|
"Average Revenue per User" (ARPU) is, today, the accepted metric
for measuring the success of mobile operators. However, in our most recent
study, entitled "AMPU not ARPU: A BETTER METRIC FOR THE WIRELESS
INDUSTRY," we argue that the ARPU metric is insufficient and that operators
may limit their profits by focusing exclusively on it.
"Average Margin per User" (AMPU) provides a basic – but
generally unacknowledged – criterion for measuring the success of wireless
operators. By focusing on AMPU, operators can generate profits sooner –
and at higher rates – than otherwise would be the case.
Over the past decade, the industry has expressed dismay at declining ARPU,
notwithstanding actual increases. Part of this dismay stems from the assumption
that declining ARPU implies a loss in profits.
Few financial analysts have questioned whether low ARPU customers actually
lose money for operators – or how operators could profit even if ARPU
would continue to decline. Nor have they questioned whether high ARPU customers
are consistently profitable. As a more thoughtful observer pointed out,
"part of the problem [of measuring profitability] is [that] the Street
doesn't believe carriers have the data to prove who's a profitable
customer."
With a renewed focus on profits, operators and financial analysts have
shifted their focus to ARPU. This focus, however, is misplaced.
The importance they attribute to ARPU rests on two widely articulated
assumptions: (1) that margins for the lowest ARPU customers are inherently
unprofitable and (2) that new data services will lift ARPU, and with that
profitability. However, both of these assumptions are flawed. By adopting
either, operators may extend losses rather than increase profits. There are two
reasons for this.
First, low revenue per user need not preclude a positive AMPU. In other
words, low revenue users can still be profitable as long as ARPU exceeds average
cost per user. For example, prepay customers have been widely assumed as
unprofitable. Indeed, NTT DoCoMo lists "low rate of prepaid customers"
as one of its advantages in the Japanese market. Prepaid customers may be low
ARPU. However, they may generate higher revenue per minute than do post-pay
customers. In addition, they require no handset subsidies, no billing and
collection costs, and produce minimal bad debt. For these reasons, they can
generate positive AMPU and with that, profits.
Second, even though data services will raise revenues, the full costs of
delivering data may exceed such revenues. If so, AMPU will be negative. As an
illustration, numerous trade press articles have extolled the revenues –
and supposed profits – inherent in UMTS services. Yet, to date, the costs
of providing UMTS have proven a financial disaster. For such reasons,
understanding AMPU becomes essential if operators are to increase profits.
DEFINING AVERAGE MARGIN PER USER (AMPU)
Average Margin per User (AMPU) is the difference between the cost of serving
a user and the revenue that user generates. AMPU can be negative or positive.
The greater the AMPU, the greater the profit. The key to increasing AMPU comes
from understanding the full costs of providing specific offerings.
A multitude of cost factors may affect the AMPU of a given offering. These
might include: (1) network robustness and coverage, (2) the mix of end-users,
(3) the flexibility of billing platforms, (4) customer service and technical
support (or lack thereof), (5) support from sales channels (or lack thereof),
(6) roaming between networks, (7) interoperability of services across networks,
(8) latency in transmitting messages, (9) handset design (user friendly or
unfriendly interface), (10) "quality" of color displays (resolution,
color saturation, color discrimination, and brightness), (11) precision of
position location systems, (12) alternative competitive services (landline long
distance and Internet access), and (13) a regulatory environment that dictates
and/or limits offerings and/or tariffs.
Handset subsidies, infrastructure investment, network operating costs, and
marketing and advertising costs are among the most important.
In some cases, all of these factors will not be relevant. Operators, such as
Virgin Mobile, do not subsidize handsets. Some applications, such as SMS, may
use so few network resources that the costs of the infrastructure required to
deliver them are minimal.
An operator may increase traffic by lavishly subsidizing an expensive
handset. However, the incrementally greater traffic may not generate the revenue
needed to recover the cost of that subsidy.
A new offering may require additional and/or specialized infrastructure. To
the extent that it does, the costs associated with it increase. Given this,
operators must balance the cost of the infrastructure against the revenues the
new offering will produce.
The operating costs associated with different data offerings can vary by
orders of magnitude. Increasingly, operators will face the issue of balancing
what end-users will pay for offerings against the costs to deliver such
offerings.
THE SEARCH FOR THE "KILLER APPLICATION"
The mobile world has talked forever about the "killer application"
– the Holy Grail of operator profits – but no one has ever defined
it until now.
First, the "killer app" generates inordinately high traffic
volumes. Second, it produces especially high revenue (ARPU). Third, it produces
high margins (AMPU). Fourth, its traffic volumes and revenues drive the
construction of sufficient infrastructure, RF and network, to support yet
unknown future applications.
Frequently, voice has been dubbed the killer app. It generates high traffic
volumes, 95 percent or more of network traffic. It produces high revenues, 80
percent or more of operator ARPU. However, in the face of competition, voice
fails to produce sufficiently high margins. If margins were sufficient, neither
the balance sheets nor share prices of so many operators would be as distressed
as they are. Nor is voice driving construction of network infrastructure capable
of supporting future applications. If it were, the voice networks in Europe and
the U.S. would not be so stressed as they are, nor would operators show such
reluctance to construct networks, in particular those to support UMTS.
But the concept of a killer application is a myth and likely an unobtainable
goal.
For operators in a competitive market, there is no magical service that will
lead to excess profits. They must build their networks to enable a plethora of
offerings, each of which contributes to profit. Key to this is selecting
offerings that can produce profits. An AMPU model and analysis would support
this goal.
Objectives of the Study
This study has three objectives. The first is to identify, define, and
analyze major cost factors that affect AMPU, in particular, those over which
operators can exercise control. The second is to quantify the relative monetary
importance of those factors. The third is to develop a spreadsheet model that
enables operators to quantify the likely impact of those factors on potential
offerings and, thereby, to estimate likely profits and when they can be
expected.
This interactive spreadsheet model – included with the study -- enables
operators to pinpoint the precise monetary gain (or loss) they would likely
experience by introducing specific products, applications, and/or services. It
also enables vendors and the financial community to better understand the
challenges to profitability for operators.
Organization of the Study
The first chapter introduces the concept of AMPU and places AMPU within the
context of profitability. It also outlines the objectives of the study,
summarizes its organization, and discusses its strategic value.
The second chapter discusses subscribers, revenues per subscriber (ARPU), and
four cost factors that may affect AMPU – handset subsidies, infrastructure
costs, network operating costs, and marketing and advertising costs. (Future
issues of this study will expand on these.)
The third chapter discusses the difficulties in defining end-user categories
and the challenges of incorporating them into an AMPU analysis, but it also
shows how doing so can enhance the quality of the analysis.
The fourth chapter describes the model itself, both in text and mathematical
terms. It lists the variables – subscriber numbers, revenue per
subscriber, and the four cost factors– which the model encompasses,
discusses the calculations that the model performs, and provides the equation
for the AMPU output.
The fifth, and final, chapter uses UMTS-based offerings to provide an example
of an AMPU analysis. It provides hypothetical values for each of the factors and
calculates and discusses the AMPU for the hypothetical offerings. It shows how
AMPU shifts from negative to positive as more users come onto the network
(therefore spreading the base for allocating depreciation costs), and as the
cost of handsets falls.
The Strategic Value of the Study
Many operators, vendors, investors, financial analysts and others have
in-house strategic personnel who struggle with these issues. This study will
assist them in, at least, four ways.
First, and most importantly, it can provide an independent validation of the
work that such groups may be pursuing. Second, by consistently asking, "Can
you make money doing this?" it provides a commercial counter-balance to
avid technology advocates – whether on the staff of the operators or of
their vendors. Third, by focusing on non-voice (data) applications, it treats
the services that are most relevant today. Fourth, by requiring quantified
inputs, it facilitates strategic and planning personnel in refining input values
for proprietary models that they may already use.
AMPU (AVERAGE MARGIN PER USER) – NOT ARPU (AVERAGE REVENUE PER
USER): A BETTER METRIC FOR THE WIRELESS INDUSTRY – including the
interactive forecasting model -- is available to order now. | 
View Cart
