What is Service Level Management (SLM)?

According to Sturm, Morris and Jander in Foundations of Service Level Management, service level management (SLM) is "the disciplined, proactive methodology and procedures used to ensure that adequate levels of service are delivered to all IT users in accordance with business priorities and at acceptable cost." Effective SLM depends on IT or the service provider providing the particular service that the customer needs.

(Note: in this article we use IT to mean internal service provider; it could just as easily be an external service provider)

Why SLM?

Because of corporate growth and more competition, departments within corporations are learning how to function more like freestanding businesses. They must be accountable for managing their own budgets and for successfully providing goods or services to internal, or sometimes external, customers, just as businesses provide goods or services to consumers. In a business world where companies have a clear choice to outsource services to independent contractors, departments, including IT, can't just tell customers to "take it or leave it."

There are six compelling reasons to establish a SLM discipline within your company or with your service provider:

• Satisfying clients
• Managing expectations
• Regulating resources
• Marketing internal IT services
• Controlling costs
• Establishing a defensive strategy

Satisfying clients

Communication between internal customers and IT/service provider is essential, and SLM requires clear communication. IT/service provider must understand what the customer perceives as good service (e.g., email availability as opposed to availability of another application). The customer must understand what's reasonable to expect of IT given limitations in hardware, staffing, etc. Everyone must agree on benchmarks for what constitutes acceptable service against which service levels can be measured. When services miss the mark, it's apparent why customers aren't happy. When services meet the mark, customers can clearly see that their requirements are being met, and confidence in IT goes up.

It's only human to continually want more and better -- call it "expectation creep." Often, customers who were satisfied with service yesterday want better service today, and will want even better tomorrow. Or, customers may just want to maintain service levels -- despite the fact that more users are accessing the system, more applications being managed and new technology's being incorporated.

To avoid expectation creep, IT and its customers must negotiate a service level agreement (SLA) based on a given, clearly documented set of circumstances. If service levels erode, IT can point to changes in these circumstances as clear reasons why, and both parties can go back to the table to renegotiate the agreement.

No service provider, including IT, wants to assume the role of "resource policeman." But sometimes that's exactly what has to happen when an especially aggressive department demands services that tie up or limit IT resources available to other departments. In this case, having clear SLAs with all its service customers can provide IT with a strong argument against unreasonable demands from one of them. And there's another side of regulating resources: When IT monitors service levels closely, it can become aware of developing problems in overcapacity and lack of bandwidth, and it can take action before service degrades.

In the old days, the only contact between IT and internal customers happened when something went wrong. In its customers' eyes, IT was seen as a roadblock to achieving business goals, a necessary evil. However, when IT can document that it's providing good service to enhance business, customers can clearly see that it is an asset for the company.

Since outsourcing emerged as a clear alternative to in-house IT, management has been under more pressure than ever to deliver first-rate service. But without measurable agreements with those customers, how can managers decide what constitutes 'first-rate service'? In such cases, it's easy to indulge in "overkill": purchasing more hardware, upgrading systems, hiring more staff than necessary to meet what's perceived as the customers' needs. A good SLA might have clarified in which areas service did need improvement and in which areas service levels were satisfactory, so resources could be channeled accordingly and efficiently.

Conversely, an SLA might also help the customer and IT decide whether or not the investments necessary to upgrade service levels actually can be justified in business terms. Sometimes, compromise would be more beneficial to the company's bottom line in the long run.

There's an old saying that the best defense is a strong offense. IT departments that are fending off subjective, negative perceptions can make use of SLM to demonstrate in measurable terms that they are worth the company's investment. If SLAs are precise between IT and its customers, and if service levels are documented, facts can take the place of perceptions and can diffuse arguments for outsourcing. One more reason for SLM.

Not many years ago, gathering the data and generating the reports necessary for SLM was labor-intensive and slow, and not always very accurate. There were not many tools available for companies to be able to measure IT service delivery. In 1996, an Enterprise Management Associates survey of IT managers found only 12 products that were being used for service level management.

However, in May 1998, the number of companies that identified themselves as offering products to be used for service level management jumped to 62, and in March of 1999, the number increased to 89. Technology has made SLM reporting much easier, and it's likely that technology will continue to make it even more so. (See our vendors list.)

There's a strong case for implementing SLM, for the good of the IT department and the company. And now that technology is available to make efficient SLM measurement possible, what's left is to begin the careful process of mapping out and negotiating the SLM.

home