Page 104 - Performance Leadership

Chapter 6 One Version of the Truth • 93

more these allocations would be needed to catch all facets of revenue.
ARPU then becomes a complete black box.
Now we come to the one context of the truth. It starts with the
assumption that there are different definitions for a reason. In the case
of E-plus, because there are different revenue streams and different user
types, it makes sense to list the various forms of ARPU as E-plus does
in some of its reports as it builds up to the overall ARPU that is reported
to the parent company. Let’s try to take it to the next level. At the high-
est level, there are four different revenue components: fees (including
various roaming fees), discounts, incoming revenues, and the correc-
tions. There are also three different user types: active users, total users,
and non-E-plus users. The full context unravels when we plot the rev-
enue items and user types in a matrix (see Table 6.5).
The ARPU matrix doesn’t mean there is no ARPU performance indi-
cator anymore; it is still an important indicator. But now we have a sin-
gle context of the truth, and we can analyze the various components
for deeper insight. The higher the percentage of total revenue and
ARPU for fees and incoming revenues from active users, the better it
is; this is revenue you can influence with customers you engage with.
This is called controlled revenue. The higher the roaming revenues
from non-E-plus customers and the higher the incoming revenues, the
more you depend on others, uncontrolled revenue. Although uncon-
trolled revenue is part of the business model, that revenue needs to be

Table 6.5

E-plus User Matrix
Active Users Total Non-E-plus
(in millions) Customers (in Users (in
millions) millions)
Fees 1,900 +60 +250
Discounts –170 –90 N/A
Incoming Revenue 480 NA +30
Corrections 20 +15 +5
Total 2,230 2,215 2,500
AIPU = 1,900 million divided by the number of active users
Clean ARPU = 2,215 million divided by the total number of users
ARPU including allocations = 2,500 million divided by the total number of users