Litmus Test

Comment

Digg

Del.icio.us

Quality of services is emerging as a major issue for telecom service providers. With an increasing number of subscribers and limited spectrum, it is a challenge for the operators to maintain the basic quality of services. The end consumer is constantly complaining of call drops and blocked calls.

In this background, for the first time in India, Nielsen brings a survey which tests and reports the performance and quality of networks of the top six operators in different markets in the country. The survey was conducted in a mobile van. The Nielsen India Consumer Experience Mobile Survey measures wireless network performance for all national and selected regional carriers in markets covering a population of 84 mn people in India.

Interestingly, the perception that the Indian customer is extremely price-conscious has been proven wrong, with around 42% of subscribers rating network performance as an important criteria for selecting their mobile operator. It is in fact ahead of the price-driven choice, which stands at 34%. After initial success, satisfaction with network performance is the single most important driver for shaping overall satisfaction with the operator. Clearly, the operators need to ensure loyalty of their customers instead of focusing on tariff war.

The network quality of operators is measured on three parameters-reliability, coverage, and audio quality. The benchmarking study shows that most cities have a clear network quality leader and there is a significant variation in network performance across individual Indian cities. However, Nielsen refused to share any data related to the operators performance.

According to the report, Indian mobile subscribers in Panipat, Indore, Amritsar, Vadodara, Ernakulum, and Meerut enjoy the best voice networks in the country. Surprisingly, the list does not include any of the metros. Cities with top 10 voice networks have an average 0.62% block call rate; and an average 0.81% drop call rate. In the same vein, Nasik, Varanasi, Pune, Coimbatore and Chennai have the worst voice networks in the country.

The survey reports that Ahmedabad experiences the minimum average drop rate of 0.9%, with Delhi following closely at 1.5% (See Chart). There's only one out of six tier-1 cities, and fourteen out of twenty-seven tier-2 cities where all six operators tested have an average drop call rate of less than 2%.

Delhi again comes in the limelight by recording the minimum average block rate of 0.71%, followed by Ahmedabad at 0.73%. Chennai experiences the maximum average block rate among the metros (See Chart). According to the survey, there are four out of six tier-1 cities and nineteen out of twenty-seven tier-2 cities where all six operators tested have an average block call rate of less than 2%.

The competitive network performance data for six out of eight tier-1 markets tested indicate that the average call block rate is 1% and the average call drop rate is 2.14% among these metros.

Methodology

For this study, Nielsen partnered with Sasken (a company specializing in providing engineering services to network OEMs) for the delivery function including-data collection, field operations, post processing and QC, and report generation. .

Nielsen-Sasken partnership used Invex3G benchmarking tool from Andrew Corporation, capable of multiple phone interfaces, for network benchmarking. The handsets/devices were placed inside isolation chambers in order to minimize/mitigate any type of external interference to the devices. Each handset was connected to an external, roof-mounted antenna. The antennae were placed at the minimum distance in order to minimize the interference between the handsets.

The data collection was done at the mobile end and was aimed at assessing the network performance from the subscriber's perspective. Testing was based on mobile originated voice calls and mobile originated data calls.

The data for this analysis was captured by a fleet of four vans, which log around 1,80,000 calls each year. The data is integrated with Nielsen's consumer research to provide an understanding and measurement of mobile consumers.

The following handsets/devices were used for testing:

Voice
Technology	Phone/Device
CDMA	Kyocera KX2 / 2235
GSM	Motorola V557 / V 547/ V 551, Samsung Z 560
Data
Technology	Phone/Device
1xRTT	Kyocera KX2
GPRS/EDGE	Motorola V557 / V 547/ V 551, Samsung Z 560

Voice services benchmark test scenarios comprised mobile originated calls of 90 seconds (8 MOS/ASR cycles as defined in Invex3G) that were terminated at a landline system. MOS audio quality testing using PESQ algorithm for Uplink and Downlink MOS measurements were conducted. All testings were conducted in a mobile environment.

The landline system (fixed end system) was centrally located at Bangalore. For the NLD backbone carrying the calls to the fixed end server, the following fixed line providers were used; this is based on good connectivity/POI between the operator and the NLD provider or whoever has a broad presence throughout the country.

Operator Tested	NLD Provider
RCOM	RCOM
BSNL	BSNL
Airtel	Airtel
TATA	BSNL
Voda	BSNL
Spice	BSNL
IDEA	BSNL

�Opportunities lie for an operator to pull away from the competition and create a nationally or regionally differentiated network position in the market by using the competitive benchmarks to guide their investments,� says Shankari Panchapakesan, executive director, telecom practice, Nielsen.

The best practice drop and block rates from the US tier-1 (top 10) and tier-2 (top 35) markets have been compared with the weighted average drop and block rates for tier-1 and tier-2 markets in India in the report (see chart).

Interestingly, though US best practice shows a distinct edge over the Indian network performance, Indian mobile subscribers are relatively more satisfied with the network. This is mostly driven by the fact that an average Indian subscriber has gone from no phone to a cell phone and is less discerning and more tolerant. An Indian subscriber is less discerning because of the absence of any service/performance benchmarks. This lack of differentiation, clearly creates an exciting opportunity for the Indian operators. The report suggests that just as in the US, 'Network Leadership' will become the path for Indian operators to create differentiation and extend their brand strength to gain market share.

The Nielsen survey defines a 'successful call' as the one established and maintained for at least 90 seconds. The findings of the survey are based on an average of over 5000 calls per market in the top eight tier-1 cities and 2000 calls per market across the top thirty two tier-2 cities. This also takes into account the number of ineffective call attempts (in which a customer tries to make a call, but can't get on the network) and calls dropped within ninety seconds of being established on the network.

The survey will help the service providers to manage their performance besides taking corrective measures and making new investments. In the end, the program is designed to serve as an industry standard by which operators will benchmark themselves against their competitors.

Gagandeep Kaur
gagandeepk@cybermedia.co.in