1. Mobile money contribution may be small compared to current MNO total revenue but could be important for future revenue growth. Mobile money is expected to reach 10% of total MNO revenue within 10 years and be cash flow positive within 3 years for many operators. 2. Mobile money success is highly dependent on the size of the MNO's existing voice customer base. Having a large pre-existing customer base is important for driving adoption of mobile money services and reaching critical mass. 3. Modeling shows that even if not the first mover, the revenue potential is greatest for the largest MNO in a given market due to their larger existing customer base driving higher adoption of mobile money services.

1. Five business case insights on mobile money
April, 2011

2. Five business case insights on mobile money
How to think about the overall revenue potential?
1. Mobile money contribution may be small compared to current MNO total
revenue but could be important for future revenue growth
2. Mobile money success is highly dependent on the size of the MNO’s voice
customer base
What are the most critical business case drivers?
3. Direct profit from mobile money depends on growth in “electronic-only”
transactions, i.e., more transactions per deposit
4. There are indirect benefits of mobile money to MNOs, but these only
become significant when mobile money reaches scale
How should MNOs think about scale and profitability?
5. To capture long-term profits beyond domestic transfers, mobile money
implementations will need to “leave money on the table” in the short term
Note: This report was informed by CGAP’s Mobile Money Expectations Survey, CGAP M-PESA profitability analysis,
CGAP/Dalberg analysis of mobile money business case, CGAP/BFA/AfricaNext research on external market effects on
mobile money, and other published data/research
2

3. 1. Mobile money contribution may be
small compared to current MNO
overall revenue but could be
important for future revenue growth
3

4. Mobile money is expected to be cash flow positive within 3 years of
launch and represent 10% of total MNO revenue within 10
• Mobile money is expected to be cash flow positive within 3 years
– 43% of respondents to the Mobile Money Expectations Survey believe their
implementations will be cash flow positive in under 3 years
– These expectations match some actual experience: (1) CGAP analysis estimates that
M-PESA Kenya reached positive cash flows in year 3; (2) GSMA MMU estimates that
MTN Uganda will reach positive cash flows in year 2 or 3
• Mobile money is expected to be 10% of overall revenue in 3-5 years
– 80% of respondents to the Mobile Money Expectations Survey expected mobile
money to comprise 10% total MNO revenue within 5 years of launch
– Even at 10% of total revenue, mobile money can comprise up to 100% of current non-
voice revenue
MM as 10% of Non-voice
(2009 revenue) Voice Revenue
Overall Revenue Revenue
Airtel (India) $713,948,148 $6,523,750,959 $713,264,421
Globe (Philippines) $119,401,482 $609,983,230 $614,942,421
MTN (Ghana) $81,755,100 $760,609,292 $56,941708
4

5. M-PESA has exceeded these expectations, but most others are not on
track to reach 10% of total MNO revenue in 3 years
• M-PESA Kenya revenue surpassed USD 94 million by year 3, equal
to 9% of Safaricom’s total annual revenue
• CGAP estimates that other mobile money services are not on track
to meet these direct revenue expectations
– Even MTN Uganda, considered a success in many ways, is not likely to meet those
revenue expectations
Mobile money revenue compared with expectations
(year 3 as % total MNO revenue)
12%
10%
8%
6%
4%
2%
0%
Expected Rev from Mobile M-PESA Rev (Actual) MTN Uganda MM Rev Major African MNO MM
Money* (Projected)† Rev (Projected)‡
*CGAP Mobile Money Expectations Survey, † MMU research 5
estimates including indirect benefits, ‡CGAP research estimates

6. Even with diminished revenue expectations mobile money is a key
source of new revenue growth for MNOs as voice margins fall
• While direct contribution of M-PESA to Safaricom revenue was 9%
in year 3, M-PESA’s contribution to revenue growth exceeded 30%
Core business revenues for MNOs have been This makes mobile money revenue increasingly
declining precipitously across markets… important for MNO growth
Declining MNO ARPU, 2005-2010 40% % contribution of M-PESA to
30 Safaricom revenue & revenue growth*
35%
30%
25
25%
20
20%
USD
15 15%
10%
10
5%
5 0%
2008 2009 2010
0
Africa Americas Asia/Pacific To Incremental Revenue Growth To Total Revenue
*Based on analysis by CGAP/AfricaNext
6

7. 2. Mobile money success is highly
dependent on the existing size of
the MNO’s voice customer base
7

8. Pre-existing MNO market share key to mobile money success
Safaricom Kenya had dominant position in the Pre-existing voice market share important
most concentrated mobile market among all for 4 main reasons:
MNOs implementing mobile money in Africa
1. The vast majority of mobile money
Relative fragmentation in African customers are likely to come from voice base.
mobile money markets (HHI*)
0.6
2. Mobile money by itself has not been shown
0.5 to be a powerful tool for voice customer
growth and acquisition, so it is even more
0.4 important to start mobile money with a large
pre-existing customer base
0.3
0.2
3. Successful payment systems require a
critical mass of adoption in order to succeed.
0.1 Large pre-existing customer base important to
reaching adequate level of adoption
0
Q2 2007 – Q4 2010
4. Dominant position in a market usually
Average (all Africa) Kenya
Ghana Uganda means greater power and control over airtime
Mali Cote d'Ivoire distributors, which can be important to getting
Tanzania Senegal an agent network scaled up effectively
South Africa
* Herfindahl-Hirschman Index: measure of market concentration
8

9. Mobile money launched in environments with critical mass of voice
subscribers
• Overall mobile penetration was at least 20 percent but no more than
60 percent at the time of the first service launch in select pioneering
mobile money markets
• First movers had at least 30 percent voice market share at the time
of launching mobile money
90%
80%
Safaricom – M-PESA
70%
SMART Philippines
60% – Smart Money
Market Share
50%
Orange Money
Cote-d’Ivoire
40% Globe Philippines
– GCash MTN Mobile Money
Cote-d’Ivoire
30% Zain - Sokotele
20%
Zain - ZAP
10%
Yu - YuCash
20% 40% 60% 80%
Market Penetration
First mover
9

10. Modeling exercise showed that revenue potential is greatest for the
largest MNO in a market, even when it is not the first mover
• Because existing voice market determines adoption of mobile
money, even if the second largest MNO moves first, it will likely
make less than the largest MNO launching second
• Largest MNO could secure 60% more in revenue when second
entrants come on its heels (2-7 months)*
Value of mobile money over first 18 months by % total
MNO customers registered
$ NPV
1% 5% 10% 20% 30% 40% 50% 60% 70% 80% 90%
% MM uptake
First mover alone First mover with second entrant
Second mover, T0 Second mover, T18
*Based on CGAP/BFA modeling exercise voice market factor influence on MM success
10

11. MNO market structure determines mobile money success factors and
strategy
CONCENTRATED MARKET: FRAGMENTED MARKET 1: FRAGMENTED MARKET 2:
dominant MNO captures most partnerships are key to success second mover among largest
value as either 1st or 2nd mover but limited revenue must be split MNOs has chance to catch up
Single dominant player No dominant player, 2 market leaders with
Desc.
Desc.
with a number of much Desc. multiple players with similar market shares
smaller MNOs similar market shares
Dominant player could Unlikely that any single Ability of mobile money to
capture significant value player could capture take hold in this market
in direct transfer revenue significant value from and capture significant
and indirect revenue as money transfer business value is uncertain
Strategy
Strategy
Strategy
first or second mover
Success determined by
Smaller players unlikely to which MNOs are able to Among the 2 market
capture significant value form successful leaders, second mover
from MM, even if they partnerships to expand still has a good chance to
partner with other small reach of MM catch up
players implementation
11

12. 3. Direct profit from mobile money
depends on growth in “electronic-
only” transactions, i.e., more
transactions per deposit
12

13. Looking at M-PESA Kenya case, revenue grew 6.6 times faster than
costs between years 1 and 3 causing huge leap in profits*
M-PESA Kenya EBITDA grew from USD - 6.6 times faster revenue growth compared
10 to 30 million between years 1 and 3 to costs in the same time period
$120.0
Revenue
$35.0
$100.0
$30.0
$94.5
$80.0
$25.0
M-PESA Estimated EBITDA
158%
$20.0 $60.0
(millions USD)
$15.0 692%
$40.0
Millions USD
$10.0 $36.6
$20.0
$5.0
$4.6
$- $-
Yr 1 Yr 2 Yr 3 Yr $(16.2)
1 Yr 2 Yr 3
$(5.0) $(20.0)
$(10.0) Breakeven at 5.3 million $(40.4)
customers assuming Yr $(40.0)
150%
$(15.0) 2 avg transaction
numbers & breakdown $(63.2)
$(60.0)
81%
Notes:
$(80.0)
- EBITDA estimates do not include M-PESA Management personnel or G&A
costs Costs
-EBITDA for FY08-10 are based on our low-end cost estimates
*From M-PESA Profitability Analysis by CGAP based on publically
available data
13

14. There are 3 main drivers behind this level of direct profitability growth
for mobile money
1. Growth in overall transactions/customer
Direct profit 2. Change in cost structure away from fixed
drivers marketing costs towards variable costs
3. Growth in ratio of “electronic-only”
transactions to agent-based transactions
Indirect profit
drivers
14

15. Growth in overall transactions per customer is important to mobile
money profitability growth
• Growth in transactions per registered customer comes from growth
in ratio of active customers to inactive customers and growth in
transactions per active customers
High inactive customer rates mean each active Rapid profit growth comes when not only the
customer must generate large amounts of ratio of active customers increases, but each
revenue for mobile money to see profits active customer also transacts more
Operating Costs per Active Customer M-PESA Kenya growth in
Technology Marketing SG&A Customer Service transactions/customer/month
6.0
$16.00
$14.00 5.0
$12.00
4.0
$10.00
$8.00 3.0
$6.00
2.0
$4.00
$2.00 1.0
$0.00 0.0
50% Active 10% Active 1% Active Yr 1 Yr 2 Yr 3
15

16. Mobile money profit increases as cost structure shifts away from fixed
marketing costs towards “revenue-generating” costs
M-PESA Kenya cost structure changes
Variable Agent Set-up Agent Management Customer Service
Technology Customer Acquisition Marketing
Agent Commission
100%
Agent CGAP estimates that marketing &
90% Commission
customer acquisition* costs more than
80% halved between FY 2008 and FY 2010
70% Marketing
Agent as a percentage of total cost
Commission
60%
50%
40% As customers and transactions have
Customer Marketing
Acquisition grown, agent commissions have
30% Customer
Acquisition surpassed all other cost types
20% • agent commissions are variable
Technology Technology
10%
costs directly tied to revenue
generation
0%
Yr 1 Yr 3
16

17. Direct profit growth largely attributable to growth in ratio of “electronic-
only” transactions to agent-based transactions
• Growth in “electronic-only” transactions means customers are
performing more transactions for each deposit at the agent
Change in Tx/Customer/Month • In our estimate, M-PESA Kenya
6.0 “electronic-only” transactions grew
5.0
35% faster than agent transactions
Electronic-
1.9 only value
4.0 transactions
1.4 • “Electronic-only” transactions are
Balance
3.0 inquiry cheaper. In our estimate, M-PESA
2.0 0.7 earns an estimated 18% weighted
Agent
transactions
(cash in/out)
average gross margin on agent-
1.0
based transactions compared with
0.0 almost 100% gross margin on
Yr 1 Yr 2 Yr 3
electronic-only transactions
17

18. 4. There are indirect benefits of
mobile money to MNOs, but these
only become significant when
mobile money reaches scale
18

19. There are 2 main drivers of indirect profit stemming from mobile money
Direct profit
drivers
1. Airtime purchased through mobile money
reducing cost of sales
Indirect profit
drivers
2. Use of mobile money reducing customer
churn rates
19

20. Airtime cost of sales and churn are primary cost drivers for MNOs,
making them the main targets for indirect benefits from MM
Selling airtime direct to consumers through Churn (or the percent of customers an
mobile money can save MNOs over 20% in MNO loses every month) is rising, costing
cost of sales in scratchcards MNOs millions in lost revenue
Airtime sales thru Airtime sales thru
scratch-cards mobile money Global MNO churn rate, 2005-2010
3.40%
MNO MNO
3.20%
Card printing Airtime sold direct 46% increase in
to consumer
3.00%
5 years
Airtime dealer 2.80%
Top dealers buy at up to
20% discount 2.60%
2.40%
Small retail shop
2.20%
Customer Customer 2.00%
Probably more important as indirect impacts of Churn reduction could have very significant
mobile money from reductions in airtime cost benefits for MNOs, but very difficult to attribute
of sales are easily measured causal effects of mobile money on churn
20

21. Indirect benefits can represent up to 50% of overall benefits to the
MNO from mobile money
Estimated Direct and Indirect Profits Estimated Direct and Indirect Profits
from MTN Uganda (from GSMA) from another major African mobile
money implementation
7% 7%
31%
62%
Retained ARPU from churn reduction
Airtime distribution savings
Direct revenue
Uplift in voice/data usage
21

22. …but indirect benefits are only significant when mobile money reaches
a large proportion of total MNO voice customers
• Comparing averages suggests that mobile money reduces churn…
– Analysis of a major Africa mobile money service shows subscribers churn 60% less
than general subscribers (2.18% versus 5.71% monthly churn)
– Safaricom churn has remained flat since M-PESA launch, but as competition has
increased in the same period M-PESA may have prevented increase in churn
• …but more than 20% of the voice customers may need to be using mobile money to
see a meaningful impact on revenue
– Registered users in most markets are under 10% of subscriber base unlike 77% for
Safaricom in Kenya and 17% for MTN Uganda
Cost savings from reduced churn increases linearly
with % of MNO customers utilizing MM
4.0%
Cost savings as % total MNO rev.
Almost 60% of total
3.5%
MNO customers
3.0%
must be registered
2.5%
for MM before
2.0%
indirect benefits
1.5%
from churn
1.0%
reduction surpass
0.5%
2% of MNO
0.0%
revenue
% total MNO customers registered for mobile money
22

23. …but indirect benefits are only significant when mobile money reaches
a large proportion of total MNO customers
• Cost savings from airtime sales through mobile money are highly dependent on the
scale of the mobile money business
– 25% of total MNO airtime must be converted from scratchcard sales to mobile money
sales before cost savings surpass 2% of MNO revenue
Airtime sales cost savings increase linearly with % of
airtime sold through MM
10.0%
Cost savings as % total MNO rev.
9.0%
8.0%
7.0%
6.0%
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
% total airtime sold through mobile money
– 20% of total Safaricom airtime was sold through M-PESA Kenya in its 3rd year of
operation, but other MNOs have not matched this growth rate
23

24. 5. To capture long-term profits
beyond domestic transfers, mobile
money implementations will need to
“leave money on the table” in the
short term
24

25. To capture long-term profits beyond domestic transfers, mobile money
implementations will need to “leave money on the table” in the short term
None of these pricing strategies above preclude what MNOs need to do operationally to create a
successful mobile money business, i.e., agent network management, marketing etc.

26. Volume of everyday small business or merchant transactions are likely
to be at least 10x the size of domestic transfers
1. As is typical of most developing countries, in
Kenya, medium and small scale enterprises make
up 90% of enterprises and account for over 20%
of GDP.
Everyday small
business or merchant
transactions 2. In India consumer to business payments are
far more than 10x the volume of domestic
transfers
~ at least 10x
in volume
Domestic
transfers
26

27. Current mobile money pricing aims to maximize profits from domestic
transfers, but hinders growth into the merchant payments market
For a large domestic transfer MM pricing is
low, but for smaller payment sizes the cost At a basic level, pricing that makes
transacting at lower transaction sizes more
is prohibitive
expensive will make it harder for low-income
customers who in many markets are likely to
Estimated total cost of MM transfer
be the larger share of customers
as percent of transfer volume
12.0%
This pricing will make it harder to scale in
10.0% 10.9% domestic transfer but will also make it harder
Lower value to drive growth in small business or
transactions merchant payments
8.0%
5x expensive
• Avg. transactions between merchants and
6.0% consumers are smaller than domestic
transfers, making many prohibitively
4.0% expensive with current pricing schemes
• Domestic transfers typically happen across
2.0% 2.5% large distances where the cost of moving
cash is high. Merchant payments happen
0.0% face to face where the cost of cash is less
USD 10 Transfer USD 61 Transfer tangible, reducing willingness to pay
From CGAP Branchless Banking Pricing Survey
27

28. This short-term focus may hinder customer adoption from ever
reach “tipping point” scale to capture small merchant payments
Merchant payments require a “tipping Pricing to maximize P2P profit may prevent
point” scale of adoption to be valuable: a mobile money service from ever reaching
this level of scale
• Merchants need to see lots of consumers
using a payment system and consumers
need to see lots of merchants Price Illustrative P2P Demand Curve
Consumers
max profit price
Anchor product
Tipping point
scale
Scale for anchor Customers
Merchants product max profit
28

29. Similarly, the volume of payments between consumers & institutions
(gov’t & corporations) dwarf those from consumer to consumer
In India, payment flows between consumers and In Kenya M-PESA, while exhibiting massive
government are 40 times the size of total payment growth in the P2P space, has likely captured less
flows from consumer to consumer than 7% of the total Kenya payments market
India payments landscape
60000 Kenya GDP times
estimated velocity
of money (1.63)
50000
40000
Millions USD
30000
20000
10000
0
Total Kenyan payments Total M-PESA transfers
market (2009) (FY2010)
29

30. MM implementations will only be able to capture institutional
payments once they reach a “tipping point” scale of consumers
• M-PESA Kenya corporate accounts for disbursing and receiving payments to
consumers grew exponentially once M-PESA reached 7 million customers
M-PESA corporate account growth related to end
customer acquisitions
12,000,000 350
300
10,000,000
M-PESA registered customers
M-PESA corporate account
250
8,000,000
200
6,000,000
150
4,000,000
100
2,000,000
50
0 0
Oct-06 Apr-07 Nov-07 Jun-08 Dec-08 Jul-09 Jan-10 Aug-10
Subscribers Corporate Accts
30

31. Merchant & institutional payments increase exponentially with
customer growth while P2P transfer revenue & indirect benefits
increase only linearly with customers
Growth in revenue sources related to customer growth
30.0 9.0
Revenue Breakdown (millions USD)
8.0
25.0 If revenue at scale from
Millions of customers
Consumer adoption as
affected by price per 7.0
indirect benefits and
20.0 domestic transfer 6.0 capturing
5.0 merchant/institutional
15.0
4.0 payments dwarfs the
10.0 3.0 domestic transfer
2.0
opportunity, maximizing
5.0 customer adoption by
1.0
offering “anchor products”
- -
for free may result in
$0.40 $0.20 $0.00
greater overall returns to
Price per transfer (USD) MNOs
P2P transfer revenue Indirect benefits (churn, airtime)
Institutional payments (freemium) Small merchant payments
Price/transfer $0.40 $0.20 $0.00
Active Customers 2,000,000 4,000,000 8,000,000
31

32. Significant indirect benefits of mobile money are also only
realized with large scale customer adoption
• Indirect benefits can represent up to 30% to 50% of overall benefits
to the MNO…
1. Airtime purchased through mobile
money reducing cost of sales
Indirect profit
drivers
2. Use of mobile money reducing
customer churn rates
• …but indirect benefits are only significant when mobile money
reaches a large proportion of total MNO customers
– More than 20% of the voice customers need to be using mobile money to see a
meaningful impact on churn
– 25% of total MNO airtime must be converted from scratchcard sales to mobile
money sales before cost savings surpass 2% of MNO revenue
32

33. There are additional revenue opportunities beyond transaction
fees and indirect benefits that may open up at scale
• Mobile money and its usage data could be utilized as a targeted advertisement
delivery channel to bring in additional revenue
The mobile advertising market in Africa is already Mobile advertising opportunity may represent over
among the largest in the world 33% of total direct EBITDA from mobile money
Total available mobile ad impressions Ad revenue opportunity compared with
by region (InMobi) other M-PESA benefits
35.0
India
30.0
25.0
North America
Millions USD
20.0
Africa 15.0
10.0
Europe
5.0
0 1000 2000 3000 4000 5000 6000
0.0
Millions of impressions per month M-PESA direct Airtime cost of Ad Revenue
EBITDA sales savings
African consumer’s acceptance of mobile
advertising is the highest in the world Note: Assuming 65 million M-PESA transactions per month each
represent an opportunity for a mobile ad impression
33

34. Advancing financial access for the world’s poor
www.cgap.org
www.microfinancegateway.org