Service Desk Tension Metrics

Sets of best practices for service management – like ITIL for example – are full of good ideas and good advice, and all that good information is valuable, but sometimes you are able to generate added value by combining two elements of advice from different parts of the guidance.

If you look in ITIL^® 2011’s Service Operations book (Chapter 6.3.5), you’ll see a list of relevant metrics about service desk/help desk performance. Now go to ITIL’s Continual Service Improvement (CSI) book (Chapter 5.5.2) and you’ll find some words about tension metrics, which are different metrics that effectively compete with each other. Each book has some pretty good stuff on their own, but put them together and you really start getting somewhere!

Metrics, of course, are just things you can measure to give you an idea of how well something is performing. We use metrics every day – for example we might measure a car by how fast it goes or how much fuel it uses to cover 100km. This is, in fact, tension metrics. The faster your car, the more fuel it is likely to use. You might choose to drive slowly to save fuel, or quickly to save time. You won’t be able to do both because the two compete with each other – there is ‘tension’ between them.

---

Metrics for a Service Desk

If you focus only on one metric, it is usually simple enough to meet, because you are basically disregarding all other factors.  So that may not be the perfect scenario.

Let’s take a look at two traditional metrics that are used to measure service desk performance:

• First-time fix rate. Easy to meet – just keep callers hanging on for as long as it takes.
• Short call duration. Just as easy – take their number, hang up, call back, and disturb them each and every time you need more information.

If you judge a service desk simply on one of the above metrics, you’re going to get behavior guaranteed to upset rather than delight customers. What the customers want is a balance of first-time fixing in a sensible amount of time. Just like you don’t want the world’s fastest car (way too expensive) nor the most fuel efficient (far too small and slow) for everyday motoring. You want something fast enough and cheap enough to run – along with other factors too, of course, like enough seats and luggage space.

So with the service desk, both first-time fix rate and short call duration are important, and both need to be measured (as do even more metrics besides those). This, however, then sets up a challenge that requires thought and consideration by service desk staff and managers. Do you keep the caller hanging on while you try a quick solution? Do you give up on the first-time fix rather than extend the call too long?  It makes the service desk work harder on decision making, but also delivers a much better and balanced service overall.

This is just an overly simple illustration of course. In real life you would want to choose more than just two metrics that fight and challenge each other – making the decisions even more complex but hopefully also making the service better matched to the client’s concerns, wishes, character, and culture. There isn’t a single right answer, but here is my list of optimal service desk metrics – as an opening or food for thought:

• First time resolution rate
• Average call duration
• Percentage and number of abandoned calls
• Percentage and number of calls miss-assigned (e.g. to wrong support team)
• User and customer satisfaction survey results (that’s two different surveys)
• Staff turnover
• Total cost of operation: staff, equipment, resources, etc.

It’s not easy to measure everything on this list, and likely you don’t need to measure it all, but surely you would benefit from measuring across the implicit tension that exists between many of these measurements.

Some of these suggested metrics might surprise you – staff turnover for example. Each new worker on the desk can take days or even weeks to reach a reasonable operating performance level, while you are paying them; so keeping the turnover down, keeps costs down. Of course, it also means more experienced staff, and that experience should deliver faster and better service.

Using Metrics to Measure Value

No one set of metrics is right for every situation – sometimes speed matters more, sometimes economy, sometimes safety. That applies to cars or service desks. What matters is measuring the value that the customers get from the service. To do that for your particular situation, the first step is to establish what those customers would see as value. Start with possible metrics and discuss with the customers to work out the best bundle of measurements for your situation.

What is universal though is the need to try and make sure you have that tension element in your chosen set of metrics, so that the compromise between speed and cost – or whatever matters to your situation – is rewarded rather than just one target that can be easily reached at the expense of overall service quality.

We all use the tension metric concept in everyday life (as with the car). You probably also do so at work, perhaps without even realizing it. Can you think of any examples? Please share.

Image credit