Building rural telecoms
Around 80 percent of the Bangladeshi population still lives in rural areas and contrary to perception, outgoing calls between urban and rural users are roughly the same.Photo: STAR
Telecoms industry analysts are obsessed with a four-letter abbreviation: ARPU. The great and the good are forever pontificating over how to arrest the decline of average revenue per user (ARPU). In Bangladesh's mobile telecoms industry with average ARPU at $3, the numbers are closely following India, which has the lowest ARPU in the world.
Industry executives worry less about ARPU and more about “usage”. They realise there are limits to what they can do to influence upwards the price of what they sell, and more to encourage consumption. As penetration passes 50 percent in Bangladesh, the dark arts of selling more minutes and driving more data through clever uses of bundling, packaging and marketing characterise the industry we now find ourselves in.
Across emerging markets, the lenses of price and usage for viewing the business opportunity have made rural customers in particular look less and less appealing. When it comes to rural Bangladesh, there certainly are some significant challenges:
* Forty-eight percent literacy rate: the lack of literacy dilutes campaign effectiveness in many media.
* Less than 1 percent English literacy: the low preference for English limits the uptake of most services.
* Lack of infrastructure: all-weather roads and electricity connections are not yet a common feature of many Bangladeshi villages: over 70 percent users rely on electric generators for their electricity needs at least once a day.
These features create common perceptions that rural users are likely to spend less money, not be interested in data services, receive more incoming calls than outgoing calls and cost more to acquire, connect and serve. Added to that, insight from one of the major Indian operators indicates that it costs 50 percent more to acquire a rural customer compared to an urban one due to higher distribution payouts.
Further, the average monthly servicing cost for a rural user is around 25 percent higher than for an urban one. At the outset, this does indicate adverse economics.
Yet the rural opportunity cannot be ignored and is potentially very significant -- around 80 percent of the Bangladeshi population still lives in rural areas and contrary to perception, outgoing calls between urban and rural users are roughly the same, meaning that rural users appear to be generating as much outgoing call revenue as urban. Recently a leading Indian TV channel recorded that in the state of Andhra Pradesh as much as 25 percent of interactive TV responses came from rural areas -- indicating non-urban consumers already form a sizable segment of engaged and interactive audiences.
There is more that can be done to nurture a profitable rural telecoms business, and to do so requires operators to take three steps:
* Understand the breadth of the customer base in rural areas better
* Address users' needs for services to be micro-local and in their language of comfort
* Develop service models fit for a varied user base
Understand the breadth of the customer base in rural areas better
Usage is the major marketing focus in Bangladesh, but in rural areas, adoption is in fact the bigger problem. Consider the data services experience in India: while the majority of smartphone users in urban areas access a data service at least once a month from their handsets, spenders using basic phones the figure is closer to 20 percent. Only 36percent of low ARPU users are both aware of non-voice and non-SMS services being available as well as knowing how to use them, and against such low visibility, 20 percent uptake looks pretty good. So the first problem is awareness, not usage.
While some operators understand this, none has yet scaled a solution for driving data services awareness.
One challenge is the problem of the free rider: “If I spend millions of taka educating the base, then what if half the customers leave and use the services offered by another operator?” The government of Bangladesh could draw upon a Universal Service Obligation Fund model to fill in for such “market failure”.
Then there is the matter of understanding rural customers. Rural users, partly because of their remoteness, commonly have a greater dependence on mobile communications for being connected, a greater need for channels to deliver them all range of services and goods. Despite having terabytes of customer usage data that could help untangle the rural mass into usefully understood and targeted segments, operators remain unable to use this information flexibly enough to generate distinctions between different customers with similar spend profiles. This is not simply a failure to know one's customer, but a result of the complexity of harnessing IT systems, which cannot generate uniform, analysable, customer data.
Address users' needs for services to be micro-local and in their language of comfort
Trying to attract developers to create “apps” for rural use in emerging markets has not worked. Seventy-five percent of developers are motivated to write apps to a large market penetration, which explains their preference for iOS and Android platforms to create apps to be used across multiple geographies. Weaning developers onto writing for a Bangla local user base, whose usage is unproven and whose handsets have multiple operating systems, is not an easy sell. But the Bangladeshi mass user market may not need 200,000 apps as the iOS and Android platforms have 5-10 “hero services” whose content can be micro-localised and translated, and which address specific, understood needs, might be a better place to start. Operators can gather more local insight, work out exactly what people want, trial a small number of services, and avoid having to “pick winners”. The operator who searches for hero services diligently may well find a handful that does have the potential to win over millions of users.
Localisation has to be precise and some operators around the emerging world are translating local service requirements into revenue streams that have encouraging prospects.
* Chandamama (Uncle Moon), a children's magazine running in India, started its own app with a $1 monthly subscription. The app, currently running in nine languages, has reached 3,000 subscribers in the first few months since its release.
* Kisan Raja (Farmer King) is a GSM-based irrigation controller selling at around $100 in the southern states of India. It allows farmers to check and control their irrigation pumps from their houses using either a landline or a mobile phone.
* In Cairo, Egypt a live traffic alert service whose content is generated daily by commuters typing in traffic news onto their handsets as they travel to and from work, has now become one very popular -- the content is in Arabic, and the information is highly localised.
Develop service models fit for a varied user base
Bangladesh's mobile operators are doing a heroic job to achieve what they do in rural areas through operations that are fit to deliver growth in metros. But they have stretched the operating model too far, and it is time to think of better, more granular ways to acquire, retain and serve customers in the rural market. We believe it is time for the mobile industry to give birth to a tiered service model, for example through the creation of “VillageCo”. Such a business should be designed to:
* Deploy networks more cheaply, for example using off-grid network solutions;
* Differentiate service levels and components to pre- and post-paid users depending on the user type and spend,
* Work more with NGOs. At the local level to enable better services, and further reaching distribution and sales. BRAC, the world's largest NGO by beneficiaries covered, is embracing mobile in providing healthcare, and through this its workers are encouraging mobile diffusion.
* Spend less on sophisticated customer service and care, for example offering more care through retail outlets and orienting the same staff to both solve problems and up-sell at once;
* Brand more in localised themes, for example by creating a sub-brand of the current one;
* Offer handsets customised for village users' preferences, such as torch lights and loudspeakers; and
* Offer a subset of micro-localised services, and collaborating across operators to offer common platforms, toolkits and programmes to offer developers a larger opportunity.
Over the next few years we expect to see the refinement of the telecoms operating model at the rural level through such initiatives, to enable the mobile revolution to truly expand its reach into remote, and often low income, segments, and to dispel current notions by doing so profitably.