Qualcomm's Ambitious Mobile TV platform - Blueprint for Others?
Last month, Qualcomm purchased at auction 40MHz of spectrum (1452-1492 MHz, known as 'The L-Band') for £8.3m ($16m). Since then there has much speculation about Qualcomm's motives and the services that they will deploy, focussing upon Mobile TV. The answer tells us a lot about how new platforms and intermediaries emerge, and could hold some keys to the future of a far wider set of services than just television content.
Qualcomm has a long track record of both buying spectrum and vendor financing, in order to seed and kick start the market for new technologies that they have developed. We believe that Qualcomm will use the L-Band spectrum as their personal seed-bed for mobile TV in the UK, and as a potential beachhead into mainland Europe. In addition, 40MHz is more than enough spectrum to deploy not just mobile TV, but also other services in the future.
We believe Qualcomm understand the needs of a platform business much better than other comparable companies, having experienced the ups and downs of initiatives like BREW. We do not expect Qualcomm to deploy either a typical "over the top" offer that bypasses the operators, or a vertically integrated solution. Rather, the story behind the L-Band spectrum is how Qualcomm will try to create a platform which offers opportunity and profit for all the main players in the value chain: consumers, broadcasters, networks and device makers.
Here's why, and how...
Network Build: plenty of options
Cell sites are typically a major cost hurdle to overcome in deploying wireless networks. For a new wireless entrant hoping to deploy a competing technology, such as WIMAX, getting the necessary planning permissions and building out the cell sites could be a business case killer. Not so for Qualcomm and mobile TV, as every mobile operator in the UK is looking at sharing costs to reduce opex. We're sure that between the major UK cell site owners (T-Mobile/3UK, Vodafone/Orange, O2 and Arqiva/NGW) there will be enough competition for Qualcomm's business not only to produce near-ubiquitous coverage, but also reasonable site rental charges.
Most of the UK mobile operators are currently investing lots of money getting more backhaul capacity from their cell sites, especially as increasing 3G data demand are creating much more traffic than historically was required for pure voice and SMS services. This backhaul requirement could definitely be a bottleneck. But for Qualcomm's mobile tv solution, MediaFLO, there remains the option of delivery the broadcast signals via satellite. FLO stands for "Forward-Link-Only", and the return path for interactivity is delivered via operators' 3G networks. This type of satellite delivery mechanism is in fact behind Alcatel's extension to the mediaFLO competitor, DVB-H, called DVB-SH.
We seriously doubt that Qualcomm would have to employ an army of engineers as outsourced network deployment is a typical service offering from nearly every major equipment vendor and whole swaths of sub-contractors. Qualcomm has even more options for operating the network once deployed with companies such as BT and NGW/Arqiva offering fully outsourced managed operations as well as the usual suspects from the equipment world.
Minimal New Technology Risk
Qualcomm already has a mediaFLO network in the USA providing services to both CDMA (Verizon Wireless) and GSM (AT&T) operators, therefore the technology for handsets, base stations and NOC (Network Operations Centre) is already tried and tested.
The effort in developing the NOC technology should not be underestimated as this is the nerve centre for the whole platform: from taking broadcast signals for the content provider and converting them to mediaFLO; managing non-broadcast data, such as the Electronic Programming Guide (EPG), billing transactions and usage data; and to provisioning services on end-user handsets.
The UK spectrum is at a different frequency to the USA and therefore some work would be required to optimise to the transmitters, but this would should minimal.
Qualcomm has a great advantage with getting services into handsets, in that it is a chip maker in its own right. Qualcomm already produces a UBM (Universal Broadcast Modem) chip which supports not only mediaFLO, but the other major broadcast standards (ISDB-T and DVB-H). The availability of this chip reduces the risk for handset makers in getting-to-market, vitally it reduces the need for many different handsets serving different countries and thereby improves the economies of scale for the handset makers.
Qualcomm also offers technology options for the content providers in supporting multiple different conditional access systems: Irdeto, Nagravision and NDS. Content providers are extremely nervous about piracy and by providing the options that are currently known and in use removes the worry that new age DRM solutions bring - not only the risk of being cracked, but also introducing a new gatekeeper.
Getting the Content
Qualcomm will need to get compelling content onto the platform before consumers will even contemplate watching mobile TV in any form. Here we believe that choice and flexibility for the broadcasters will be absolutely vital - not only in DRM, but also in the business model.
The UK has a wide mix of content providers using a combination of state licence fees, subscription revenue and advertising as their funding method. In addition, pay TV players such as Sky and Virgin Media already play a rich role in aggregation.
We expect that Qualcomm will be very innovative, especially compared to other mobile TV platforms, in making their platform attractive to the content provider community. For instance: the BBC may just wish to rent a set of channels from them to broadcast free to all content; Sky may wish to rent a set of channels some of which are broadcast free to their pay TV subscribers; ITV may wish to rent a couple of channels, but pay for them with a share of advertising revenue; and others, especially new entrants, may wish to have a hybrid type of arrangement sharing both subscription, pay-as-you-go, and advertising revenues.
Again, Qualcomm will be able to learn lessons from the USA market, where the market is at least as complicated as the UK.
What is the carrot for the mobile operators?
Mobile Operators are the key gatekeeper for the whole platform:
- some of them have visions of complete vertically integrated fixed and mobile solutions, e.g. OrangeTV;
- mediaFLO requires a 3G bearer to act as the return path for interactivity and ordering;
- mobile operators usually define the handset specifications to run on their networks and furthermore fund a lot of handsets with subsidies;
- operators' assets such as customer care, billing (especially prepaid) and authentication will be extremely valuable.
At a bare minimum, Qualcomm needs to find a way of compensating for the handset subsidy, data traffic and any use of Operator assets. Beyond this operators may want some type of deal for content exclusivity over a period of time, which would probably be negotiated directly with the content players, but still requires enabling platform features.
Operators may request a share in any upstream revenues, such as advertising and subscription, especially for the privilege of direct marketing services to their customers with a handset that supports mobile TV.
Of course changing video from unicast to broadcast will remove strain (and therefore cost) on the operators own 3G networks - one day these networks will be filling up and anyway unicast is a hopelessly inefficient delivery mechanism to any reasonable sized audience.
There is also a big case for mobile TV as a future solution to in-car entertainment, whether served by a mobile operator, new start-up or the car manufacturers themselves.
Consumer demand - where the real risk is
Consumer Demand remains the greatest risk of all - the hype behind initial deployments of mobile TV is over. We sense a growing apathy within the operator community for such services.
Italy is claimed to be the most successful mobile TV market in Europe with 700k paying subscribers as at August 2007. Public numbers in other countries are hard to find.
There is plenty of demand for mobile TV, if compelling content is available at a reasonable price. One of the core uses of a mobile phone is to fill dead time. However, we believe that MobileTV will not be an overnight success story - an optimistic scenario would be 20% penetration within 5 years of launch, i.e. 10 million UK users. Most of this would be snack-type consumption and be nowhere near the time that people spend glued in front of the box in the home. Perhaps 20 minutes per day on average is all that can be expected. Since the platform supports rich interactivity, is highly personalised, and offers a return path for marketing messages, the revenues will reflect this accordingly.
Whether this type of demand is enough to sustain mobile TV is seriously open to question, but this level of uncertainty actually favours the emergence of a single platform open to all players in the value chain. We believe this is the Qualcomm proposition.
We also believe that the direct revenues for mobile TV will be hard to determine - hidden within mobile, pay TV and licence fee bundles. The indirect revenues such as advertising and handset prices will also be hidden in bundles that will be even harder to determine.
In the MediaFlo technical specifications Qualcomm claim to be able to broadcast 32 channels in an 8MHz slice. Even if we reduce this to 20 channels, because of the higher frequency of the L-band compared to the UHF spectrum, Qualcomm has tonnes of capacity. We seriously doubt in the medium term that 100 channels of capacity are required for MobileTV.
Because of this excess capacity, we would expect Qualcomm to reserve some capacity for their own use in the future for other services, whether tailored to media content delivery type services, or more general two-way communications services. Qualcomm effectively have their own test bed for next generation services for very little cost, something that other wireless technology companies would only dream of.
One to watch
Qualcomm's use of the L-Band will be a fascinating case study of the development of a multi-sided platform. The initial interest will be how Qualcomm makes the platform attractive enough to all players to generate enough consumer demand. Observe how Qualcomm tinkers with the platform price levels to create the externalities over time: more content, more value, more eyeballs.
However, the revenues and costs of the platform will be hidden from analysts eyes and therefore to determine whether the platform is a success or not will difficult to determine for a long time to come.
Qualcomm's history with the BREW platform shows that they have the track record to satisfy enough players to make a platform work - and we have very little trouble in believing that somehow they will make a success of mobile TV in the UK and make it a mass market service.
However, there is a word of warning to all existing players - new services can create new roles for powerful middlemen. Just as Google at the beginning looked like an innocuous search engine and then morphed in some all-consuming monster with an insatiable appetite for advertising budgets...