Network Fault Management: Beyond MTTR

Service disruption, billing errors and unresponsive customer service drive customers away. But now carriers are recognizing that network-performance issues are the critical determinant of customer satisfaction. In doing so, they realize they must focus on network management and measure end-to-end performance in terms of time, quality and cost. Typically, carriers do this by monitoring:

* Mean time to restore (MTTR), which is the total elapsed time from a customer-reported fault to service restoration * Total average cost to restore network faults

* The number or percent of repeat failures within a defined time frame.

However, rapid response to service disruption and monitoring network-performance data is not enough. As a carrier, you have strategic considerations that require a broader perspective on network fault management. Network-performance measures belong in a framework of key business indicators, along with customer-satisfaction results and new-product introduction metrics, that become a part of the regular score-card for top management's review and action planning.

Research shows that many carriers need to overcome a common set of operational problems with respect to data integrity and cross-functional cooperation before top executives can use network-performance measures as a key indicator for success. There is a basic set of issues related to data integrity and cross-functional cooperation that are table stakes in network management. The final step of network management is putting the data to use effectively.

Network-performance data needs to be credible and relevant. Data-integrity problems reduce the reliability of network-performance measures. Many carriers have similar data-integrity problems:

* The customer-service IT platform is not integrated with the network-management trouble- or fault-reporting platform. This impedes the effective monitoring of customer-focused MTTR because you cannot capture the full-time cycle from when the customer reports a fault to service restoration. Also, you cannot forward the salient details of the customer report in adequate detail to the technical staff.

* Trouble tickets are not fully completed on a consistent basis or promptly closed out, which leads to a lack of confidence in the MTTR statistics.

* Lack of accurate activity-based costs across operations, materials management and engineering leads to a lack of confidence in the average cost statistics.

Unless these data-integrity issues are addressed, there is a natural reluctance to publish network-performance results, and executives will doubt the validi ty of the underlying data. Therefore, you don't get an accurate view of overall network performance.

Although data-integrity issues are a basic requirement, given the nature of network faults or troubles, you also need a customer-focused, cross-functional view of network management to optimize performance. There are common cross-functional issues you should follow. Technical training needs to be in sync with technology introductions. A lack of standardized troubleshooting methods across divisions and regions can lead to missed opportunities in applying best practices in fault management. Comprehensive planning, including a back-up plan, for moves, adds and changes to the network will help avoid unnecessary network troubles. Automatic feedback to customer service about the progress of network-problem restoration is a critical issue.

Failure to address these issues results in cross-departmental finger pointing and stands in the way of an open assessment of network-performance results. You need to resolve the data-integrity and operational issues before technology management will have the confidence to share network-performance measures and results. In the end, progressive organizations will go beyond the basics to establish capabilities that can lead to improved root-cause analysis and a base for continuous improvement.

BEYOND THE BASICSSub-process performance measurement, cross-departmental service-level agreements and reporting capabilities of the trouble-report system are just three of the capabilities that go beyond the basics in establishing the foundation for pro-active fault management.

Measuring sub-process performance in fault management is necessary to fine-tune root-cause analysis. Typically, the sub-processes are defined as monitoring/surveillance, diagnosis, dispatch and service restoration. To facilitate root-cause analysis and provide the framework for assigning performance targets, you need the ability to measure time cycle, cost and quality for each of these sub-processes, and define and capture relevant data.

Designing the trouble ticket is key in creating a database for learning. Trouble tickets need to capture the type of fault and equipment in a succinct coding system that facilitates accurate completion of the tickets and generates management reports on a monthly or quarterly basis. Summary reports on fault and equipment type can be instrumental in providing hard data for feedback to equipment vendors and making decisions on equipment upgrades.

Finally, developing cross-departmental and key supplier service-level agreements can increase understanding of the interconnected nature of network-service-restoration requirements and give you a yardstick to measure response.

Taking a process view of fault management can help you better understand and assess performance. This view can lead to a better understanding of cross-departmental interdependencies and customer focus. It also establishes a framework for cross-functional goal alignment and resource allocation.

Many companies are realizing that you cannot find the leading indicators of network performance in technical and financial data alone. You have to track non-financial measures, create the necessary information architecture and acquire the management discipline to take action on the results.

Only when your company has the ability to improve network performance continuously can you argue that you are taking appropriate action to minimize network-failure impact to customers. Only then can you really begin to position the network as a competitive advantage.

Taking a process view of fault management can help you better understand and assess the impact of network faults. Although a technology manager often is the prime executive accountable for network performance, to achieve the required focus on customer impact and make decisions on resource allocation, the organization's top executive team needs to be reviewing relevant network-performance data and measures regularly.