How many Incidents should we expect?

An open discussion on issues related directly or primarily to the service or help desk.
Post Reply
User avatar
Beanietdc
Itiler
Itiler
Posts: 8
Joined: Tue Sep 25, 2007 8:00 pm
Location: UK

Fri Apr 25, 2014 5:34 am

Hello, I am trying to find out if there is a standard for the amount of calls that a Service Desk should expect for x number of users.

I am the (fairly new) Problem Manager for my company and am trying to work out what level of calls we could consider "normal" for a given application as opposed to various members of management thinking that it should be around the level of none 8O

I am aware that particular applications do have Problems associated with them and we are investigating these but some sort of benchmark figure would be helpful when it gets to the point of trying to persuade the management that the amount of time that a decent size group of fairly well paid people spend investigating an issue isn't worth the effort for how many incidents it might save.

I hope that makes sense! Does anyone have any rough figures that I might be able to reference?

Thank you
Tina


User avatar
ChangingMan
Itiler
Itiler
Posts: 27
Joined: Sun Apr 27, 2014 8:00 pm

Thu May 01, 2014 7:18 am

There are some gartner figures on the internet that say x amount of users should require x amount of service desk head count but they are generic and not much use at all in my opinion.

The method of persuasion for management to invest in a help desk is simple. If your service desk is logging and flogging you paying a lot MORE for each incident.

Let me explain that a little:

Every single call has a $ cost when coming into the service desk, that cost is made up from the cost in the loss of productivity caused by the problem + time taken by person to log + time taken (and cost of resource) by analyst to log.

Should the issue not be resolved in that first instance that cost is then increased.

+ continued loss of productivity
+ time and resource of 2nd analyst and potentially of user testing the fix

Good service desks fix 90% of issues first time. Doing so reduces overall cost to the business.

I understand this is very simplified but using this logic you should be able to show a financial benefit to increased investment in the Service Desk.
User avatar
Beanietdc
Itiler
Itiler
Posts: 8
Joined: Tue Sep 25, 2007 8:00 pm
Location: UK

Thu May 01, 2014 10:40 am

Thanks for that but I am not sure that I explained myself very well. :(

What I am trying to drive towards is showing that the number of incidents that we get for a particular system is within the levels that we might expect for the number of people using it.

Senior management seem to be determined that there are lots of issues with the system and I need to keep gathering (expensive) analysts/techies together to go through each issue and try to find a resolution. I am becoming more and more convinced that, a couple of Problems aside, the calls that we are getting are consistent with the number we should expect from the system under normal operation.

The issue is that in order to be able to take this argument to the Senior Management I need to have numbers or I know they won't believe me.
User avatar
ChangingMan
Itiler
Itiler
Posts: 27
Joined: Sun Apr 27, 2014 8:00 pm

Thu May 01, 2014 11:16 am

Beanietdc wrote:Thanks for that but I am not sure that I explained myself very well. :(

What I am trying to drive towards is showing that the number of incidents that we get for a particular system is within the levels that we might expect for the number of people using it.

Senior management seem to be determined that there are lots of issues with the system and I need to keep gathering (expensive) analysts/techies together to go through each issue and try to find a resolution. I am becoming more and more convinced that, a couple of Problems aside, the calls that we are getting are consistent with the number we should expect from the system under normal operation.

The issue is that in order to be able to take this argument to the Senior Management I need to have numbers or I know they won't believe me.
So the very short answer to your question is "no" those numbers do not exist.

The number of users on any given system is an irrelevance, you can't make a correlation between number of users / number of incidents because user behaviour on the system will be inconsistent.
Essentially one user could take up 90% of the systems resource while 1000 could use less that 1%.

The only realistic comparisons you could make would be with other simialr levels of deployment of that system (so for example another CRM Oracle deployment in a similar sized company) but as the information is very rarely public I could see that being difficult to do.

You have to remember that to the business the only acceptable level of failure within any system is 0.
User avatar
jbednarik
Newbie
Newbie
Posts: 1
Joined: Sun May 18, 2014 8:00 pm

Mon May 19, 2014 3:06 pm

Agreed, no number for that. What you can do is striving to minimize the Incidents by applying the
  • preventive maintenance and
  • training
The first one keeps in good shape the hardware (and software), the second one is for the minds. I'm talking about internal heldpesk for the users. Helpdesk for external customers should be more difficult.

Rather use your resources and time for avoiding the Incident than fix them.
Post Reply