Mobile TV is engendering strongly held - and opposing - positions on the chances of its successful adoption. Dr Windsor Holden examines the detail behind the posturing
Much of the literature that has been written about mobile TV falls roughly into one of two camps. In the blue corner are the evangelists for whom it is the killer app to which the populace will subscribe in their millions; weaving and ducking in the red corner are the nay-sayers, amongst whom the general consensus appears to be that they would not touch mobile TV, no Sirree, not even with a barge-pole, and anyone that does will be turned into a pillar of salt, or at the least see their enterprise go belly-up due to lack of customers.
Now then. Mobile TV clearly represents a tremendous opportunity for the various participants in the value chain. People will pay a premium for mobile TV. So far so good. Now for the caveats. It will generate significant revenues, if... (You should sit back and make yourself comfortable at this point, for there are quite a few "ifs".) If it is properly marketed; if prices of handsets and services are not prohibitive; if operators are allowed a free hand to pick their standard of choice; if it is if delivered via a cost-effective but robust solution; if the service can be received nationwide; if the service quality is acceptable at high-speed; if service providers can obtain the mobile rights to key content; if prime spectrum is made available; if chipsets are fitted in mass-market handsets. And so on.
Read on, and we'll go through a few of those "ifs" in a little more detail...
One of the most pressing concerns for adherents of the various standards is that regulatory bodies will mandate one of their competitors, effectively ending their interest in a given market. Chinese regulator SARFT moved down this path in 2006 when it mandated STiMi as the de jure standard in the country, but a similar (if not yet so emphatic) stance has been taken by the European Commission (EC). The Commission report, Strengthening the Internal Market for Mobile TV, argued that DVB-H should be introduced as a common standard for the following reasons:
- DVB-H is the most widely-used standard in Europe and is also becoming popular worldwide; and,
- DVB-H is fully backwards compatible with DVB-T, a core advantage as DVB-T is used for terrestrial digital transmission everywhere in Europe, and network operators have experience in building and operating DVB-T networks
While the EC's report makes some valid points, most notably over a single standard delivering economies of scale, the effective imposition of a standard both flies in the face of policy elsewhere (the EC calls, for example, for a "level playing field" in terms of content obligations for mobile TV) and hamstring players who feel that MediaFLO or another standard might be better suited to their particular circumstances.
The availability of spectrum is a key factor in determining the success of Mobile TV. This is especially true in the case of DVB-H, which plans to operate in the UHF spectrum band currently occupied (depending on the country/region) by analogue TV. However, the availability of UHF spectrum varies widely across Europe: this is partly a reflection on the relative popularity of analogue terrestrial services in the member states. Thus, in the Netherlands and Luxembourg for example, where less than two per cent of customers used terrestrial TV services, relatively little disruption was caused to consumers when analogue signals were switched off. By contrast, the UK has traditionally relied on terrestrial broadcast TV services with a comparatively low adoption of cable, and even though the country has the highest level of digitalization in the EC - 80.5 per cent of households were digital at the end of March 2007 - that left more than 4.6m households dependent upon analogue terrestrial transmissions for their television.
A consultation paper released by Ofcom in December 2006 on the Digital Dividend Review (DDR) suggested that L-Band spectrum (scheduled to be auctioned in the UK by the end of 2007) might be suitable for mobile TV. However, UHF spectrum is perceived as ideal for DVB-H because signals can travel comparatively long distances (thus minimising the required number of repeaters), and can penetrate walls without noticeable degradation. By contrast, signals in the L-Band (1452-1492MHz) travel shorter distances (more repeaters, therefore a more expensive network) and are less effective at penetrating buildings Furthermore, if the UK unilaterally decided to utilise L-Band spectrum, this would immediately put it out of kilter with neighbouring countries committed to using UHF spectrum for DVB-H. Thus, in the UK at least, there appears to be an impasse: Ofcom will not release the optimal spectrum; L-Band is perceived as too expensive.
The start-up costs for a mobile TV network are not insignificant, regardless of the broadcast standard employed. In the first instance, operators may be obliged to pay a fee for spectrum over which the service will operate. Secondly, there are the network infrastructure (construction and implementation) costs. Thirdly, the backhaul costs of distributing the mobile TV content to the broadcast stations will be substantial. The cost of the spectrum will vary significantly by market and the relative desirability of the spectrum in question. In the US, Crown Castle acquired nationwide spectrum in the L-Band for $13m; the UHF (channel 55) spectrum acquired by MediaFLO cost $38m. In the UK, Arqiva recently acquired four lots of 4MHz spectrum at 400MHz for a total of £1.5m ($3.0m).
However, these costs are minimal when compared with the network rollout costs. The La 3, Hutchison Italia DVB-H network cost around $280m, while Mediaset's DVB-H network cost around $320m. The MediaFLO network in the US was estimated to cost somewhere between $800m, while Modeo would have been obliged to spend in excess of $1bn. Costs also vary depending upon the spectrum utilized: as noted on page, the different propagation attributes of the spectrum bands means that L-Band services require a greater number of terrestrial repeaters. Even for relatively small countries, certain topographical characteristics can ramp up costs.
The sheer scale of these rollout costs make it imperative for operators to share infrastructure. Let us assume a single network rollout in a medium-sized country (e.g. France), costing around $400m (DVB-H UHF spectrum) or $600m (L-Band). In either case, it will take an operator five years until cumulative service revenues exceed these rollout costs alone (regardless of other expense such as backhaul costs, programming costs, etc). The construction of two sets of competing infrastructure would therefore be uneconomic, suggesting that the best model would be for competing service operators to share network infrastructure to minimize costs.
Clearly the costs outlined above are dependent upon the breadth and intensity of network coverage involved. One of the key factors that can ramp up the cost is providing an acceptable level of quality of in-building coverage. This has been an issue for most cellular networks, and with higher 3G frequencies and a higher degradation of the signal through thick walls and glass, an increasing number of cellular operators and building owners are resorting to In-Building solutions like Distributed Antenna Systems, Pico Cells and Repeaters. The same issue of indoor coverage applies to broadcast mobile TV technologies like DVB-H, DAB/DMB, ISDB-T and MediaFLO. Although it is too early to say, the degree of the problem and which technology will eventually face an indoor coverage issue, we believe that the success of broadcast mobile TV is very much dependent of the user experience. As the various trials of mobile TV have illustrated, the majority of mobile TV viewing is expected to happen at the office, at home or while in a bus or train. This means that the user will at most times be inside an enclosed space (whether stationary or at speed). Thus it becomes very important from a network design point of view.
However, the comparative analysis of mobile TV standards is very much a game of claim and counter-claim, with the adherents of each particular standard citing research that purports to show that their technology is superior to that of the competition. Regardless of the veracity of these various claims, the fact remains that mobile TV will not achieve its full potential unless the selected technology offers an acceptable audio and video quality over the majority of a country's territory and to subscribers watching its service within buildings and moving vehicles. Ultimately, customers are paying for the universal availability of a mobile service; the premium they will pay is to receive that service anywhere, anytime. This is a problem that has yet to be fully addressed by 3G networks, where streamed services are regularly interrupted when users move out of areas with good coverage. It is an issue that must be addressed at the outset when establishing a dedicated mobile TV network. To put it another way: customers of a DTH or cable pay TV service would not renew their subscriptions if such services regularly and repeatedly broke down.
Key to the mass adoption of mobile TV is finding an acceptable price point for services: too high and users will be disinclined to subscribe; too low and service providers run the risk of failing to extract maximum value from their service. In January 2007, the Mobile DTV Alliance released a white paper claiming that:
"When prospective subscribers are asked if they are willing to pay $20/month for the service, before experiencing the service first-hand, only 10 per cent on average respond positively. However, once holding phones in their hands with real, live, high-quality broadcast services available, this number should change, and more than 50 per cent will be willing to pay for the service, as worldwide commercial trials (Italy, Finland, UK) have shown. What will the mobile broadcast TV revenue of US operators be if 25 per cent of their subscribers are willing to pay $20/month? With close to 200m US subscribers, the annual revenue will be $12bn. Even if we assume that 50 per cent of that revenue will go directly to the content owners, and the entire investment in infrastructure (estimated at anywhere between $500m and $2bn) is to be amortized over one year - there is plenty of profit to share."
This argument is flawed because it places too much faith in comparatively small survey groups, and secondly, in reality, adoption of such technologies is almost always much lower when the reality (I must now pay $20) takes over from the hypothetical (I will pay $20 in the future). It is true that, in Italy, a monthly subscription to the service retails at P19 ($24.77) which on the surface would suggest that the Mobile DTV Alliance case holds water. However, this price also includes 1GB of bundled data, and - most pertinently - more customers are tempted by offers which bundles mobile TV with low cost telephony, or else (seeing that they have made the investment to buy a broadcast TV-enabled set in the first instance) are purchasing three month subscription at which point the effective monthly spend falls to P9.67 ($12.61). Bearing in mind there are already some casual (daily/weekly) users of the service in the market, and taking 3 Italia's own figures, the effective spend attributable to broadcast mobile TV is around P8.3 ($10.8) per user per month.
Certainly, within Western Europe at least, customers seem unlikely to baulk at a monthly subscription of P8-10 ($10.43-13.04), provided that the service offers good coverage and service quality. Furthermore, operators seeking to retain customers could emulate 3 Italia and bundle mobile TV in as part of a wider content/data bundle, although one caveat is that in so doing they should ensure that they do not undervalue the mobile TV element of the bundle (particularly given that in most cases they will have been required to invest significant capital in any mobile TV venture).
Given the high cost of handsets, it is likely that most initial customers will be postpaid, with operators in a position to offer further incentives for customers to renew subscriptions by offering the high-end handsets either free of charge or (more likely in the first instance) at a significant discount as part of the subscription package. In addition, operators should make mobile TV available on a one-off basis to encourage revenues from casual users of its postpaid services, who might just want to watch occasional sporting events or might be tempted to "dip their toes in the water" with one or two viewings before signing up to a monthly subscription. In such cases it makes sense to offer a varied tariff, with certain programmes (i.e. the aforementioned sporting events) priced at a premium to news and soap operas.
Conversely, for mobile prepaid users, it makes sense for operators to seek to emulate postpaid subscriptions by offering subscription-type tariffs, with customers paying in advance (a la 3 Italia) for access to the service for 24 hours, a week, or a month.
One additional factor to bear in mind is that for simulcasts of pay TV services, the premium charged to existing subscribers of those services (i.e. those who already receive them via DTH, cable or DTT) will need to be lower than that charged to new customers: i.e., the premium they pay will be a mobility premium rather than a content premium. Thus, customers who already pay $50-80 per month for a Comcast, Viacom or BSkyB package will be reluctant to pay an additional $15 per month purely for the benefit of receiving that content over a mobile.
However, operators and service providers should instead see this an opportunity, ultimately offering the opportunity to bundle services as part of a triple or quad play package by locking customers in to a pay TV package across different media. Thus, for example, in the UK Sky Mobile offers existing Sky TV subscribers the opportunity to purchase themed bundles of Sky channels at £5 ($10) per bundle, thereby reinforcing its relationship with those customers but within the mobile environment. While not currently offered to non-Sky TV subscribers, it is possible that such bundles might be offered in the future although at a considerable premium.
The above "ifs" are not insignificant, and there are others, equally problematic, which are, unfortunately, not going to disappear overnight. That said, if and when operators and service providers have addressed those "ifs", have ticked them off on their "to do" lists, then the opportunity afforded by the small screen of mobile TV appears large indeed.
Dr Windsor Holden is Principal Analyst with Juniper Research, and can be contacted via: firstname.lastname@example.org