This isn’t as simple a question as it might first seem. Let’s look at the issues which need to be factored in to your decision.
IT help desk response times – average and worst case
The number of check-out staff at the local supermarket varies throughout the day. Some marts will try hard to ensure the lines never exceed a threshold – say no lines with more than four people waiting – by deploying staff to the tills in an on-demand mode. At quiet times some of those operatives will either be sitting idle at their tills, or taking a rest break, or gone home having completed their shift. So dynamic are the demand patterns through the day that it’s not possible to match the need for staff through shift patterns alone. The policy of ensuring that customer delays don’t breach a threshold means that the staffing level will at times exceed requirement. (That excess capacity may of course be assigned to other tasks in the supermarket when not needed at the check-out, to gain some return from the employment costs.) At the other end of the scale of customer-orientation, the supermarket may employ just enough check-out staff to process a typical number of customers for that day of the week, and rely on lengthening customer lines and delays to absorb fluctuations during the day. Our IT help desk manager faces just the same dilema.
IT help desk Service Level Agreements
Whereas all the grocery customers can do is grumble if they have to wait a long time, most IT help desks operate under a regime where they will be penalised for failing to meet a resolution time defined in an SLA. All quality help desk and issue tracking software tools will provide means to define multiple SLAs, and will provide reporting on performance relative to SLAs. This should include export in spreadsheet format to facilitate further processing such as calculation of any penalties in regard of failures to meet targets set. The help desk manager now has to consider the relationship between his/her employment costs and the consequences of under- or over-performing against SLAs.
The organization or the help desk? What is the right context for setting staffing levels?
Is your IT help desk regarded purely as a cost center, or does it effectively trade with user departments through charge-back or any other form of levy? If you are a simple cost center, then a model can be built that relates the cost of the size of your team to the service levels to the organization and hence the value to the organization. (That may be calculated any way that is appropriate to the activities of your organization and hence the impact of timeliness of incident resolution.)
If you operate as a business unit, then the model is different, as we discussed above. Your optimum help desk team size will be determined by the parameters of your charging arrangement and any penalties – or rewards – arising through the operation of the SLAs. For that team size to also deliver optimum results for the organization points to the greatest challenge in devising SLAs in the first place, which is to contruct SLAs to truly reflect the organization’s needs. We’ll think about that in another post to this blog.