symptoms and the other is more classified in to Access problem, functional problem, connection problem.

We seek consistency in our incident types. should we have a list of all possible symptoms or have them more generic and less descript?
There are benefits and drawbacks to each , but we dont want a mixture of the two.

We are nearly there, so any advice on this would be most useful!!!
Many thanks in advance,
Paul.
_________________
Paul R Baxter
IT Support Analyst/DBA
University of Leeds, UK