To print: Click here or Select File and then Print from your browser's menu

This story was printed from silicon.com, located at http://www.silicon.com/

Story URL: http://management.silicon.com/itdirector/0,39024673,39256349,00.htm

The Naked CIO: Service level disagreements
SLAs - not worth the paper they're written on?

By Naked CIO

Published: Monday 07 July 2008

Service level agreements are great at perpetuating mediocrity and blocking excellence. And that's exactly why some fresh thinking is required, says the Naked CIO.

Service level agreements (SLAs) are now an antiquated way of managing IT standards and support.

Exclusive column: The Naked CIO

See what this CIO really thinks…

The Naked CIO: What makes a great IT leader?

The Naked CIO: Business misintelligence

The Naked CIO: Price of panic

The Naked CIO: The skills drain needs fixing

The Naked CIO: Madness in the method

In most cases SLAs are not worth the paper they're written on. That's because companies often fail to make the terms enforceable, should standards not be met.

These agreements are becoming more and more irrelevant and less and less effective in assisting modern IT organisations. What's more, against the backdrop of ever-increasing offshoring, these agreements contain the potential for disaster.

In an industry trying to improve overall service, SLAs provide a licence to adopt a lowest-common-denominator mentality that acts against the concept of exceeding customer satisfaction.

They are increasingly being used to justify poor service: the lowest standard has been met, so why should we try any harder?

We must try and find another mechanism for internal and external service delivery. I rather like the concept of matrix-driven SLAs or introducing a risk-and-reward element to service level standards.

This approach revolves around compensation, increasing or decreasing according to how well customers are satisfied, measured by a number of service indicators.

The difficulty with this type of agreement is getting external vendors to agree to risk elements that are hardly ever covered by their business model and the complexity of identifying multiple degrees and indicators of service that is often difficult to measure.

While I may not have all the answers I am increasingly frustrated by the dilution of service standards that derives from lawyers rather than IT professionals negotiating complicated contracts.

The end product of these negotiations is not related to realistic services but rather to a reduction of exposure for outside vendors.

As much as I would like to rid myself of the SLAs in my organisation I suspect they are here to stay - at least until we can move over to a more effective service contract formula.

So if you're managing SLAs, try to ensure they are enforceable and you have some form of comeback if your supplier lets you down.

Additionally, try to set the benchmark for service at as high a level as possible. Of course, agreements need to have minimums - whether they are response times and other standard finite measurements.

But if possible also try to include other indicators, such as customer survey responses or some sort of satisfaction indicator as a benchmark.

These indicators will prevent the agreement being just about some obscure statistic that does not tie in to the ability to improve service.

Finally, if you really want to breed customer service in your department, assess whether you can manage the service yourself. In my experience external companies never drive customer satisfaction to the level that your own staff can.