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

If 'pure' 3G players are to have any chance of making a significant profit, they must get their network rollout plans right. Needless to say, the network and marketing challenges confronting a 3G operator with no legacy infrastructure will markedly differ from incumbents.

By definition, the pure 3G players will be last to enter the market. And when they do arrive on the scene, they will only have a small window of opportunity to make their presence felt on an already mature market.

But there are similar challenges confronting incumbents and the pure 3G players. Both, for example, need to invest considerable sums of money on 3G infrastructure and live with the pressure of achieving an ROI (return on investment) in as short a time as possible (although you could argue that this task is made more difficult for the pure 3G operator as its target market will already be equipped with mobile phones by rival operators). And finally, every 3G player must contend with the technical uncertainties (and availability issues) surrounding this new technology.

The purpose of this article, however, is to look specifically at the challenges confronting the pure 3G players. How can they predict traffic usage patterns in order to tailor their network accordingly and deliver service in the most cost-effective way?

The methodology for market projections described below is the result of work undertaken by consultants from Sofrecom, which have worked with Orange SA to support its efforts on UMTS licence applications for different Orange member companies.

In particular, Sofrecom has implemented this methodology for one of Orange's member companies in Scandinavia, which is scheduled to launch 3G services -- from scratch -- in the first half of 2003.

Methodology guidelines

To optimise the dimensioning of the different 3G network elements -- radio access (UTRAN), the core network and the services platform -- the operator needs to forecast data traffic usage.

The most obvious way to do this is to list different services and forecast usages per customer segment by benchmarking behaviour currently observed (on fixed internet access, for instance). This method, of course, is only applicable for basic services such as e-mail -- customers know more or less what an e-mail is and operators have access to a full set of benchmarks.

However, this method is very limited when it comes to brand new mobile multimedia services for three major reasons: the mobile multimedia market is not a subset of the current internet market; mobile operators do not know what they will really be able to deliver at launch (primarily due to uncertainties on customer equipment features and network performance); and customers do not have a good perception of what these new services will look like (which creates a bias in demand forecasting).

To address these issues, we decided to explore another methodology by asking the following questions:

* What are customer expectations for 3G services and usage?

* How much is the customer willing to pay for 3G services?

* How can a 3G operator attract these revenues and what will be the consequences in terms of the network handling increased data traffic?

To answer these questions, the starting point -- as always -- is market research. But how can the potential demand for services that do not exist be assessed? To have any chance of evaluating future 3G take-up accurately, market researchers must interview face-to-face and illustrate the potential services as much as possible (through the use of prototype handsets or live demonstrations, for example).

Once this has been done, the results of the survey should produce market segmentations for those most likely (and most unlikely) to accept the new 3G services. Then, for each identified segment, it will be necessary to get a clear picture on current purchasing patterns for similar services and goods (such as current telecoms budget and purchasing power for entertainment services).

It is important for market researchers to cross check the willingness of interviewees to use the service with a willingness to spend, and then benchmark it by assessing current consumption patterns for similar services or goods. The disposable income per inhabitant won't grow dramatically in the future, but mobile service providers can reasonably expect that new purchasing habits will most likely be encouraged by new entertainment products.

Finally, the projected amount spent on these new services needs to be looked at in the context of the overall economy (a percentage of GDP, for instance]. This check is necessary to avoid over-estimating the mobile multimedia market.

Survey results

By employing the market research methodology as described above, Sofrecom has found that the mobile multimedia market is not a subset of the multimedia market -- it is a brand new one.

We can summarise customer expectations for new mobile multimedia life services by the 4Ps 'killer cocktail', where future applications will have to address the following parameters: