Your browser is not supported. Please update it.

9 August 2021

Rakuten’s bid for commercial dominance of Open RAN is laid bare

Thousands of column inches have been devoted to Rakuten Mobile and its cloud-native 4G/5G network in recent years, but from a commercial point of view, being a new entrant in a mature market like Japan is very tough. The network may be the world’s most cloud-based and open large-scale deployment, but the MNO business results are far less spectacular. Rakuten Mobile has resorted to commercial tactics that are far more conventional than its architecture, notably price wars that are starting to impact on its rivals, notably Softbank (see below) – but without propelling Rakuten to hit its customer share targets.

But the mobile business, if analyzed as a conventional MNO, has always been a red herring. The cellular network is designed to enable new services and differentiation for the parent business, which provides cloud, ecommerce and content offerings in Japan and many other markets. It needs to attract new consumers to achieve those benefits, but they do not need, necessarily, to deliver high ARPU or revenues directly to the MNO unit. Like Amazon, Rakuten understands that connectivity is worth offering almost for free, if it drives higher additional spending on other services that may carry better margins and stickiness.

Rakuten Mobile may or may not deliver the targeted benefits to its parent company. Far more ambitious, though, is the creation of a brand new business that turns Rakuten into a global platform provider, drawing on its own experience in deploying an open RAN. It announced Rakuten Communication Platform (RCP) in 2020, promising pre-integrated and pre-optimized combinations of hardware and software, in each layer of a cloud-based end-to-end network, that would reduce the time, cost and effort to deploy a multivendor open 5G network – especially for operators without significant inhouse engineering resources like Rakuten itself, or unwilling to pay high prices for a systems integrator.

Rakuten executives, notably CTO Tareq Amin, have discussed RCP extensively in the past year, but the strength of the commercial significance to the group has not been clear. Sometimes RCP appeared to be a loose federation of suppliers and supportive operators, such as Bell Canada and Orange, designed to accelerate progress and deployability by stimulating an ecosystem. At others, RCP looked more like an attempt to define a common platform, which would be well integrated and easily deployable, but would create barriers to entry to suppliers that were not included, and would put some form of lock-in into Rakuten’s hands.

The latter model has clearly won out – quite prescriptive, with Rakuten’s favored inner circle in pole position (it has even acquired its key vRAN software provider, Altiostar), and completely commercial. RCP has been placed into a new business unit called Symphony, which is separate from the MNO business, and will be headed by Amin. This looks primarily designed to deliver new revenues to Rakuten Group, rather than to ensure the dreams of Open RAN are kept alive.

Rakuten Group CEO Hiroshi (Mickey) Mikitani said last week that he sees the addressable market for the Symphony business as over $100bn a year, worldwide, with Rakuten offering integrated platforms, as well as network-as-a-service and software-as-a-service, to suit a range of cellular network deployment and operations models among operators, enterprises and governments.

Given the geopolitics that surrounds Open RAN, the fact that Symphony is an overt power play and bid for market power will please the Japanese government, which like those of the USA, India and other countries sees Open RAN as a way to encourage a local ecosystem, assert hi-tech power and disadvantage China.

But will Symphony deliver the goals of huge revenues and ecosystem dominance? There are significant question marks, one of which is how far large operators will support the platform. Open RAN supporters such as AT&T (and, indeed, China Mobile) have the resources and culture to continue to develop their own platforms, and many large players may be wary of being as dependent on Rakuten’s technology choices as they currently are on Nokia’s or Ericsson’s.

Without significant real world deployment support from large operators, Symphony could still succeed commercially as a platform for greenfield operators, non-telco deployers such as enterprise private network operators, and smaller MNOs. That would provide a valuable way to ease the path for these companies, and would force Symphony to remain fully supportive of global open specifications, and to keep its ecosystem doors open, rather than risk being separated from the standards adopted by the Tier 1s.

It is not impossible that a separate Open RAN will evolve for small cells and enterprise networks, while the 5G and future 6G macro network platforms develop separately. A broad ecosystem, comparable to that in enterprise WiFi, could support the large numbers of smaller networks expected to emerge in the coming years to support industrial and government requirements. That would be able to focus on those requirements, rather than the never-ending battle to drive even more top end performance and responsiveness into the macro network. In turn, it would be able to achieve scale without waiting for the huge challenges of macro Open RAN to be fully addressed.

To be successful, such a platform would need to embrace not just O-RAN, but other specifications that are becoming popular in enterprise environments, such as Small Cell Forum’s nFAPI interface, based on 3GPP Split 6 for fronthaul.

However, if there is a bifurcation in Open RAN, that would deprive the platform of much of its scale, and of the driving force that comes from the innovation and deployments of operators such as AT&T. Rakuten is certainly not constraining its ambitions to the relatively new and forgiving world of small cell, enterprise and rural networks. But for success across the whole 5G landscape, it will need to attract more than just the warm words of Tier 1s like Orange – it will need to see these companies actively using Symphony.

So far, active macro deployments of Open RAN have come from greenfields, notably Rakuten Mobile itself and Dish (which is creating its own ecosystem with its own preferred suppliers, only a few overlapping with Rakuten’s). Reliance Jio is also building its own platform and partnerships. The first national MNO in Europe to commit to use RCP is Germany’s new entrant, Drillisch 1&1, and that was a significant endorsement to accompany the announcement of Symphony. But 1&1 is also a greenfield, and will face similar commercial challenges, in a mature and competitive market, to those of Rakuten Mobile.

Rakuten claims it has many commercial deals for RCP, including some with established telcos for large-scale deployments, but so far these have not been public. The NaaS and SaaS offerings will certainly help to accelerate progress towards those new models of deploying 5G services, and even small or greenfield operators, if successful, can be valuable proof points for a new approach. But like Open RAN itself, Symphony will not be an overnight success, and its path will be smoother if it can attract full support from a significant number of large operators, despite the obvious concerns about lock-ins and geopolitics.

Eyes will be on Europe, whose Tier 1 operators have historically been less likely than their US, Japanese or Chinese counterparts to create their own platforms and ecosystems from scratch. However, these telcos, especially the ‘Gang of Five’ supporting a local Open RAN industry, are cooperating to an unusual degree on shaping the market to their own requirements, and potentially even defining and building their own cloud network platform. Symphony needs to be part of such efforts or risk driving Open RAN down the path of geographically fragmented platforms.