Talk to Me: Reporting on Service Levels

In the final analysis, SLM is all about good communication -- between the IT department, end users and the company's business units. When IT actively measures its service delivery, it must communicate these statistics to customers using reports that are appropriate to the audience, easy to understand, and that clearly address areas where the customer reading the report has a vested interest. If service lags, even proactively delivering less-than-favorable news about service puts IT in a better light with customers than letting customers draw their own, less-than-favorable conclusions.

Starting at the top

IT should keep in mind one thing when tailoring reports for executive managers: they have no time!

Reports should compare service levels achieved overall with service-level objectives. They should be very concise and to the point, and should make use of visual aids like graphs and charts to further speed comprehension.

Because of the scope of executive managers' responsibilities, they're interested in looking at the big picture -- how is the IT department adding value to the company and contributing to business success? How are service levels improving employee productivity? Conversely, if there were outages or performance slowdowns, these should be reported in terms of real costs as well as lost opportunity in revenue and productivity.

Talking business

If the audience for a service-level report is a particular business group, that report should show how IT's service increased business opportunities for that group. These reports should relate to:

• transaction volume
• personnel productivity
• (if possible) customer satisfaction

Of these, customer satisfaction is the most subjective measure, and probably the toughest to track. However, there is value in conducting regular customer surveys to chart satisfaction levels and highlight areas where improvement's needed as well as reinforcing IT's commitment to customer service. And to help IT realize its commitment to quality, IT managers should all receive copies of whatever reports are given to the business groups.

Outside, looking in

If an IT department is providing any services to customers outside the company, it should provide those customers with summarized reports on service delivery. These reports should also cite any steps it takes to improve customer service.

Types of reports

Just as the audience for certain information affects the format of service-level reports, different reports may be required to cover different aspects of service quality and to satisfy the interests and focus of various audiences.

Service availability reporting

When reporting on availability, IT should map the actual measures against its availability objectives. The report should show availability by service levels or application instead of system components and should reflect the end-user experience by organization, location and line of business.

Tracking performance

To report on system performance, IT must relate performance to the end-user experience by specifying online transaction responsiveness and how quickly batch jobs are completed (turnaround). And the report should show system responsiveness by application, user group, location and line of business.

IT can gain an even better idea of how overall system performance is affected by specific system components by tracking response times and propagation delays by each technology layer (e.g., network, middleware, database) and correlating these measurements against overall end-user response times.

How much workload

To provide senior management and the lines of business with a better understanding about the value IT provides, it should provide information about workload volumes in terms of business transactions. This way, IT and non-IT departments can "speak the same language." When IT reports on workload external to the IT department, the business transaction volume should be shown by user group, location and line of business.

Security

Any breach of system security, or attempted breach of security, should be reported to the lines of business. This includes reporting on any virus infections and how these impacted the system.

Recoveries

When outages do occur, it's important to plan how to prevent another. That's why IT should report on both how long it took to recover from an outage and how it was done as well as how IT could implement procedures to avoid future outages. If these procedures would require additional staff time or new technology, the cost for these should be compared with the cost of the outage in the report so that the lines of business can make informed decisions for the future.

Allocating costs

If the IT department charges back the costs of its services to the lines of business, it should provide reports outlining:

• how IT calculated the charges
• total charges as calculated by this method
• how IT allocated costs to the lines of business
• calculated cost for each business line

IT can either distribute to its business customers a single, master report containing this information for all the lines of business or create individualized cost allocation reports for each business line.

How often to report

The audience as well as the detail of information contained in IT reports should determine how often it generates any given report. IT should provide its customers with reports on its service levels daily, weekly, monthly and quarterly.

Reporting in real time

Real-time reports on IT's service levels as well as proactive notification of problems increases customers' confidence and flexibility when outages or slowdowns do occur. When IT communicates information about outages, it should show which end users are affected as well as applications, locations and lines of business that are affected. It should also show the nature of the problem and its symptoms along with when service is anticipated to return to normal.

And now for the bad news …

In this 24x7 Internet world, scheduling any downtime for maintenance or system upgrades is becoming more difficult and impacts more and more users. That's why it's important that IT alert its customers about any planned downtime and should include details such as which users will be affected and the reason for the disruption in the alert. Along with planned service slowdowns inevitably come the unplanned ones, and it's important that IT report on these as well. This may not be easy, since tracing performance degradations can be complex, but in the end, clear reporting will reduce calls to the help desk and reports of problems.

Other realities of today's global Internet that affects service are the threat of heavy attacks on system security or spreading viruses. When these scenarios occur, IT should immediately alert its users to reduce the damage that can be done. Since averting system damage can also cause significant slowdowns in the enterprise, IT should apprise users about the nature of the attack, what it's doing about the problem and how long the system might be slow or unavailable before returning to normal. In the case of viruses, IT should also notify its customers about how to recover any lost or corrupted data and how to avoid being victimized by any more viruses of that type.

home