mobileSQUARED

Twenty five years on from the inception of the mobile phone and a major problem still plagues the industry: patchy network coverage. Consumers face a plethora of coverage black holes, dropped calls and frustratingly slow mobile Internet access. Even in advanced Western markets in-building coverage can be minimal and rural communities disconnected. If you thought we’d left coverage problems behind with the demise of the Rabbit phone – the 1980s system that only allowed users to make a mobile call when in sight of a ‘rabbit’ sign – think again. Indeed, of the mobile population polled for this month’s consumed many said that network coverage was a major bugbear. And consumers are becoming increasingly frustrated at the regularity of ending a mobile call with “Hello? Hello?”

“The coverage problem is enormous,” says Mike Flanagan, CTO of network planning company Arieso. “No operator in the world is relieved of this problem. What distinguishes one operator from another are the steps they take to quantify it and then improve coverage.”

Where networks are still growing in developing markets and customer acquisition is the primary operator motive, filling in the coverage gaps may not be a priority. But for consumers in highly mature saturated markets there’s only one question when it comes to poor coverage: why?

While not all blame can be laid at the doors of the operators – all radio networks suffer from coverage issues – they have it within their remit to solve the problem. Walls, hills, network capacity, and power all work to place inherent limits on where and when calls can be made. Factor data into the equation and the picture only becomes more complex. But for many consumers the problem appears to be getting worse as they attempt to access ever more sophisticated services and rely on their mobile phones for anywhere, anytime communications and mobile Internet surfing.

Coverage
Operators’ legal obligation, as set out by the UK telecoms regulator OFCOM, was to roll out 3G coverage to 80% of the population by 2008. In the US the Federal Communications Commission (FCC) has no such requirement at all, meaning that large population areas are often devoid any mobile broadband. In practice even where obligations exist, it means that consumers must be able to receive a signal if they’re standing outside their house. Hardly a realistic mandate when 70% of calls are made indoors.

“When they [operators] say 95% coverage, it doesn’t mean you get good coverage in 95% of the areas,” says Bernard Breton of network planning company Mentum. “It doesn’t mean you’re going to get a good call inside the building, or 95% coverage for data either.”

Coverage, Breton explains, is limited by radio propagation – the amount radio waves are able to penetrate walls, buildings and natural obstacles. In 2G networks, often at 900Mhz, radio propagates further and issues are generally limited to dense buildings or rural areas with a complete coverage blackout. Move to 3G, often at 1900MHz or 2100MHz, and the radio waves travel less far and are more sensitive to obstacles, creating a greater problem for the consumer on the street.

Operator strategies
Operators have reacted to the coverage problem by building more macro cell sites, often incorporated into street furniture in urban settings, at vast expense. Other vendors offer distributed antennae systems to improve in-building coverage, fibre to base stations to increase network capacity or handset side software. It seems almost inevitable that many operators will look to deploy 3G at 900Mhz to solve the problem.

It’s a scattergun industry strategy that has addressed only some of the coverage gaps and left consumers in limbo.

Moreover, as the industry increasingly moves to 3G, mobile cells begin to ‘shrink’ as more data is put through the system, creating an even less satisfactory experience for the consumer.

“Operators have rolled out these thin and crispy networks,” says Simon Shaw, chairman of the Femto Forum. “And coverage load has ramped up extremely quickly. Networks are not capacity constrained just yet but it will happen soon.”

Ironically, Shaw adds that the problem with 3G services is often not too little coverage, but too much. This is a major issue considering the recent explosion in dongle sales and introduction of HSDPA smartphones into the market. This vast growth in data input has put significant strain on operators’ networks – although none will admit it – resulting in a worsening consumer experience. It’s a problem that may deteriorate as consumers move to 3G, which will surpass total 2G subscriptions in the UK during 2009. While 4G LTE (Long Term Evolution) technology will increase network capacity from 2010, data throughput could also be exponentially higher.

The solution to the coverage problem depends on the strategy of each operator. UK operator 3, a Greenfield 3G operator with no 2G legacy infrastructure issues, has around twice the number of cell sites of rival UK carriers.

“3 built their network from scratch because of what they had – 2100Mhz,” said Ricky Watts of network planning company Aircom. While incumbent 2G operators “designed their networks from 900Mhz – and then added on 3G. That’s why you have these holes in coverage,” he added.

Smarter network planning, based on new technology that allows operators to track real-life consumer usage in real-time, and marry that to subscriber information can alleviate at least part of the coverage problem by building an ROI model for carriers’ infrastructure spend.

Even with new tools, operators are still faced with an infrastructure legacy built with expediency in mind. After all, carriers created their networks based on which sites were cheapest and where they could get planning permission for new base stations. Mapping future base-station placement based on actual use will – theoretically – offer consumers a better experience. At least if you’re trying to connect in a dense urban area where the economics of network build-out make sense.

It’s good to share
Where infrastructure build-out remains uneconomic in rural areas, network sharing has been mooted as an alternate route to fill out coverage gaps and save potentially millions in network CAPEX. In the UK, for example, operators 3 and T-Mobile agreed in late 2007 to share infrastructure planning and build-out under the Mobile Broadband Network Ltd (MBNL) banner. This followed a similar agreement between Orange and Vodafone UK.

The operators had even trialled technology with London Underground – subsequently rejected somewhat absurdly on grounds of cost – that would have provided talk and text coverage on the world’s oldest subway system, which is used by millions of commuters every day.

In Sweden, too, rural coverage is being expanded by two consortia formed of the leading four operators: TeliaSonera, Tele2, Telenor and 3 Sweden. In the majority of instances, all four operators share one mast. It’s a policy that benefits the rural consumer by offering the economies of scale that operators need to build greater network coverage. But this is not a pattern that is being repeated globally due to operators’ desire to remain operationally independent or suspicions of rivals’ strategies.

The real cost of churn
The inevitable risk for operators that fail to invest in network improvement is that poor coverage leads to churn. Mobile operator churn varies hugely around the world. After all while “operators always point to high satisfaction, and good coverage,” says Ben McCahill of Tellabs. “Consumers will point to the problems.”

Japanese operators – whom have invested heavily in network performance – can point to churn as low as 10%, with Europe and North America averaging around 20% according to mobile messaging company Acision. In APAC churn can be anything from 50-100% upwards, with Sri-Lanka sitting at the lower end and Indonesia experiencing over 100% churn. South American churn levels are around 40-50%.

“The reason for churn largely falls to two key causes,” says Mike Beech, vice president product management, intelligent charging, Acision. “The first being marketing pull, where attractive price and service propositions pull customers to new operators; the second is operational push, where failing networks and a negative customer experience is the root cause.”

And churn is highly expensive, Beech adds, because “in more developed markets, the cost is inevitably higher with operators bundling handsets free with price plans.”

Ovum’s Churn Management 2009 report puts the cost of replacing a customer at an eye watering US$200-300 per customer churn in developed markets such as Europe and North America.

The lesson being that those operators whom believe in quality of service as a point of differentiation will reduce their churn and therefore costs. It could even be argued that the growth in subscriber numbers at O2 UK, who are now the nation’s largest operator with 19.47 million consumers, for example, is in no small part down to the quality and (99% GSM / 80%+ 3G) coverage of their network. In the US Verizon Wireless’ marketing focuses heavily on network quality.

Femto
Some operators are looking to femtocell technology as an anti-churn tool. Femto offers consumers an improved 3G experience in the locality of their home but doubts remain about the economic viability of the technology. Femto will, after all, solve only one problem: coverage in the home. Will consumers buy a new box when they are happy to churn for free at the end of their contracts?

“The critical inflection point for mass consumer deployment is the cost of the box,” says Eric Org, vice president marketing, at US semiconductor firm BitWave. “[The femtocell] has to be sub-US$100 to be a mass deployment and until we get to a US$15 baseband chip and US$5 radio chip its hard to get below US$100.”

Question marks must also be raised about operators’ support structures for mass deployment of femto, Org adds. In a city the size of New York, for example, it could take literally hundreds of thousands of femtocells to have any kind of deep impact on the network.

But carriers may use femto as a targeted anti-churn tool; a sweetener to limited numbers of frustrated consumers who are ready to switch. In rural areas, “operators are spending US$200,000 on deploying a macrocell and the return isn’t enough,” Shaw says. Subsidising a sub-US$100 box would enable operators to reduce both churn and CAPEX.

Whether the solution to this very real consumer problem is femto, or new macrocells, or distributed antaenas, one thing is obvious – carrier processes don’t evolve quickly. In the short term, users shouldn’t be surprised if they still can’t receive services in their area when mobile^SQUARED next visits the topic.

In the meantime the best thing for consumers to do is complain and operators may hear. “Some people do that, but it’s a minority,” says Finnegan. Just might be best to use the fixed-line phone for the time being.