LPWA connectivity - a Paradox of Choice?
Having worked with LPWA (Low Power Wide Area) since 2015 and with mobile data since 1997, it was very interesting for me to attend the LPWA World 2018 IoT conference in London this week. People told me that this year's event was much smaller than last year, and a look at the video from last year´s event confirms this. Does that mean the LPWA business has shrunk?
Rather not, as it is not actually a major business yet - certainly not when it comes to connectivity. And there is reason to doubt that LPWA connectivity ever will become a major business compared to the mobile operators’ core offering which is mobile communication and data access, sold through different packages and propositions. The fact that LPWA connectivity provides much more basic value than high-speed mobile networks do is one of the reasons why.
For end users, the value of mobile connectivity lies in immediateness (high speed and low latency, often better than with fixed broadband) and in mobility (services should work in the same way in any place, in any country and also on the move). Since the commercial launch of mobile data in the 1990s, network speeds have constantly been pushed to higher limits. As of November 2017, 14 Gigabit LTE networks were commercially launched (Source: Ericsson Mobility Report) - an amazing achievement if we look back at the early days of GSM, when 9,600 bit/s were the end of the line.
LPWA networks are designed for completely different purposes - to connect massive amounts of ultra low cost devices, most of which are stationary and just occasionally transmit small amounts of data. Reflecting on this, it seems obvious that if customers are willing to pay for LPWA connectivity at all, they will only be prepared to pay a fraction of what a 4G mobile subscription costs today. Or perhaps nothing? It depends.
LoRa connectivity can be provided without service subscriptions, as anyone can buy and install a gateway and operate it on unlicensed spectrum, in most countries - in a way similar to Wi-Fi (although the latter is a short-range technology not optimized for low power consumption). Then, there are mobile operators (e.g. KPN, Proximus or SK Telecom) and initiatives (such as The Things Network) offering LoRa-centric propositions. In a similar approach, Sigfox charges a small annual connection fee for its basic service, operating wide area networks in many countries.
Mobile operators offering Narrowband-IoT should be able to charge higher prices: they can achieve nationwide coverage more easily (thanks to the possibility of software-upgrading LTE base stations), they use licensed spectrum (higher customer confidence in the service quality), the technology supports higher speeds (still a sales argument, even if most LPWA applications do not need high speeds) and the network is based on 3GPP standards (again, higher customer confidence in future-proofness of their decision).
NB-IoT rollouts have taken far more time than expected, and there are still not many operators offering nationwide coverage. But there are positive signs of NB-IoT take-up. As a GSMA representative stated at the conference, there are more than 10 million NB-IoT connections in China today. That number appears astonishing compared to the 3 million connections claimed by Sigfox, a company that started its network deployment almost six years ago and that has now 18 countries fully covered by its network.
What is the conclusion for LPWA users? It is a buyer’s market, which Wikipedia characterizes as “an excess of supply over demand, leading to abnormally low prices”. On the other hand, the abundance of technology options creates uncertainty at many companies, leading to a “paradox of choice” effect at IoT users. On the supply side, mobile operators realize that they cannot avoid price competition - the recently unveiled 1NCE initiative (powered by Deutsche Telekom) illustrates this: Once launched later this year, business customers will be offered connectivity packages including 500 MB of data (2G, 3G, NB-IoT) valid for ten years and throughout the EU, for a price of €10 per SIM.
Unless customers purchase top-ups or value-added services, this equals an operator revenue per connection of €0.083 per month. For comparison, the average revenue per M2M connection in Sweden was 9 kronor (€0.88) per month in the first half of 2017 (down from 28 kronor in 2008, Source: PTS) which is over ten times more than 1NCE’s offer would cost. Apparently, the arrival of LPWA has changed the M2M market dramatically, as there is now a choice of dedicated long-range networks able to connect sensors and tracking devices very cost-efficiently. But will volumes be high enough for network operators to stay in this business?
Despite the extremely low price levels, mobile operators should look at the LPWA business in a positive way: They can provide nationwide coverage at lower cost by software-updating their existing network infrastructure, if it is recent enough. They have a sales force which regularly visits business customers. Unlike rivals offering only LPWA services, mobile operators can design customer propositions that utilize a mix of technologies - there are many IoT applications that require high speeds and mobility, and those generate much higher revenues (for instance, NB-IoT can often be used as a backup connectivity option). Finally, many operators have already made moves to widen their value chain reach beyond connectivity, e.g. by acquiring IoT specialists or by creating strong partner networks.
LPWA will keep on growing, as more and more companies and organizations will realize its benefits. Let's continue working on it and check the progress at next year's LPWA World!