Your browser is not supported. Please update it.

24 November 2020

As Dish puts flesh on its 5G strategy, its challenges are becoming clearer too

Dish Network has the potential to become a poster child for the USA’s effort to define a new 5G architecture, one that will be open, US-centric and will put the country’s industry once more in the vanguard of advanced mobile platforms. Dish’s greenfield mobile status gives it the opportunity to emulate Rakuten in building a fully cloud-based network from day one in order to optimize service agility and cost-effectiveness, and to work with suppliers of its choice.

This opportunity brings with it many challenges, which Dish will have to meet in the full glare of publicity and other operators’ expectations, as Rakuten is doing. There are four main challenges with Dish has to address, if it is to be the valuable proof point that the open RAN community hope for. These are covered in more detail in the analyses below, but can be summarized as:

  • Commercial – existing business

Like Rakuten, Dish has entered the mobile market from a well-established adjacent sector, but while the Japanese player has a growing business in cloud and ecommerce services, Dish’s core pay-TV activities are under intense pressure. There has been persistent speculation that it would offload its satellite TV business and concentrate on a mobile broadband multiplay, but it is unclear who would be ready to buy. And if Dish, as now seems to be the case, is finally committed to full 5G build-out and services, it will no longer have the cushion of spectrum assets to monetise directly to boost its balance sheet.

  • Commercial – new business

Even more than for established MNOs, Dish therefore needs to ensure its 5G investment generates new revenue streams quickly, to offset TV decline. Some of these will come from new mobile services, and from improved ability to offer multiplay bundles cost-effectively over the same 5G infrastructure. Dish already has one new, if modest, revenue stream from its acquisition of Sprint’s Boost prepaid customer base, as a result of conditions of the Sprint/T-Mobile merger. However, the US consumer market is more competitive than it has been for years, especially with the commercially aggressive TMO now strengthened in terms of network technology and spectrum, and the growing impact of the cablecos – another group of disruptive new entrants vying for the same gains as Dish.

This has turned Dish’s attention to wholesale, business-to-business and direct enterprise models. There is certainly more new demand here, and the established telcos have yet to build enterprise bases that cannot be reasonably challenged (at least outside, perhaps, the automotive sector). Dish has discussed building a wholesale platform for enterprises and diverse service providers, rather on the lines envisaged by the beleaguered Ligado, since its 4G days. Its first shot at build-out, of a (now-suspended) NB-IoT network, was firmly geared to B2B enterprise and IoT opportunities.

Now it is painting a vision of supporting network slicing from day one, courtesy of its cloud-native 5G core, and stealing a march on rivals in terms of supporting diverse enterprise needs from a common platform. This would be a strong and much-needed commercial approach, but comes with challenges, including Dish’s lack of experience in enterprise markets, and the immaturity of the technology (see below).

  • Infrastructure costs

However cost-effective and agile a cloud-native, software-centric RAN may be compared to integrated appliances, an operator still needs a lot of heavy, and potentially expensive, infrastructure. The increased density envisaged for 5G will only increase the numbers of cell sites, edge nodes, xHaul, power and physical active equipment such as antenna/radios.

The passive infrastructure is far less susceptible to price erosion than the active networks with their virtualized elements, and Dish will need to rely on increasing competition in the US infrastructure and fiber markets, and on some creative partnerships with cloud and network partners, to make its economics work. It said, when it first started acquiring spectrum, that its cost model would only be viable with a network hosting partner, and while it now has far more cost reduction levers to pull, such as its open RAN network plan, it is still likely to need to rely heavily on infrastructure and asset sharing to achieve its 5G coverage targets.

  • Is cloud-native cheaper?

Reduced total cost of ownership is a key objective for most operators migrating their RAN and core to the cloud, but those cost efficiencies may be a case of delayed gratification for early adopters. As Rakuten, and the highly customized virtual networks in South Korea and Japan, indicate, while COTS hardware may be economical and scalable, there are many other set-up costs associated with cloud migration.

These include sourcing the right skills, inhouse or through partners; relying heavily on systems integration to build and optimize multivendor systems before all the interfaces are fully standardized and commercialized; building out or renting an edge cloud that is sufficiently distributed and optimized to support RAN performance at the same level as an integrated platform. Dish’s capex projections, like Rakuten’s, are aggressive and widely questioned by analysts, and are highly likely to exclude some categories of initial deployment expense (which may be accounted as opex, but will still weigh on the first couple of years of roll-out and ROI).

Greenfield operators are in a strong position to address these challenges in their networks, of course, and will be useful templates for those that are more encumbered by legacy network maintenance and coexistence. But new entrants have greater costs and challenges on the commercial side, since they need to create a subscriber base, a brand and customer trust without any existing foundations. To do this, and to offset any negative perceptions of the Dish TV brand, the improved performance, pricing and applications enabled by a cloud-native 5G network will have to be very demonstrable, and easily visible to potential users, from day one.