INTERVIEW: Alex Nourouzi,  Senior Consultant, Ovum "Packet is going to drive the mobile data market."

by Judith Berck, GSM Data Today, October 10, 1998.

Alex Nourouzi is a Senior Consultant for Ovum’s Mobile and Satellite Group. He was an author along with John Davison of the recent study, Cellular Data Services: Market Opportunities for Packet Upgrades (1998). GSM Data Today interviewed Alex to talk about the report.

Nourouzi discusses GPRS and the impact it may have on the data market, the potential for growth in the data market as well as barriers to it, the investment decisions operators face in upgrading their networks, vendor initiatives and much more.   Included below is a short summary and excerpts from the report, followed by our interview, which includes excerpted graphs from the report.

To purchase the full report, Cellular Data Services: Market Opportunities for Packet Upgrades (1998), visit Ovum’s website.

Report Overview:

With mobile operator revenues per subscriber falling despite a general increase in subscriber numbers, operators are looking to data applications to raise revenues. Current 2nd generation mobile systems have the capacity to support only voice and low-speed data applications such as messaging. While 3rd generation systems promise to deliver high speeds, they are still some way off. In the meantime, cellular operators have the option to invest in network upgrades, enabling them to use higher speeds and Internet protocols and deliver early multi-media services. For those upgrading from GSM, the preferred option is currently GPRS - however more than one upgrade option exists. The financial rewards are great but so are the risks.

Ovum’s new report looks at the decision operators face in relation to these upgrades, maps out Ovum’s views on different market scenarios and identifies the triggers that should govern strategic decisions. It forecast the global market in fifty of the key countries around the world as well as the different technologies.

GDT:     What’s the purpose of Ovum's new report?

AN:        Our report is really looking at packet data generally for all cellular networks. It looks at the possibilities facing mobile operators to upgrade to offer these packet data services. In the GSM environment, we’re talking about GPRS, and how operators should implement it straight away. And how packet is going to drive the mobile data market.

There has been talk about the mobile data market growing for the last five to ten years. It’s been one of the most disappointing areas for GSM. Certainly, in terms of mobile data forecasts, they have consistently under-performed. And I think that now we’re seeing a number of things appearing on the horizon which suggests that this time it’s actually going to go ahead.

GDT:     What do you see appearing on the horizon that suggests wider adoption of data is coming?

AN:        One of the key things is the introduction of packet data capabilities such as GPRS within the networks -we’re seeing the network architecture coming into place. Another key is the Internet. People are increasingly using the Internet for data services and browsing. But also, corporates are installing their own intranets for storing information and people are regularly accessing that sort of information. People are becoming more familiar and comfortable with using data services. And that’ll obviously spread onto the mobile networks. The Internet will certainly be an important platform for delivering mobile data services. So the fact that GPRS is IP-based and lives well within the Internet world is quite key.

Some of the other elements are the handsets. Up to now they’ve had SMS [short messaging service], which is pretty clumsy -- there’s no keypad, you’ve only got three lines or so on a display that surely isn’t big enough for information services. Now there’s a lot of experimentation on the handset side of things. On the computing side, you’re seeing movement towards the palm tops and smaller laptop devices made with the cut down version of the Microsoft operating system running WinCE. So we’re seeing a movement towards smaller, data friendly terminals.

So those are the types of things building up on the horizon. Plus the mere fact that there has already been some experimentation with packet data in the U.S. with CDPD. It’s been pretty much a failure from what we can gather, but there are lessons to be learned there, and experiences that people can draw upon.

And also you’ve got third generation coming up as well, which is going to enable us to have two megabits to the terminals. For people to invest in third generation, they have to prove that there is a market for data services. And by getting GPRS and getting people interested in their mobile terminals for things other than voice, and getting them dabbling into data, they are getting their interest levels built up before they start to invest. It’s a stacked process.

GDT:     What year do you think will be the year mobile data takes off?

AN:        I think most of the operators are talking about implementing packet data in a big way in the year 2000. And I think that from 2000 onwards, that’s when it’s really going to take off. It’s interesting. It happens when the operators become convinced that there’s a market. I think it’s coming together now. The operators are starting to hold pretty serious discussions with the vendors with regards to packet data. I think that toward the end of 1999, 2000, that’s when we’re going to start to see the first services.

GDT:     What needs to change for mobile data to take off?

AN:        It’s not subconscious at the moment for the end user. You have to think about getting the software configured right. Is it a problem with the cable? Is it a problem with the phone? Is it a problem with the software? Is it a problem with the network? It’s not subconscious. Once all those ease of use problems are overcome, then it’s subconscious. It’ll just be another service. And that’s the sort of stage we’re going to get to.

We can see GPRS and packet generally, whether over CDMA or TDMA networks, becoming inherent within the phone, and becoming quite subconscious, a little bit like the way SMS is being used at the moment. People are going to become subconscious data users. That’s when we’re going to see the big growth in data services, when they get used to using the devices.

And also once users become more accustomed to using the Internet for business transactions or e-commerce, then again they’ll want to do that over their mobile devices. A lot of this is a perception problem, a perception about the security element. It’s perhaps just as secure giving out your VISA card number over the phone to someone that you don’t know.

Another problem is that users’ expectations are set by the fixed networks, and those expectations will have to be dampened in terms of data speed, reliability and quality of service. I think that the fixed networks are going to act as both friends or foes as it were for mobile networks.

Another crucial area is the service integration - whether the standardization process is going to be split - whether WAP goes one way, and someone like Microsoft goes another way. Those two will really be key in determining the market.

The operators’ willingness is another element, getting the operators interested. They’re still very much focused on voice and pricing, which is true because at the end of the day, their core business is voice and will continue to be voice. The users don’t know what they want and the operators don’t know why they need data services, and they’re going around in a loop. On the other hand, the markets have been growing very fast for traditional voice connections, and operators are coming up with problems with high churn rates and falling average revenue per subscriber. It’s now that they’re starting to get a little bit twitchy maybe, and turning to value-added services and data services. I think that’s quite a key element. They’re now looking at other revenue streams and mobile data seems to be one.

In terms of the operators, lots of cultural elements have to be handled -- the mobile operators are not familiar with IP and the Internet. That’s one of the key things they’ve got to get into - starting to understand IP and the Internet, energizing the sales channels, getting them interested in sending data. And just following generally the movements of the Internet over fixed networks such as e-commerce with a view that this is what will happen over the mobile networks,. It will mean a dramatic change in their business. You can draw parallels to cable TV companies moving from not only being in the entertainment industry, but also getting into telephony.

With mobile data, the network will just become the actual pipe to deliver the information. And it’s the actual information, the content which will be interesting. And whether operators, mobile operators, are going to start to embrace full partnerships with content providers. AT&T Wireless Services is doing that with that Pocket Net service using CDPD. They’ve got a number of partnerships with content providers like Bloomberg providing financial information. So those are the types of things we’re starting to see.

Data’s a very complicated thing; it’s unlike voice. Ultimately, the distribution channels have got to be interested and they’re still all geared up to sell voice. You go into a retail outlet which claims to be serving GSM services and you mention the word "data" and they’ll rummage around at the bottom of their drawers and they’ll have one little sheet that has data, and says "well, this is data. We don’t know what it means, but here it is."

So, it’s got to be simplified, maybe a simple bundle package, with all the software integrated onto the terminal, with the PC card integrated into the terminal. Off the shelf devices, which they haven’t been in the past. A number of these little pieces in the jigsaw have got to come together, and the vendors are working very hard to put them together and present them to the operators.

It’s no surprise that it’s really the infrastructure vendors who are driving that, the Ericssons, the Nokias, Motorolas of this world who are keen to sell packet data upgrades, but also to insure that there’s going to be end to end connectivity with the Internet. The vendors are doing their utmost in terms of providing examples of applications which have worked, putting together standards forums, which will help the movement towards packet data, such as WAP (wireless access protocol). Bluetooth technology is another vendor-driven example - again it’s driven by the likes of Ericsson and Nokia to simplify the whole process for the users and operators. So they’re doing everything to encourage the adoption of these upgrades, and to reassure the operators.

 GDT:  What about HSCSD (high speed circuit switched data) which will appear even sooner than GPRS - will that take off?

AN:        It’s an interesting one. There is certainly a small market for HSCSD. Arguably the most valuable customers are going to be wanting these types of services, the high-end users, the corporate customers. But I’m not convinced that many operators are going for it. High speed circuit switched data doesn’t do anything to ease spectrum capacity constraints that operators are facing.

We believe that the key with GPRS is the packet data element, rather than high speed, because it uses the spectrum in a better way. You’re not tying up a whole channel end-to-end for one user. You can get multiple users in there. And operators should be looking to build their networks that way, rather than offering very high speed services.

GDT:     What services might operators offer with GPRS?
AN:        It could be for the information services we’re already seeing -- that could be done through GPRS. A lot of the SMS-based services at the moment could be done over GPRS. That’s a natural evolution path. And we’re now seeing early introduction of location-dependent information services, which between now and 2000 I think is going to be very important and beyond that perhaps.

GDT:     Do you think that GPRS will also drive processor intensive devices like laptops?

AN:        Yes. Our theory is that people are going to have lots of different terminals depending on their applications. You might have a wrist watch for voice applications. You might have a small calculator-sized terminal which only needs 9.6 kilobits per second just for general information services. And a laptop which might need 64 kilobits per second. GPRS will handle that, or some of that might be on HSCSD.

It’s like the way that people have lots of different types of bags. If you’re going away for a holiday, you’ll take a briefcase. If you’re going to a conference, you might take a laptop with a big QWERTY keyboard. If you’re going on a one day business trip, you might just get away with a small WinCE-type device, that might be comparable with an overnight suitcase. And then you have your wallet, which you carry around all the time, which is just for carrying your money and your cards. It’ll depend on what you want and your requirements.

GDT:     What is your outlook for data beyond the year 2000?

AN:        In terms of some of our projections, we think that in about 2007, thirty percent or so of mobile users will be using data. That’s a high number. Whoever has a mobile subscription, one in three mobile users will be using data of some sort. But in terms of how important data is, I think voice is still going to be the key in 2007. Data is only really going to be about ten percent of revenues we believe.

GDT:     You’re talking about GPRS in the year 2000, but do you think the US is going to be on the same timeline as the rest of the world?

AN:        I think the U.S. is perhaps a little bit behind Europe. Not to say there isn’t pent-up demand. But it seems that as a lot of second generation networks are only just being rolled out, operators are focusing perhaps more on voice. So, I think that perhaps a little bit behind, but I think it’ll happen at a relatively similar time. A lot of that again, is user acceptance and how competitive the market is again.

In terms of the operators, it has very much to do with what’s our competitor doing? He’s installing GPRS from day one. We’ve got to do it, because otherwise, we don’t want to appear to be the slow coach in terms of rolling out new services. Some of the companies which were first to roll out CDPD, for instance, have not benefited massively from being one of the first to offer these services. There is a case definitely for watching what other people are doing in perhaps less competitive markets.

GDT:     How do third generation issues play into the discussion about upgrading to packet data?

AN:        Some operators are asking "do we go through the second generation upgrades or do we wait for third generation? Are second generation upgrades definitely a step towards third generation or will it be a completely different network?" The vendors are saying, "go with the packet upgrades. It’ll be a step in the direction towards third generation, you’ll be able to reuse that network".

Well, the third generation standardization process hasn’t been completed yet. So the vendors cannot say at this stage, until the third generation standardization process has been completed, whether you be able to re-use your whole network. It is a step in the direction definitely of moving to third generation. A key thing is that GPRS is essentially a separate network. You have the movement towards IP and the movement away from the switched environment towards the router environment. And obviously third generation will build on that, but how much of the network you can actually re-use is a key question, and something that perhaps eating into the operator’s mind. It’s the reassurances that the operator needs.

What’s clouding the situation is that companies like Ericsson are saying, "we propose third generation speeds over second generation networks - with changing modulation schemes, EDGE, you’ll be able to offer 384 kilobits to a moving terminal, the same as third generation to a moving terminal." So why invest in third generation? And I think that some of the TDMA operators are going down that route. They’re going to want it to be an evolutionary process rather than a revolutionary process, reusing as much of the network as possible. And I think that they’ve perhaps been scared off buying.

For some operators the third generation drivers are capacity constraints in city areas. For them third generation has nothing to do with high speeds or mobile data multimedia but with more capacity for voice services. You could argue that in Japan. NTT DoCoMo is suffering now with massive capacity constraints, particularly in Tokyo. Some operators see it as the immediate, short-term solution for spectrum for voice.

GDT:     What do you think third generation standards will look like?

AN:        I   think you’ll have different flavors, a family of IMT-2000 standards which will hopefully be compatible. And I think that’s the way it’s going to have to be. The vendors at the moment are developing solutions for packet data. The operators will only buy that if that’s a step towards third generation. If third generation means something completely different, then they’re not going to go for that.

Another key element is interworking between second generation and third generation. UMTS and GSM interworking is key because there are going to be islands of coverage initially, so they’re going to be falling back onto second generation networks. And so, there’s a lot of baggage being brought into these discussions.

© 1998 Intel Corporation. Judith Berck is an employee of Intel corporation.