What Went Wrong with 5G?

Some market insiders have a saying odd-numbered cellular generations are destined to fail. Indeed, while 2G and 4G have been instrumental in introducing voice and mobile broadband applications to the mass market – and generating the financial returns that came with these – 3G only provided a lukewarm foundation for 4G that came after it. The killer application for 3G was video calls, something that never came to market until Apple and other OTT companies tied these applications to the device ecosystem. It is also for this reason that 3G networks are being decommissioned in many markets, before 2G which is still providing simple voice and SMS services across a nationwide footprint nearly across the whole globe.

The success of 5G is also a controversial topic, nearly three years after the first 5G networks were lit. In the consumer domain, mobile operators have accelerated their deployments with Non-Standalone Access (NSA) that allowed them to tether a shiny new radio access network to an existing 4G core network. This has been a blessing and a curse: a blessing because it allowed them to deploy the new network without the more complicated Next Generation Core (NG Core) that was not technically and commercially ready in time, but a curse because NSA downgrades 5G to a faster pipe compared to 4G. This fact, coupled with the fact that 5G deployments are scarce, means that subscribers will be spending most of their time on 4G and only using 5G a fraction of their time online. Indeed, OpenSignal’s latest analysis of 5G in the United States indicates that users spend ~10-15% of their time connected on a mobile network on 5G (excluding T-Mobile’s 600MHz 5G network that offers wide coverage, but overall lower data rates)[1].

Nevertheless, 5G was designed as a grand upgrade to 4G that was supposed to enrich people’s lives while accelerating digital transformation. Three years after its initial deployments, real 5G is only a shell of what it was supposed to be. What went wrong? And what should the industry learn from it?

[1] OpenSignal 5G User Experience Report – April 2021

5G corresponds to the technical characteristics the International Telecommunications (ITU) has put together under IMT-2020. This translates to the following specifications and the famed 5G triangle that outlines eMBB, URLLC, and mMTC: Maximum downlink data rates of 20Gbps, typical user-perceived data rates of 100Mbps, latency down to 1ms, and a connection density of 1 million devices per square km. 3GPP has spent a tremendous effort to deliver specifications and standards to provide for these capabilities, and the supply chain has delivered equipment to the market for eMBB, but not yet for URLLC and mMTC which will be fully specified in subsequent 3GPP releases.

However, 5G was designed in a “build it and they will come” basis, for both consumer and enterprise domains. The telecoms community assumed that 5G capabilities would offer a lucrative incentive for enterprise vertical companies to be involved, especially when offering the network specifications that were discussed above. But enterprise verticals – and the consumer domain to a certain extent – do not develop new capabilities this way anymore, especially when hyperscalers and enterprise digitization efforts are taking huge steps forward and predominantly use subscription-based business models. The enterprise vertical technology stack is also fragmented and governed by dozens of competing technical specifications, that include networking domains most often. Can 5G displace, replace, or even augment these? It is unlikely, and mobile operators as well as infrastructure vendors are now coming to terms that they need to address the long-tail of applications rather than generate scale in a few key use cases of applications. 5G promised much to enterprises but failed to understand the intricacies and legacy technology stacks that were already there. Coupled with the bi-annual cadence of 3GPP specifications – that many enterprise CTIOs do not have patience to understand and follow – means that 5G is destined to fail in enterprise verticals. Moreover, 5G has failed to deliver even what was initially promised to enterprise verticals and the bi-annual cadence fails to take into account technology refresh cycles in enterprise verticals. The telecoms community has thus failed to adapt in terms of business structure and has attempted to extend consumer-based technology and business development to the enterprise domain. This has failed to create significant opportunity for 5G in verticals.

5G has yet to make an impact on the enterprise market and remains a faster pipe on the consumer domain. Here are three ways 5G can become a success in both domains and create scale beyond niche applications:

1. Adapt to subscription-based models: 5G enterprise business models need to address the Small-Medium Business (SMB) space and to do this, it needs to operate on a subscription-based model that doesn’t require a high upfront cost to be deployed. In this model, mobile operators and vendors would have to build 5G platforms that could be used by other vendors for enterprise vertical customization. An example initiative is Singtel’s GENIE 5G in a box.
2. Partnerships with hyperscalers: AWS, Azure and Google Cloud are aggressively targeting both enterprise and telecom verticals to become the platform of choice. 5G can augment these initiatives with great connectivity features that include high quality, even nearly deterministic wireless capabilities. For example, 5G can be used to extend SD-WAN capabilities over the wireless domain and extend existing initiatives to a much broader addressable market.
3. Government-backed initiatives: China’s most recent Sailing plan is a great example of how government-backed initiatives aim to promote 5G as a key technology for a nation’s digital transformation. It remains to be seen whether 5G will succeed through government- or market-led conditions, but China has been aggressive in deploying 5G and it is very likely that the Sailing plan will accelerate this even further and help 5G create a critical mass of enterprise applications.

These options are not exclusive and the only way 5G can succeed, but it is becoming very clear that 5G needs to adapt to market conditions and move well past the “build it and they will come” mentality. The enterprise world is now starting to operate in a software, hybrid cloud and subscription-based mindset, and 5G will need to adapt to that to succeed.