How to minimise the 3G risk: By accurately forecasting data usage, it will help the 'pure' 3G players to roll out their networks cost effectively - 3G Networks

Telecommunications International, Feb, 2002 by Damien Rouchouse, Nicolas Faesch

* permanent services: since the mobile phone offers a permanent 'on-the-move' connection, operators need to design their services accordingly (ie, they are in real-time, interactive, are easy to respond to, and have 'alerts' where necessary).

* personal services: given that a more or less identified customer is behind each mobile subscription (even for prepaid), personal services to a mobile phone is a key differentiator from any other access device -- with the PC or TV, you never know who is behind the screen!

* positional services: operators who leverage the location-based information of their users are more likely to have attractive services for their customers.

* payment services: operators need to remove all the barriers for m-payment, making it easy, secured and guaranteed.

Market research will not bring 'off-the-shelf' answers, but it should provide enough information for the operator to undertake network and service rollout decisions with conviction.

Traffic modelling

By undertaking market research, it should be possible to identify probable future service usage patterns for mobile multimedia services. The next problem is how to establish projected network traffic patterns from the information gathered and then dimension all the technical elements accordingly.

To illustrate how difficult this task is, let's take a look at music services over 3G networks.

Such services will enable customers to do various things, such as listen to extracts from the top ten; download mp3 music files; watch video clips; and receive news updates from favourite bands. However, because such services are not yet available, forecasting their usage (and implications for the network) is not easy.

Nevertheless, as revealed by marketing research conducted in Scandinavia, it is possible to establish -- with conviction -- what the probable usage pattern for music services will be. If we focus an the youth segment, for example, you can analyse what kind of information they look at in the newspaper, how many compact disks they buy every month, if they attend music concerts, and -- most importantly -- how much they already spend on 'music' in a month.

Extrapolating from primary market research conducted by our client, it was found that teenagers are ready to spend around [euro] 6 per month on 3G services. Of that sum, they expect to spend around ten per cent on 'music'. The issue then is how this music budget will be spent. Will it be on reading music news? Listening to music extracts? Receiving the latest information? By understanding this distribution it is possible to define what music services should be in terms of delivery method.

In the past, dimensioning GSM networks by translating marketing usage into technical inputs was not so difficult given that voice traffic and usage for basic value-added services (such as voice mail) are well known today, backed by years of commercial experience. With a basic knowledge of what a busy hour is, dimensioning a 20 network became a piece of cake (at least compared to 3G issues!).