.gif)
IMS CUSTOMER EXPERIENCE - Going to the next level
The many bells and whistles promised by IMS make it essential for operators to understand and monitor all the device-types used by customers, if they are to ensure high standards of customer experience, says Matt Herdlein
As emerging IMS platforms open the doors to real-time, interactive multimedia services, taking care of the customer experience becomes an even more critical ingredient for achieving success. Investments made to support the emerging service complexity could be wasted if customers cannot derive the intended value. Moreover, the assessment of service quality would be misleading unless the actual performance of user devices is considered as well. As services become more sophisticated and complex, more functionality and features are migrating to user devices, thus making them an integral element in the overall service quality equation.
With 3G handsets based on open operating systems, the numbers of both device makers and third-party application vendors have skyrocketed. Handango, a leading supplier of applications for handsets and Personal Digital Assistants (PDAs), reported more than 11,000 new applications in 2005 from more than 1,200 new vendors, and “type approval” has moved to vendor certification. The GSM Suppliers Association reported that in 2006 there were 212 GSM/EDGE terminal devices available in the market from 33 vendors. Handset operating systems come from companies such as Microsoft, Symbian, and Qualcomm, among many others.
In this environment, the challenge for operators is clear. They must be able to deploy an enormous and quickly growing range of services on a host of intelligent devices, and ensure that those services operate successfully on each device. In short, operators need to augment the scope of service quality to include user device performance to better assess the actual customer experience derived from their IMS investments.
To further understand the problem, consider the following simple case: if a new service fails 90 per cent of the time on a user device that serves 10 per cent of the market, then an analysis of the service will only show a one per cent failure rate. However, the reality is that 10 per cent of the customers are unhappy and may move to another operator or stop using the service.
MobileGuru, a UK company that sells mobile phones and accessories, compared handset performance on a UK network and found that call drop rates can vary from about two per cent for the best performing devices to nearly 10 per cent for the worst performing devices. Even then, within a single device type, there may be significant variations in performance caused by batch problems in manufacturing, user configuration errors, or software download problems.
These issues are not new. GSM operators faced the dilemma of whether to issue recalls or modify their networks in the mid-nineties, when two of the leading handset vendors were found to have compatibility issues with their networks. At that time, they had to make software changes under controlled conditions since type approval would be invalidated if the changes were done incorrectly. The recall was avoided in those cases, but smaller manufacturers did have to recall devices.
Identifying the problem
The advent of Universal Serial Bus (USB), and changes in component prices, along with the relaxation of type approval, made home-based upgrades to handset software more common. A handset's International Mobile Equipment Identity (IMEI) can be used to identify the model and even the place of manufacture, but it is no longer a reliable indicator of software build or application set.
Further complicating the matter is the fact that handset operating systems are available to suit the preferences of any manufacturer or vendor. If you like Java, try Savaje; if you prefer Linux, then look at MontaVisto; or if you are Microsoft fan, there is a version of Windows available. The market leader, Symbian, grew out of UK PDA innovator Psion, and if you don't like the Symbian software, then Nokia supplies Series 60. Handango reports having more than 190,000 titles from 16,000 content partners supporting nine different handset operating systems.
Vodafone has reported a significant linkage between the rate of churn for residential customers and the numbers of services they use on a regular basis – the more services used, the more likely customers will stay. Of course, services will only be used if they work reliably.
The cost of unreliable service to operators can be measured not just in lost revenue but also in lost handset investment. Nokia, which supplies one in three of the world's handsets, reported that the average selling price of its handsets is US$125 (103EUR), while Sony Ericsson, which has more high end products, has an average selling price of $180 (149EUR). Some retailers in the UK are offering all Nokia handsets for free when bundled with a post-paid tariff, so the cost to the operator is around $146 (120EUR) for each handset.
The abundance of software and device options presents a number of challenges for operators. It also provides a unique opportunity for operators to better serve valuable customers. When root causes of problems can be traced to individual customers or device types, operators can develop new application versions or make changes to operating systems to correct the problems.
Consider, for example, the value to enterprise customers who, according to Yankee Group, make up 28 per cent of mobile operators' revenue. These customers can be advised on which mobile phones perform best for their services or can be given upgrades. Customer satisfaction would increase, as would retention levels for a valuable market segment.
Inspecting the packets
To monitor service quality, it has always been imperative that operators have ready access to reliable data. For voice calls, there are many options, from Call Detail Records (CDRs) to signalling probes. But for data services, and to support the move to IMS, more sophisticated, deep packet inspection probes are required.
Data service quality is determined by three main variables: network performance, device performance, and portal performance. Degradation in any of these variables will result in a poor customer experience.
Any effective analytical tools should permit early identification of trends so that solutions can be developed and customers informed before service is affected. Hotspots may occur when changes are made to services, access networks, or core networks, or when handset operating system upgrades are introduced, but these may be difficult to identify among millions of users, and operators must take care that “normal” user actions are not misinterpreted.
For example, with voice, a short call (one quickly terminated by the user) may raise no alarms but actually involve unacceptable voice quality, whereas with data, short sessions may be the result of high throughput, and long sessions may indicate problems.
By comparing different handset models running the same service, patterns can be established. Analysts should also consider whether particular models are only available to a limited user group, such as prepaid. Traditional service quality monitoring gets a view of specific services based on consolidated data extracted from the network, and by using field probes that perform synthetic transactions depicting various services and user behaviours. Although these approaches have merits, they fail to analyse service quality from actual service transactions that customers make. In other words, they fail to capture the trends and nuances of the true customer experience.
Human behaviour
It is, therefore, essential for operators to understand and constantly monitor the “behaviour” of all the various device-types used by their customers. Specifically, it is important to understand service performance by device-type as well as by specific device configuration, to understand how customers are affected by network or device issues.
Some of the typical issues that mobile operators face on a daily basis are:
• How to choose a device/s for a new service
• During trials, how to measure device performance by type, service, and configuration
• How to view device performance to identify service bottlenecks before they can affect the service, and how to identify affected (and potentially affected) customers
• How to know if a new configuration or update is performing as expected
• How to identify devices with high support overhead.
To answer these questions, operators need to go beyond probes that make educated guesses about performance based on small, “synthetic” samples. Operators need a solution that aggregates actual device performance around the clock from every service transaction.
Back to reality
The reality is that virtually every mobile operator supports dozens of device-types and millions of active devices. That means, to manage service quality and customer experience, operators must do more than monitor their networks and operations. They have to know, at any given time, the capabilities and limitations of all of their user devices, how those devices are performing, how they will handle new service offerings, and how customers are using them.
When operators have access to this level of user device performance intelligence, the business benefits are invaluable. For example, operators can: “see” how customers are reacting to marketing campaigns and special offers; recommend the best devices for customers when orders for new or expanded services are received; respond to customer calls with a holistic understanding of the customers' experience; offer targeted promotions to individuals or groups; provide incentives to device vendors based on verifiable performance and offer more focused SLAs.
As the choices for new and more complex services continues to grow, and user devices that offer more functionality emerge, mobile operators need to understand service quality from the device perspective. Operators can expand their traditional service monitoring arsenals and realise the business benefits that can result from higher customer satisfaction.
Matt Herdlein is Executive Director, Service Management, Telcordia
Share this article with others
Post to del.icio.us
Printed from http://www.eurocomms.com/features/111793/IMS_CUSTOMER_EXPERIENCE_-_Going_to_the_next_level.html
Comment on this article
Skip to comments
We encourage users to analyse, comment on and even challenge European Communications's articles, including the one above - 'IMS CUSTOMER EXPERIENCE - Going to the next level'
User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site.
Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. We will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site.