Service Desk Escalation Rates

[tombradley]

Hi all,

We're going through some changes here and will be implementing a level 1 service desk to provide technical support to external customers.

In our planning, we are discussing what we should target as an escalation rate from level 1 to level 2. Does anyone have any general direction on what industry best practices are in this area? Can we expect 40% of the tickets to be escalated to level 2, or 10%. We previously functioned as a "flat" organization so I do not have historical metrics to help with this.

Any direction is greatly appeciated.

Thanks,
Tom

[UKVIKING]

Direction huh.

follow the yellow brick road
_________________
John Hardesty
ITSM Manager's Certificate (Red Badge)

Change Management is POWER & CONTROL. /....evil laughter

[Diarmid]

I'm going to answer, but, let me emphasise, I prefer John's answer.

What you are describing as "escalation rate" sounds very similar to what many people refer to as "fix on first call", or "fix at level one" if you like. I'm not making a fuss about your terminology. It is fine, but I think others will recognize other ways of describing it.

Okay. Level one can resolve calls that come within its capability (skills, experience, available time, etc.). So the more finely you tune (including the arrival rate) level one staff to the calls that will come in, the higher you can expect the fix rate.

You shouild be able to see from this that the idea of an "industry standard", or even an "industry ballpark" here is absurd.

There are two approaches:

1.
a) decide on a target figure
b) run the desk with increasing numbers and skills until you reach your target.

2.
a) populate level one with the skills and experience you can afford, making sure you have the numbers to field expected call rates even at peak.
b) measure the fix rate over a period.
c) analyse the nature of calls and the way the staff and the system are working; make some improvements based on this and measure again
d) use this measure as your baseline for the future and, from time to time, look at ways of improving the figures, always remembering that adding more, higher skilled, better paid staff ought to improve things, but at a cost.

3.
a) use mixed skills (rotating staff) until you understand the nature of calls you can expect.
b) decide what is practical for level one and estimate what proportion this will be.
c) now populate with staff trained to meet your requirements.
d) review and tweak until either you are meeting your estimated rate, or you change it to meet what you can achieve.
e) place the operation in a quality audit and continuous improvement environment.

I wouldn't do any of this. I wouldn't even bother measuring the rate at all. I would identify call types as suitable or not suitable for different levels and areras of support and manage the function to ensure that they all (99.5% anyway) get done at the right point and I would want to know why if that wasn't happening.

Some service environments function better with less than 10% first fix, others with at least 90%. If all your calls are going to be prettty much exactly the same, then it is easy to have staff geared up to resolving them there and then. If all your calls are totally different, then it is pretty hard to expect the person answering the phone to know what to do most of the time. Some service environments (most or all) change in accord with the business cycle, in response to the level of change activity, the experience of the users, the maturity of the systems, and a few hundred other things.

Now. Where's that yellow brick road gone?
_________________
"Method goes far to prevent trouble in business: for it makes the task easy, hinders confusion, saves abundance of time, and instructs those that have business depending, both what to do and what to hope."
William Penn 1644-1718