Page 1 of 1

Incident request Logging

Posted: Sat Sep 25, 2010 4:19 pm
by rhackett
Hi all,

Our servicedesk engineers suffer from "corridoor calls" i.e. people approaching them for help rather than ringing the servicedesk.

I am looking for benchmarks on what is the typical percentage of calls like this, I've not found any studies or stats.

Also do most service desks accept this user behaviour and log the call themselves or not actually log the call.


Posted: Mon Sep 27, 2010 11:41 am
by Diarmid


Firstly there is no such thing as a "typical" figure or benchmark for such behaviour. Nor is there an average or a tolerance threshold.

The issue is management and control.

As far as casual requests are concerned, the only acceptable number to deal with is none whatsoever.

Control is not retrospective. You can empower your engineers to make decisions "on the spot", but you cannot permit them to make uninformed decisions and the limits of their scope and authority must be clearly defined.

How do your service desk engineers establish that the person waylaying them is authorized to make the request?

How do they establish priority for the request?

How do they inform "HQ" that they have been delayed?

How do they account for their time?

What do they do if they are not the best person for the particular task (but think they can muddle through okay)?

How do they access the history of the service component(s) they are asked to do something to?

Of course, they are stymied if a change is required anyway.

These kind of issues need addressing. Nevertheless there is also the issue of travel and saving journeys where remote sites are involved. So, still dealing with all the control issues, it might be prudent to enable an engineer to perform a number of actions on one visit, if (to repeat myself) you still achieve full management control of the activity by some means.