Four Basic Measurements
When negotiating a new outsourcing agreement, clients face the challenge of determining the service levels that are most meaningful to the business. The intent of a service level agreement (SLA) is to measure the provider’s overall performance by virtue of concise, unambiguous metrics with targeted levels of performance that are easily understandable by the client community and are simple to validate from a client’s perspective. As the outsourcing industry has matured, providers have developed a multitude of service level measures they can propose to their clients. Some are more relevant to the client’s business than others. Without the proper alignment of IT service levels to the needs of the clients’ business, companies can fall into the trap of “seeing green but feeling red,” meaning the service level measures are exceeding their targeted performance levels yet there are still IT delivery issues. Fortunately, there are several common service levels within the outsourcing marketplace that align nicely to the perception of lines of business and end users. The following four metrics serve as a guideline for defining your service level requirements.
Service Desk – In many companies the service desk is the primary touch point into IT, and therefore measuring its performance aligns nicely to the business perception of IT. For example:
First call resolution – Directly ties to the user base experience of how well the service desk is equipped to solve their problems during the initial call.
Abandonment rate – A low abandonment rate indicates users are not getting frustrated waiting for a live agent.
End-user satisfaction surveys – Collects direct user feedback on their satisfaction with help desk services.
Install/move/add/change/delete (IMAC-D) requests – Measures how well the IT organization provisions requests from the business community.
Other service desk measures, while important to the IT group, may not be as relevant from a business perspective.
Projects – Project work performed by IT is usually the work most aligned and most visible to the business. There are many types of project-related metrics, and the following is a good subset to use to communicate project performance to the business:
On-time milestone completion – The project manager, working closely with his/her business counterparts, should develop a set of key milestones as part of the overall project plan. Measuring the on-time completion of these milestones communicates the progress being made and the maturity of the project management discipline being used.
Estimating accuracy – Measuring the accuracy of estimates that are provided to the business, especially when you can show sustained improvement over time, is a great way to build credibility with the business. The estimations could be in terms of schedule, cost and/or resource utilization.
Percent of budget/cost spent on strategic projects – This is an excellent measure to communicate how IT is driving down the “lights on” costs and reallocating to work that adds tangible business value.
There may be other metrics that better reflect your environment and what’s important to your business partners. The important point is to somehow showcase the value that IT is providing towards the execution of projects that are adding business value.
Change Management – Measuring and reporting the volume and success of changes to the environment is a good way to showcase the volume of work being done by IT “behind the curtain” and to illustrate how much goes right. This can provide good “air cover” whenever a delivery issue does occur that causes pain to the business. For example:
First time successful changes – Measures the percentage of changes that are correctly implemented the first time.
Percentage of non-emergency changes – Measures overall system stability and the maturity of the organization’s change management processes.
On-time change implementation – Measures how well IT activities are planned in advance.
Change management SLAs are effective in measuring change management effectiveness and efficiency.
General – Several other measures are closely aligned with business perception. Among these are:
On-time reporting – Many business units rely on the on-time delivery of accurate reports. This can be measured by identifying the list of critical reports and defining the time at which they must be completed (and in some cases, delivered).
Problem resolution – Ironically some providers will initially balk at this, stating that there is too much out of their control to commit to targeted resolution time-frames. However many will eventually agree to resolving a certain percentage of problems within a defined time frame. This is obviously one of the most visible signs of IT performance, and is important to demonstrate that even though problems are bound to occur, they can be quickly resolved due to the resources, tools and architecture in place.
Application availability – Most providers will supply a standard service level called application availability, so the important thing is to ensure that it is a true end-to-end availability measurement that reflects the users’ experience. In other words, the metric should comprehend any IT issue that results in the application not being available to use as planned. This includes not only issues with the application itself but also the entire underlying infrastructure including the servers, databases, network and devices used to deliver the applications to the desktop.
There are literally hundreds of “typical” IT metrics that can be reported on. While true that some IT-specific metrics should be in place (especially to ensure that issues are identified and resolved prior to affecting the business community), most should have a focus squarely on measuring the delivery of services to the end users. This list, although basic in nature, provides a good framework to evaluate the service levels you are currently using or are developing as part of a negotiation.