The Boring Telecom — Are the glorious days gone forever? (I)
Part 1 — how wireless revolutionized our world and ended as boring commodity
The background
When I started my career mobile internet was still in emerging state in most of the countries, first iPhone was yet to be released and hottest topic for wireless operators was to speed up the 3G rollout, fixed operators only slightly concerned with landline decline, Google being a search engine and Amazon a market place and not the biggest cloud provider.
At the same time there were first signs of fatigue in most developed and saturated markets. Revenues began to stall, products became very comparable commoditized. Unable to find a USP in the simple “minutes-SMS-data” paradigm, price war was the only deafferenting mechanism. I watched this happening first in several EU countries and other parts of the world with a certain price lag. The war of tariffs eroded margins and led to lackluster in innovation. Industry got mature and boring-there is a strong degree of fatigue. In the past decade, Telco sector was second worst performer preceded only by the even more boring Energy sector.
The industry that once enabled all the shiny and innovative companies to emerge and strive on the network backbone, failed to innovate and re-imagine itself. The industry failed to monetize on new technology in full, instead focusing on repackaging and reselling of same old products, while giving away its most precious and competitive assets. It seems as if the memory is getting shorter and the wheel of sansara are spinning faster and faster. Yet the biggest challenge is that the It feels very disappointing, but at the same time, when asked about possible reinvigoration means — I cannot give a clear answer, lacking ideas myself.
So I decided to write a retrospective to reflect on the 15 years I spend in this industry. My hope is that in the process I could find answer to the main question: “Can Telecom be Un-bored?”
Wireless revolution
“Without History There Is No Future”
History of telecommunication industry dates as far back as 1830 with the invention of the telegraph. Fixed telecommunication segment is the older brother, so to say. Wireless, or mobile industry, is much younger than fixed line industry, but its development are fundamental to the overall technological progress. Without mobile network many (if not most) of the services we are using today would be nearly impossible. By revenue size it is also larger being somewhere between 60–75% of total telecom market segment. Hence I will focus on the wireless industry first, and cover fixed market later. There are many similarities in the business development part and some differences.
Wireless history can be split into epochs or “mobile generations” (the “Gs”), each lasting around 10 years on average.
1G — The inception
Although mobile radio systems (or “0G”) existed since as early as mid 40s, it took another 150 years after the telegraph until we had first cellular network in Japan in 1979 or roughly 40 years ago. Within 5 years it build a nationwide coverage with analogue NMT-technology, or 1G. Within 2–3 years Europe followed with their rollouts of NMT in Nordic countries. As with every pioneering technology skepticism was high. A well known story is that in the early 1980s AT&T asked McKinsey to estimate how many cellular phones would be in use in the world at the turn of the century. The consultancy noted all the problems with the new devices — the handsets were absurdly heavy, the batteries kept running out, the coverage was patchy and the cost per minute was exorbitant — and concluded that the total market would be about 900,000. At the time this persuaded AT&T to pull out of the market, although it changed its mind later, but at a much higher entry price.
Back in the late 80s barely anyone could sanely predict the parabolic take-off of wireless adoption. After all its nearly impossible to see a trend of adoption going so fast and steep. NMT was a “heavy technology” — first phones being huge bricks weighing close to 1kg and tariffs being prohibitively expensive. Massive adoption requires product democratization, something 1G could not provide.
2G — The Revolution
Roughly ten years later in 1991 the industry became “digital” with the rollout of first GSM network in Finland by what is today the operator Elisa. It is rather anecdotal and somewhat ironical, that Telco industry is considered not being digital as compared to new age companies. On the other hand it shows how the outsiders treat the word digital as being related to being innovative, app-based service providers, rather than referring to the underlying technology
2G was far more efficient than its predecessor but more importantly it brought in two major innovations: short message service (or SMS) and mobile internet access over GPRS, followed by EDGE. Telco business was thriving: as operators expanded their service coverage to new areas it signed up more and more customers. There was a certain race to get new areas covered. The name of the game resembled pretty much the key line in the movie Field of Dreams.
Subscriber growth came indeed. First it were the business people who cared less about the monthly bill or the device price (which was quite high in early days). Whereas in 1991, there were only 16 millions cellular worldwide and the end of the decade 30-fold!
As the prices became more affordable and devices actually would fit into a pocket — mass adoption really kicked in.
It is thus fair to say that the modern wireless market, as we know it began with 2G in very early 90s and very quickly developed into mass market phenomenon by % of adoption. Toward the end of millennia, in most advanced markets every second person had a mobile phone. A decade later people would have 2 to 3 phones (SIM cards actually) to optimize on the tariffs. Thus we saw penetration rates over 100% and in some larger cities closing in on to 200%.
But there is even more than just a “mobilization of nation”. Wireless telephony brought in a true revolution in those countries where landline networks was not not available to broad population.
Mane countries made a technological leap and lead to a phenomenon of fixed-to-mobile substitution (FMS in short), where more people had a mobile phone than a landline. Today, that is virtually the case in every country
Another important mark of 2G era was the emergence of different technologies: while most European networks rolled out GSM, US operated D-AMPS and cdmaOne and Japan used its own PDC standard. The abundance of options proved to be beneficial for the industry as it spurred innovative competition and strive for continuous improvement.
Undoubtfully, it was 2G that enabled wireless technology to enter mass market state. The technology proved to be so resilient and prone that thirty years later still many 2G networks are in operation, although most operators plan to shut down this technology completely soon.
3G — The Age of Mobile Data
By 2001, the industry faced a second generational change with introduction of. It was the upgrade for 2.5G GPRS and 2.75G EDGE networks, allowing data transfer. Along with faster speeds and better internet experience another promising killer app was video calling. What was long possible on desktops should finally be made possible on mobile phones too.
It failed. People were barely using it. For one of the reasons the pricing was off, compared to regular voice calling. But most importantly, user experience was not right: using video on the go is impractical. There is a nice discussion in this 10 year (!) old thread on the reasons why video calling did not took off. Without a killer app 3G network was an evolutional development to introduce mobile internet at broadband speed experience (although arguably EDGE was faster with close to 2 Mbps than first 3G networks with only 386 kbps).
The hype was strong, the fear of missing out on the next generation revolution even more stronger with all Telco executives. 3G networks required new spectrum resources, as the existing one were used by 2G services. In some countries, operators were ready to pay spectacular prices for valuable airwaves: German operators paid a breathtaking 50bn€ and UK operators close to 40bn€ for just 15MHz. Other countries managed to spend less, but overall 3G spectrum auctions are considered to be most costliest.
Next to paying huge sums for the spectrum, network rollout in higher band (most commonly 2100 MHz), required a denser network and thus more radio equipment and site locations than 2G and thus higher investments. Securing competitive coverage became a costly exercise, as operators focused on the most profitable urban area, providing no or very poor internet coverage outside of cities. There were so technical challenges with 3G technology, like cell-breathing, that impacted on the user experience in overloaded areas. Unlike its successor, 4G, it tried to live in both world: the circuit-switched (2G) and packet-switched (4G). Like many attempts to serve-them-all, it did so not provide sufficient quality.
Still, 3G managed to attract masses to use mobile internet. First experience on the tiny screens was mediocre from todays perspective, but exiting enough back then. However, arguable, part of the success came from innovation in end-user devices.
Whereas before, the phone manufactures were mostly concerned to reduce the weight of the phone while not sacrificing battery life-time too much, new use cases spurred a wave of form-factor experiments.
A variety of phones with or without keyboard, sliding, rotating, clam-shells, with or without touch-screen and stylus. What a cheer compared to the look-a-like same-brick-design of todays smartphone, coming primarily in black and white. Without doubt there was a wave of design innovation.
Without doubt what really drove the mobile internet adoption was the rise of applications (apps). While there were Symbian and mobile windows programs for phone, it was without doubt the AppStore that brought it mass adoption. Before that, most used their phones for voice, basic messaging and browsing. With apps a whole new range of use cases emerged spurring a craze in app development. “Is there an app for that?” became the common meme, answered in Apples 3G iPhone launch campaign back in 2009.
Operators also tried to innovate by offering a broader set of services. MMS was the evolutional development of SMS that allowed media sharing and in 2008 a new protocol called Rich Communication Services (or RCS) started its journey to provide advanced communication features, like IP telephony, group chats, file sharing, location base services and more. The timing was quite right, in 2008 there was no WhatsApp or WeChat, and even Skype was most a desktop solution. It could have been the next innovative killer apps — the premise was right. But eventually the battle was lost against 3rd party new comers, who albeit coming with their product later captured their use base much faster. There are several reasons why RCS never took off, most important one being the industries attempt to find a common standard for interoperability.
Standardization was the killer of innovation. While 3rd party developers focused on bringing bare minimum early on, without looking at the competition, Telcos struggled to find common ground between each other — by end of 2000s more than 800 wireless providers were operating worldwide. Several larger groups were formed through M&A which resulted in running huge behemoths with the grindy wheel of corporate bureaucracy. Decision making became a slow multi-stage process. Ultimately, all Telcos were facing the Innovators Dilemma: their core-business was so big, sales people eventually lacked the motivation and resources that could only potentially become the next big thing. Experiments did not receive the budget and ultimately innovation and R&D functions started getting downsized.
Yet another failure was Mobile TV. Early 3G networks did not have the capacity to stream quality TV footage and so operators rushed into the newly development DVB-H standard (digital video broadcasting for handhelds).
The reasons for failure are several. For once phone producers did not provide broad support for such service. It was a catch-22: more phones would come with DVB-H functionality if producers believed in service take-up. And service take-up depended on broad availability of phones. Alas, new technology required operators to invest first to rollout broadly, with uncertainty on returns
The overprotection of the core business and complex decision hierarchy was also the reason why Telcos never been able to capture full value of VAS market (ring-tones, wallpapers, etc.). For one, they did not control the device, but more importantly, there were too slow in releasing content themselves. Instead they took a marginal share for content delivery or reselling.
Towards the end 2000s came the “Big Disillusionment”. 3G did not live up to the promise of the next revenue boost. Telco spend their war chests on spectrum and later on the race to the “best and fastest” networks for an inferior technology. No wonder that many operators, particularly in Europe decided to switch it off first, while retaining 2G for a longer while. More importantly though, is that Telcos failed to move beyond their core service of connectivity, leaving service margin to more agile newcomers.
“Dumb pipe” became the mocking term for the industry, where the only difference in product was reflected in the aggressive price wars. Virtually most of the value-add was captured by new “digital” players
4G —Faster speeds, bigger screens, slower growth
In the early 2010s next generation of wireless networks went commercial. Commonly called Long Term Evolution (albeit very first deployments were not 4G by IMT definition) it was the first true All-IP packet switched network. Next to many technical improvements the main user experience improvement was the speed: peak data rates of up to approximately 100 Mbps for high mobility such as mobile access and up to approximately 1 Gbps for local wireless access.
Industry learned from the 3G mistakes. Higher bandwidth requires more spectrum, hence for LTE many more bands were auctioned off. At the same time wireless operators were not willing to pay high sums for the spectrum anymore — 3G investments into spectrum did not pay-off for most and as the revenue began to flatten (see chart below), investments were scrutinized thoroughly. Phone industry also came along to support a wider range of spectrum bands — crucial to make more optimized network rollout, e.g. securing broader coverage in lower bands and providing hot-spot capacity in higher bands.
Slowdown of revenue growth (or even reduction in revenues) began already towards the end of first decade of the new millennia.
At the same time subscriber growth was still high — but it case primarily from Asian and African markets.
Developed markets in Europe and North America reached saturation already mid 2000s. Declining revenues were the result of decline in service prices. The industry’s metric ARPU (average revenue per user) fell significantly. Europe has been particularly hard-hit over the past five years, with the drop in service revenues standing at more than 30% between 2010 and 2015 in some countries. Outside of Europe, a country experiencing sharp declines in both mobile service revenues and monthly ARPU is Israel: income from cellular services dropped 53% between 2010 and 2015, while ARPU fell from USD37 per month in 2010 to USD17 five years later.
The main reason for the decline were price wars, between wireless operators and also through the entrance of virtual operators (MVNO), i.e. companies offering services by renting infrastructure. As I discussed above, in the early days the differentiating factor was service availability, or coverage. First entrants had a head start and for years proclaimed network leadership as their USP. That promise is still kept up today as an argument for a price premium. But late entrants have caught up in network reach and soon the difference was hard to tell to average consumer. Lacking any other USP, having comparable service portfolio and quality, the only thing remaining was price.
Price wars emerged already long before 4G network upgrades. In order to sustain ARPUs, operators tried several pricing tactics. One important shift was the introduction of bundles: fixed price package with a certain quota of minutes, SMS and Data. The tactic was to outcompete on the package size: more minutes, more SMS, more Data. Ultimately, unlimited packages emerged — that was the end of the pricing game after all.
As the quotas grew in size, so did the traffic in the network, primarily data traffic and with it the network investments and operating costs. There is an important difference to fixed networks, which can scale far cheaper than mobile network. In mobile networks, network investment is is almost a linear function of traffic. While each mobile technology generation got more efficient per each mobile generation, in absolute terms traffic growth exceeded the efficiency. In combination overall revenue stagnation (and decline), profit margins and other financial efficiency KPIs continued to decline.
Bottom line is that 4G did not bring the revenue boost to compensate for additional network and spectrum cost (Capex and Opex). Neither did it bring new customers or use cases. Admittedly, 4G networks were far more efficient and reduced the incremental costs, but traffic surge spurred by marketing-driven price war has eliminated most of the gains. Between 2010 and 2018 industry revenues were down $27 billion, while telcos invested $250 billion in the network.
So, where is the wireless industry standing today?
As the bells ring for another technology generation switch (5G), in my view the wireless industry is in a deadlock:
- There are no new customers, as mobile user penetration has topped in virtually any country
- There is no differentiation, not even in pricing as all operators offer same USP and core products. To sustain unit price decline, marketing strategy is about bundles and packages — i.e. selling more traffic
- Monetization of investments is very poor and rapid traffic growth requires constant investment and growing operational costs. As an effect, EBITDA margins are still growing
- Industry failed to innovate and bring in new revenue streams that would compensate for declining core revenues. There are few successful attempts, but double digit growth in top line is long foregone.
- The real benefiters of the past telco infrastructure investments are 3rd party “digital players”
