By Vinod Mahanta 

Companies operating in India's ISP market are encountering their own version of http 404: Profits Not Found. One reason for scarce profits is clutter: India boasts 453 companies that have secured licences to be Internet Service Providers, 83 across the country. Delhi has 22 functional service providers who between them have around 2.5 lakh subscribers. Mathematically, that works out to 11,363 subscribers a company, but if you leave out the five largest ISPs in the city, VSNL, Mantra, now, Satyam, and Dishnet, who between them have 2.1 lakh subscribers, the 17 others have an average of 2,352 subscribers a piece.

Competition has been good-as experts posit it always will be-for the consumer. In 1996, India had one ISP, the state-owned Videsh Sanchar Nigam Limited, and internet access cost Rs 50 a hour. Today, it rarely costs more than Rs 10 a hour and some ISPs give it away free. Dial-up rules the world: 95 per cent of internet connections in the US are dial-up ones; the figure is certain to be higher in India; only companies offering the service will not be making money in a hurry.

The smaller ISPs are also discovering that the business is a capital-sink of sorts. It could cost anything between Rs 70 lakh and Rs 2.5 crore to get started and offer upto 10,000 connections. Starting up operations in a city would mean setting up a node or a hub there. It needs at least 30,000 connections to justify it. The cost of linking would vary depending on the size of the network. The price tag on an international gateway normally used by all players is Rs 2.5-3 crore for 8 mbps link. As Atul Kanwar, the CEO of Mantra Online puts it: ''ISPs have realised that the cost of acquiring, servicing, and retaining a dial-up customer is far in excess of the revenue per subscriber.'' Adversity breeds innovation, and while not exactly imaginative, the larger ISPs are experimenting with business models that could help them break even sooner than later.

Geography isn't history: the numbers seem to indicate that focusing on a particular geography may actually make great business sense: 72 per cent of the country's 1.4 million internet connections are in the four metros; add Bangalore and Hyderabad to the sample and the figure goes up to 80 per cent. The smaller cities may not be able to sustain more than one ISP in the long run and the survivor will likely be the one with the deepest pockets as it could take some time to reach to magical break-even 50,000 figure. ''For long-term players it makes sense to be in small cities too,'' says Ramesh Ramanathan, President, Internet Access, Satyam Infoway.

Still, consolidation isn't such a bad idea. Mantra, for instance, has decided to freeze, for the time being, its presence to the eight cities in which it offers its services. Dataccess' now offers its service in just two cities-Delhi and Mumbai (it has 82,000 subscribers across both). And C. Sivasankaran's Dishnet DSL service focuses on the four metros and Pune. The emphasis on large cities could help these companies tap into advertising (almost all have a horizontal portal) and e-commerce revenues. Better still, their retail-ambitions fit perfectly into a B2B-plus-B2C model.

There's B2B here too: There's money to be made in B2B. Wipronet focuses exclusively on corporate connectivity; 70 per cent of Satyam's revenues arise from this; and Mantra is moving to a corporate-and-retail model. The emphasis on c-c isn't surprising: corporate customers need more bandwidth than retail customers, and will likely opt for value-added services like Virtual Private Networks (VPNs). The flip-side: companies insist on zero downtime, and guaranteeing this means the ISP will have to own and manage its own international gateway. Says Siddarth Ray, the Managing Director of now: ''It is important for a retail ISP to do corporate successfully.'' The logic? Retail customers consume bandwidth at night; business customers do so during the day; ergo, the two can share the same network. The only caveat: ISPs adopting this model need to tweak their last-mile access for corporate customers, allowing them broadband access as opposed to that of the dial-up variety.

There may be room for ASPs: it isn't happening yet, but companies like Satyam Infoway, HCL Infinet, and Bharti BT (Mantra) are looking seriously at the Application Service Provider model. That would require them to invest in data storage, and basic non-critical applications like inventory-, supply chain-, or payroll-management. Both small and medium enterprises that do not have the wherewithal to do this on their own, and large companies that believe their employees would be better off doing something far more critical, are potential customers for this kind of service. But the shift to an ASP-model won't be easy, believes P.G. Ponnapa, COO, Netkcracker: ''There needs to be radical shift in the attitude of corporates, and certainly better bandwidth quality for the ASP model to work.''

Other models will emerge as the economies of convergence force companies operating in one node of the telecom value to diversify into other nodes. That could open up opportunities to bundle services or subsidise one service offering with another. Or a wave of M&As could sweep through the domain, leaving the field clear for four of the five national level ISPs who will be left standing at the end of it all. Till either of these actually transpire, ISPs will continue trying everything they possibly can to break into the black.

Issue Contents  Write to us   Subscriptions   Syndication 

INDIA TODAY |  INDIA TODAY PLUS | COMPUTERS TODAY  |  TEENS TODAY  
THE NEWSPAPER TODAY | MUSIC TODAY | ART TODAY | CARE TODAY

© Living Media India Ltd