Analysys Mason: A structured approach to data pricing is the first step to profitability

Amrish Kacker

Figuring out how much data costs for operators is not an end in itself, but it is the basis to understanding the key drivers that affect data costs and the cost boundaries within which the business can operate.

Data profitability is a hot topic among mobile operators because data traffic is beginning to dominate mobile networks. Figure 1 captures some of the key questions being addressed by operators, covering service profitability, business profitability and dealing with multi-year costs.

Figure 1: Key questions on data profitability (illustrative charts) [Source: Analysys Mason]. Click on the image for larger version.



A structured approach to costing can provide the information required for sound decision making, and needs to address the following key areas.

  • Cost definition: what is the appropriate cost base? Operators need to decide the scope and the timeframe of the costs that are to be included--these will be determined by the objective of the costing e.g. short-term pricing opportunities (marginal costing), long term data profitability (full costing) or voice termination (regulatory costing). Operators must also select a modelling approach for costing--either a bottom-up approach based on unit costs multiplied by traffic, or a top-down model that is allocated across different services or businesses. For any long term or regulatory costing approach, multi-year costing approaches are necessary.
  • Cost recognition: what is the appropriate depreciation methodology in recognising the asset lifetimes and utilisation? Historical cost accounting provides a good view of existing assets, but it does not take into account future network expansion or asset replacement cycles. In addition, a straight-line depreciation methodology distributes cost equally over time, whereas network utilisation increases over time. This tends to increase costs in the short term, but reduce costs in the long term. The pricing challenge for operators in the short term might be market competitiveness, while in the long term, it would be ensuring that they do not under-price the service.

    An operator's approach to cost recognition must take into account practical asset lifetimes. There are a variety of approaches that an operator can take to spreading costs across multiple years for capital investments. An economic depreciation approach provides a fair representation of spreading capital costs across multiple years. In an economic depreciation approach, as the usage of an asset increases over time, so does the recognition of cost in that particular year--with the principle that there is a constant cost per unit of asset utilised across the asset lifetime. While this provides a balanced recognition of costs across an asset lifetime, it relies on accurate data usage forecasts, which is still an evolving art. More conservative operators may decide to use some form of a tilted annuity recognising higher costs upfront as future equipment tends to be cheaper than current.
  • Cost allocation: what is the appropriate basis for allocation of common costs across services? The first stage of cost allocation typically involves distribution across technologies (2G/3G/4G), followed by services. Operators must first decide whether to adopt a fully allocated basis or an incremental basis for cost allocation. It is important to identify the appropriate drivers, taking into account not only the current network view but also its evolution, in order to ensure that there is no cross-subsidisation of services. One of the biggest challenges is in dealing with network opex , a large proportion of which is site related costs. The current approach for most operators is to allocate site costs across technologies based on presence on a site. This approach could result in a large proportion of costs being allocated to 2G networks--which primarily carry voice--and therefore to voice services. Over time though, voice services are likely to make up a smaller share of total revenues and an extremely small proportion of total traffic. Therefore, there is a consideration on whether traffic-based allocation is more appropriate even for site costs.

These boundaries provide the business with a framework for focusing on service pricing as well as cost reduction, which is the most effective way of increasing business value while also protecting profitability.

Some of the key areas where understanding data costs is supporting business decisions include:

  • Tiered pricing: Migrating users from large or unlimited bundles to tiered packages is an ongoing process. The balance the operators need to strike here is to provide large enough bundles for customers to feel that their data usage will be significantly lower than the tier cap, but will also be profitable for the operator as customer usage converges with the caps over time.
  • Technology migration: Data costs must take into account technology efficiencies of migrating users from 3G to LTE. The expected customer data usage and lower costs of new technologies can inform decisions for the operators on the potential speed of technology migration.

Analysys Mason engages with mobile operators and regulators in understanding data costs by undertaking detailed modelling as well as facilitating the understanding of the appropriate trade-offs in data costing assumptions.

Amrish Kacker leads Analysys Mason's Operator Strategy Consulting with work in the Asia-Pacific region. He has been with Analysys Mason since 2000 and moved to Singapore from Analysys Mason's Cambridge office, UK, in 2006. Amrish specialises in supporting board-level investment and strategy decision-making in the telecoms sector, and has wide experience in working with fixed and mobile operators across the world.