Ten telecoms predictions for 2018

Some of this might happen, but then again it might not. Who do you think we are – Nostradamus? If you think you can do any better then go for it mate – that’s what the comments section is for.

1. There will be plenty more M&A between the telecoms and media worlds

It remains to be seen whether AT&T’s acquisition of Time Warner will go through, but regardless it is symptomatic of the underlying trend towards marrying content with the means of distributing it. As with most mega-M&A the rationale focuses on prosaic benefits such as efficiencies and economies of scale, but underlying it is a defensiveness brought about by the hijacking of both the telecoms and media business models by the internet giants. There seems to be a growing sense of ‘my enemy’s enemy is my friend’ and they all seem to be huddling together for comfort in the face of the onslaught.

But while the corporate will is there, the way forward is increasingly fraught. For one the menu is getting thin – there has already been a lot of consolidation and the very existential posed by the likes of Google and Facebook means there aren’t many startups in either world. And then there’s the possibility that the internet giants themselves could get into the media M&A game – after all YouTube didn’t go too badly for Google. But Telecoms.com believes concerns about freedom of speech and information could stand in the way of all this M&A. Already we’re seeing Facebook and Twitter cave into calls for arbitrary censorship along ill-defined pseudo-legal lines and this itself is merely a symptom of these profound concerns. Let’s see how regulators react when Chinese internet giants start buying western media companies.

2. 5G will go from being utopia to a problem in search of a solution.

While the 5G hype was mostly driven by manufacturers, and to a lesser degree academics and Governments, Telecoms author William Webb says the mobile operators have allowed themselves to be drawn in and often thought it important to be involved in trials and in boosting their credentials by repeating the rhetoric. But they have been walking an increasingly fine line between being pro-5G with regulators who might award spectrum and being anti-capital expenditure when talking to analysts and investors. As the 5G spectrum battle increasingly appears to have been won, but as financial pressures continue to grow on the back of often falling ARPUs, the time has come for a change of position.

We saw this in November 2017 with both Vodafone and BT publicly saying that they could see no business case for 5G and that it was more about network efficiency – which would in principle help reduce capital cost. Expect to see other operators increasingly following this approach during 2018, especially those with investors that need appeasing. Operators will talk more about maximising the value from their 4G assets than deploying new 5G systems. In practice this will mean radio network expenditure remaining at historically low levels and increasing focus on network management and maximising retention of the most valuable subscribers (those that pay higher ARPUs but have relatively low usage). Of course, none will actually say that 5G is bad or that it will never be useful – there is no point in taking such a stance. It will just be something of potential interest in the future.

3. 3.5 GHz will be the primary band for 5G

Telecoms consultancy Northstream believes the 3.5 GHz band occupies the 5G sweet-spot. The mmWave band, popular for pre-5G FWA deployments, presents challenges with mobility support and providing consistent user experience, while sub-1 GHz spectrum is occupied with previous generations of mobile technologies and is not suitable to deliver the high throughput needed for 5G enhanced mobile broadband. Therefore in 2018 the telecom industry will shift its attention and efforts towards the 3.5GHz band, especially in Europe, as this band brings a number of practical benefits.

Although the propagation challenge of mmWave band can to a large extent be mitigated by the use of beamforming technology and site densification, its ability to support mobility so far remains feasible only in theory. Sub-1 GHz spectrum can facilitate wide area coverage, but these low bands are crowded since they are already used for previous generations of mobile technologies. In addition, low bands have limited bandwidth and cannot benefit from massive MIMO.

Considerable amount of 3.5GHz spectrum is available and can be directly deployed onto existing grids while still taking advantage of massive MIMO technology. This in turn delivers a significant improvement in user experience that can be properly described as 5G. 3.5GHz is also an ‘almost’ globally harmonized band that’s ready for allocation. The exception is the USA, where it’s used as part of the CBRS band.

4. The march to ultra-fast broadband (UBB) will run out of steam

The industry is buzzing about operators cracking the 100 Mbps threshold, not just in a souped-up lab test, but actually delivering this level of service on a consistent every-day basis for consumers. In 2018, the reality of this challenge will settle in, says Brendan Gill, CEO of network performance tracker OpenSignal. The huge investment required, combined with no clear line of sight to incremental revenue opportunities, means more operators will realise the scales just don’t balance and will shift their focus elsewhere. So what exactly do we need 100 Mbps speeds for?

Video makes up the majority of data traffic on most networks. Surprisingly, HD video requires less than 5 Mbps in speed. Even Ultra HD ‘4K’ video needs only around 15 Mbps – which is actually less than the global average LTE speeds we recorded of 16.6 Mbps. This indicates that the LTE status quo is already well equipped to handle the video deluge. And that’s not even factoring in LTE-Advanced improvements like 256-QAM that we’ll increasingly see rolled into production in 2018.

To use an ice hockey metaphor, it’s important to skate where the puck is going – but let’s not forget the game we’re in already. We predict that 2018 will be the year in which the ‘inevitable’ march towards 100Mbps speeds starts to be questioned by the industry decision makers, and they instead put more focus on what actually moves the needle for mobile users, and operators’ bottom lines, in the nearer term.

5. The voice UI will finally become useful

The growing ubiquity of connected smart speakers in our living rooms seems to have significantly accelerated the development of the voice user interface. New technology paradigms are often closely aligned to new UIs; the PC was all about the keyboard, mouse and GUI, while the creation of a truly user-friendly touchscreen UI sparked the smartphone revolution a decade ago. But now it’s all about hands-free, whether you’re in the living room, the car or wearing a smart watch.

Until we crack telepathy, the most viable hands-free UI is likely to be voice. The growing installed base of voice UI-dependent devices should mean that companies like Amazon and Google are in an exponential phase of voice recognition development – not just accuracy, but things like natural language processing and understanding nuance. So Telecoms.com predicts that the voice UI will have its breakout year in 2018.

But there are real economic, legal and economic concerns around the voice UI and the wholesale hoovering of data it enables. Google now knows what we search for, Facebook what we share and Amazon what we by, but voice UI-enabled devices will increasingly know everything we do and that’s pretty damn creepy. Add to that the imminent GDPR moment of truth and this moment of triumph for the internet giants could also present significant peril.

6. Data trading will rise to make data great again

Thanks to digitalization and growing deployment of IoT, enterprises are collecting increasing amounts of data and looking for ways to generate more value from it, both by making it available to other companies but also by in turn getting access to the data generated by others. One way to efficiently achieve this is through data trading. Northstream believes there will increasingly be attempts to establish data trading markets where anonymized data can be exchanged. In 2018, we will see progress in technology enablers (e.g. blockchain, data mining etc.) and expect that enterprises will begin to put focus on how to adapt their business to the data economy era.

2018 will see accelerated activity from enablers of data trading, including:

7. IoT will happen, but don’t hold your breath

The Internet of Things is an area that has disappointed, insists William Webb. In 2010 Ericsson predicted that by 2020 there would be 50 billion connected devices – we are at less than 10% of that at present and only growing slowly. But it is growing. Ever-more street lamps are being connected and smart meters installed. The eco-systems are slowly growing. In 2018 there will be a significant push from some operators as they roll out their LTE-based IoT solutions such as NB-IoT. This push will help stimulate the market as will the presence of well-known brands.

But IoT will remain slowly growing. IoT is not like an iPhone. With an iPhone an individual buys it and then immediately starts using it. With IoT it is normally a business buying it, and they need to deploy sensors, software, analytics, often new ways of working. A farmer adding IoT devices to his cows will not just order 100 tags on-line. He will want to talk through solutions packages, have help in deployment and have on-going support. These support systems need to grow and develop expertise themselves. This means IoT will be a much more slowly developing area than consumer products.

We should not despair, IoT will happen, indeed is happening already. 2018 will see significant progress towards national IoT networks. But operators will find that they only get a 1-2% revenue uplift from IoT. Instead the value will be spread across the analytics, support systems and experts that make it all happen.

8. Unlimited data plans are here to stay

Northstream predicts that, in 2018, pricing plans with unlimited data will become widespread in Europe, and this time they will be here to stay. Such plans are expected to drive data usage per SIM (as has been the case in Finland) but operators will be able to manage the increase in data traffic in the next years thanks to technology advancements.

As unlimited data plans become widely adopted, operators will be under even greater pressure to transport data as cost efficiently as possible. This increased focus on cost optimization will benefit large players that already have large customer bases to justify the necessary CAPEX investments. This will in turn accelerate and strengthen the trend for operator consolidation and network sharing.

9. AR will become more useful, somehow

Augmented reality had its initial moment of glory with Pokémon Go in 2016, notes William Webb. That has not been followed by a flood of similar blockbuster applications but Apple and Google have both been working on enablers, releasing their AR suites in 2017. These give developers much more powerful tools to write AR apps. Some early examples are already emerging and 2018 will see a number of mainstream apps delivered.

Predicting which AR apps will be successful is as hard as predicting that Angry Birds would be a hit – there is often a randomness to what goes viral. However, one area that looks likely to become widespread is being able to use AR to show what a possible purchase, such as a new chair, would look like in situ. Web retailers will find this useful both as a nice-to-offer feature and because it might mean fewer purchasers feel the need to go to a store to see the real item. Adding AR to show directions superimposed on the real world is another obvious application. None of these feels like they are going to generate huge new revenue streams, but then they do not need a lot of effort to develop either.

For AR to really become valuable it might take the reappearance of glasses such as Google Glass that can superimpose images when undertaking activities such as maintenance, showing a mechanic which part to remove, for example. But 2018 may be too soon for the world to revisit glasses. That is also why not much will happen with virtual reality other than a few more people will feel sick trying it out.

10. Cisco, Ericsson, Nokia, ZTE and Juniper will merge to take on Huawei

Not really, but it would be a laugh wouldn’t it? There is a school of thought that the underwhelming Cisco/Ericsson partnership was a consolation prize for failed M&A and recently we had a trial balloon rumour about Nokia buying Juniper. Huawei has been kicking all their arses for a few years so Telecoms.com says why not have an M&A-gasm and set up a massive geopolitical vendor bun-fight?

Joking aside all of them, Huawei included, need to get their ducks in a row. As covered previously 5G is unlikely to serve up the kind of cyclical boost all of them may have been waiting for. Cisco is showing increasing signs of falling out of love with the CSP world and needs to sh*t or get off the pot. Nokia and Ericsson seem to be more focused on mucking about with their executive teams and it’s not obvious what Huawei’s strategic focus is these days. Aside from cyclical challenges they’re all threatened by an increasing tendency for the big operator groups to control their own technology destinies. Something’s got to give.