The Service Desk Is Meeting Its SLA and Failing the Business
A common vocabulary, a consolidated desk and a green set of service-level reports are evidence of control. They are not evidence of value.
The Green Dashboard at Eight O’Clock
At eight o’clock on Monday morning, the service desk report is already circulating. Ninety-two per cent of calls were answered within thirty seconds. Eighty-seven per cent of incidents were closed inside target. The backlog is down for the third month. Every indicator is green.
At the same hour, a distribution centre is re-keying orders because the overnight interface failed. The incident was logged at 06:12, classified correctly and passed to the application support queue within nine minutes. Nothing in the service desk agreement requires anyone to ask whether the 10:00 despatch run is now at risk. The ticket is moving exactly as designed; the business is standing still.
This is the uncomfortable feature of the present enthusiasm for service management discipline. Across large organisations, ITIL language is becoming common, service desk tools are replacing informal help lines, and the new ISO/IEC 20000 standard is giving boards and suppliers a clearer badge of process maturity. Much of this is welcome. But the profession is in danger of confusing a better-managed support function with a better-served business.
The Metric Has Become the Service
The service desk was meant to provide a single point of contact and restore normal operation quickly. In too many implementations, that sensible purpose has been narrowed into a production line for tickets. We count speed of answer, first-time fix, age of open incidents and compliance with severity targets because the tool can report them cleanly. We then allow those measures to define what good service means.
The mechanism is simple:
- The desk is rewarded for accepting, classifying and routing work quickly.
- Resolver groups are rewarded for closing incidents within category-based targets.
- Suppliers are governed through the same targets because they are contractible and auditable.
- Business interruption that crosses several tickets, systems or suppliers has no single owner and therefore no single measure.
The result is local efficiency and end-to-end blindness. A user who cannot print may indeed have a printer incident. Fifty users who cannot print despatch notes before the carrier cut-off have a business service failure. The technical symptom is the same; the operational consequence is not.
A composite example makes the distinction plain. Consider an organisation supporting 1,800 desktops across twelve sites. During one month, the desk records forty-eight incidents against warehouse printing. Forty-four are closed within the four-hour target, so the reported achievement is 92 per cent. Yet three of the four misses occur between 14:30 and 16:00, when despatch notes must be produced for the final collection. Those three incidents delay 260 orders and require an evening shift to recover.
The monthly report says that printing service exceeded target. The operations director says it failed at the only moments that truly mattered. Both statements are factually defensible; only one is useful.
Discipline Is Necessary, but It Is Not Enough
The strongest defence of the present model deserves respect. Before disciplined service management, support was often opaque: users called favoured technicians, priorities were negotiated by persistence, recurring faults disappeared into personal notebooks, and suppliers blamed one another. Standard incident categories, escalation paths, configuration records and agreed service levels impose order on that world. Without them, “business alignment” can become an excuse for favouritism and constant interruption.
That defence is correct. The answer is not to discard ITIL discipline or return to heroic technicians. It is to complete the design.
The service desk must connect technical events to the operating rhythm of the enterprise. That requires three changes which are modest in concept but demanding in practice.
- Define services in business terms. “Network”, “server” and “desktop” describe technology towers. “Branch opening”, “order capture”, “despatch” and “month-end close” describe work the organisation recognises. Technical components still matter, but they sit beneath a service whose failure has an owner and a consequence.
- Make impact conditional on time and volume. A two-hour interruption at midnight is not necessarily equal to a twenty-minute interruption before a payment run. Priority rules should include the number of users affected, the business timetable and the cost of delay, not merely the component and the caller’s grade.
- Review service loss, not just ticket performance. The monthly meeting should examine the few interruptions that damaged operations even when every team met its target. This reveals the gaps between queues, contracts and component measures where the most expensive failures tend to hide.
A service level is valuable only when meeting it is a reliable sign that the business received the service it needed.
There is a practical test for whether this connection exists. Ask the service desk supervisor what must not fail between now and close of business. If the answer is a list of systems, the desk is still managing technology. If the answer includes the payroll submission, the branch opening window or the despatch cut-off—and if the queue changes accordingly—the desk has begun to manage service.
The Question Leaders Should Ask Now
In 2006, the service-management debate is moving from whether organisations need process discipline to how far that discipline should extend. The risk is that leaders declare victory too early. A common vocabulary, a consolidated desk and a green set of service-level reports are evidence of control. They are not evidence of value.
The next useful management question is therefore not, “Are we compliant with the process?” Nor is it, “Did IT meet the SLA?” It is: when the business most depended on a service, did our operating model recognise that moment and act differently because of it?
If the answer is no, the service desk is not yet the bridge between technology and the business. It is a well-run reception desk at the entrance to a building whose occupants are still finding their own way through the corridors.
The service desk should remain disciplined, measurable and consistent. But its success must be judged at the point where work is done, orders move, customers are served and obligations are met. Otherwise we will spend the next few years improving the accuracy with which IT reports that the business has been inconvenienced.