Safaricom is the poster child of developing world operators, within eight years of Vodafone’s buy-in and what is now a textbook example of grassroots innovation (the M-PESA mobile money service) under its belt
Safaricom: Micro finance, macro profits
TextStart
By Joyce Fan
Having achieved 600-fold customer growth within eight years of Vodafone’s buy-in and what is now a textbook example of grassroots innovation (the M-PESA mobile money service) under its belt, Safaricom is the poster child of developing world operators. However, despite its enviable position, a price war in the voice market and new entrants into the mobile money and mobile broadband sectors are forcing the Kenyan operator to stay hungry. CEO Robert Collymore and CTO Thibault Rerolle recently sat down with WinWin to discuss what their firm has in mind for an encore.
Less reliance on voice WinWin: What competitive advantages does Safaricom have?
Collymore: We have the biggest and the best network in the country, which gives us a significant advantage. The network has the broadest geographic coverage. We made it the best by upgrading 50% of the base stations to 3G, which provides the speed of 21Mbps, and we are about to introduce 42Mbps. The other advantage that we have is M-PESA, of course. So, we have a very effective mobile money transfer system which is used by a large percentage of the population (96% of banked & 75% of unbanked in Kenya in 2011, according to Slate)
WinWin: How does Safaricom effectively increase ARPU and retain customers in the midst of a price war?
Collymore: ARPU decline comes about from two things. First is a reduction in price. We are addressing it really by making sure that our pricing accurately reflects the products that we are delivering. In a number of places, the prices of our products are more expensive than our competitors, but at the same time, the customers believe that the product gives them best value for money.
Second, as you increase the penetration, you start to increase in the lower-income customers, which will decline your ARPU. Comparing with our competitors, we are not continuing to drive further the penetration into the market, and that is sustaining our overall ARPU.
The most important thing is to make sure the customers are using more than one product. A typical customer uses voice and SMS. We also introduce them to M-PESA. Around 15 million out of nearly 19 million of our customers use M-PESA. Other than that, we are encouraging them to use data and finding some interesting applications for them. The more products you give to customers, the more they are likely to stay with you, if you are giving them overall good value for money. We are substituting the decline in the voice ARPU with people spending more on M-PESA and data.
Fully exploring the possibilities of M-PESA M-PESA (an acronym that combines mobile with “money” in Swahili) has been called a lifesaving service because it fulfills the needs of money transfer for millions of the unbanked. After five years of operation and development, M-PESA has moved far beyond its origins. Lifestyle-changing features are now in use and banks have taken part. Collymore has commented, “M-PESA is bigger than cash – anything you can do with cash you can do with M-PESA. So, the possibilities are endless.”
WinWin: The fifth anniversary of M-PESA was celebrated this March. Can you tell us what new features/applications have been added in the last few years?
Collymore: Well, we are getting more people to accept M-PESA. First is to broaden the range of uses for M-PESA. For example, individuals can pay school fees, utility bills, and shop in the supermarket through M-PESA. We extend the service to the corporate business level where companies can pay salaries by M-PESA. The second thing is that we have reduced the lower limit people can transfer with M-PESA. It was reduced from 50 Kenyan shillings (KSh) to the present level of 10 shillings, which is equivalent to about 12 cents in the U.S. We’ve also reduced the fees for those lower-level transactions, which makes it more accessible to more people. I think that we are number one in the world, in the sense that more than one in every two mobile money transfers is done on Safaricom’s M-PESA system.
WinWin: What is the next step then? How are you going to make it a fully-fledged banking service?
Collymore: Many people ask this question. Features that we are currently exploring include micro-insurance, micro-financing, micro-lending, and micro-savings.
Cooperating with Equity Bank (an East African financial group), we have got a thing called M-KESHO, which means mobile saving in Swahili. With M-KESHO, the customer can earn interest, borrow or lend money. That is a partnership that we have with one particular bank, but we are exploring how we can broaden that.
Kenya has the privilege of having around 42 different banks; 42 for a small country serving only 40 million people. We have different cooperation models with different banks. We are integrated to a lot of banks.
You can move money between the bank account and the M-PESA account seamlessly. You can currently take out your money from ATMs. I think you can take out money from about 300 ATMs across the country, which is probably most of the ATMs, actually. With companies like Equity, we have M-KESHO. We also have some banks who are expressing their loans; so if you borrow money from the bank, they will deliver the money on M-PESA. So, there’s a range of cooperation models that we have with banks. We have moved on from the days when the banks were critical of what we are doing. And I think that we have all understood that these things are complimentary.
WinWin: This gives people more access, leading to increased ARPU and user base, but do you have any challenges related to technology and/or operations?
Collymore: The technological challenges are many because the platforms weren’t designed to handle 15 million distinct users or more than 2 million transactions a day. We are handling about KSh2 billion in cash per day, and that’s a constant challenge.
Rerolle: Yes, as mentioned by Mr. Collymore, the capacity of the platform overall was not designed for this scale. So what’s happening then is that we are constantly increasing the capacity. But now we have come to a point where we not only need to increase the capacity but also change the architecture in order to evolve this platform to the next generation.
The issue lies not only in terms of volume but also volume of transaction. Currently, our run rate is around 100 transactions per second and we are peaking at 220 to 250 transactions per second. We see the growth in the coming months and years and this is a reason that we absolutely need to change the architecture. We also need to improve the relay ability of the platform and its accessibility; that also needs to be more robust in the future.
WinWin: Safaricom has announced a platform migration plan for M-PESA. How will you do that?
Rerolle: The migration, as I mentioned, is an architecture change. It’s also a geographical change. Currently it’s a Vodafone-hosted platform and service and I would say that the objective of this migration is for it to be hosted locally, in Kenya.
We will execute this over a period of time that we plan to be around a year and a half. Obviously we cannot tie this rollout with the new service evolution because the cycle of new services in the market is much quicker than this timeframe. Our roadmap of new services is not totally dependent on this migration. So, we are rolling out new services and new functionalities into the existing platform as much as we can, while we are also preparing for the migration.
Promoting mobile data usage WinWin: You mentioned that Safaricom runs the biggest 3G network in Kenya. How has your 3G service been received by the local people?
Collymore: We were the first to launch 3G in Kenya. I think it has been welcomed by local people. More than 90% of data and Internet access are primarily through the mobile phone in Kenya. So, 3G has been important.
The biggest determining factor has been the availability of 3G handsets on the market. So, although more than 50% of our network is 3G, we still only have just over one million 3G handsets, and most of the mobile accesses on laptops require a 3G modem. We have actually been very successful in our partnership with Huawei over the past year through the introduction of the Ideos phone. This has been one of the best-selling phones in the market. Through that kind of partnership, we will drive the adoption of smartphones.
WinWin: Safaricom’s website helps users choose data packages based on their using habits for different data applications such as video, email, SNS, etc. Tell us more about your data service offerings.
Collymore: What I would say is that we are still learning from international benchmarks about what the pricing models should look like. But we have attempted to provide a range of options whenever possible – unlimited-but-throttled bundles, daily bundles, weekly bundles, and monthly bundles.
We have bundles based on size, but for people who don’t understand megabytes, we price the service based on minutes – KSh3 per minute. We thought that was interesting but simple because many of our customers understand a minute but don’t understand megabyte. They were criticizing us for allegedly stealing their money because they didn’t understand that if they downloaded a big file in two minutes, and because the speed has gotten better, they were quickly exhausting their bundles. So, we have offered that as an option.
WinWin: Providing diverse data offerings in a timely manner requires a robust billing system. How has the new CBS system helped in this regard?
Rerolle: Yes, of course, the billing system is absolutely instrumental to this functionality. We have implemented, in the past year, Huawei’s Convergent Billing System. It is a long journey. We initially migrated prepaid voice, then postpaid voice, then out-of-bundle data; now we are migrating in-bundle data. And then once all data functionality is integrated, we will be able to benefit from all the functionalities that we have drafted into this migration path.
What we want is to be able to timely deliver what is inside the bundle, enable timely top-up on top of the bundle, and give adequate visibility or clarity to the customer about what is his bundle, what is his consumption. This is absolutely the key to the customer experience – to understand and to have a simple way to see or to benefit from this bundle.
I don’t know how the previous CBS performed. But for sure, the robustness and the response to the overall network system of the new system is quite fine in terms of integration. And of course, the simplicity that it brings by converging all voice & data, prepaid & postpaid, is an essential asset that we benefit from now.
Info box
M-PESA milestones: 2012
- Over 15 million active M-PESA users, out of 19 million total Safaricom customers.
2011
- Alliance with Western Union: Customers can receive international money transfers from 45 countries and territories.
- M-PESA International Prepaid Visa Card: Customers can transfer money from their M-PESA accounts into an international prepaid Visa card.
- School fee payment: Safaricom’s Lipa Karo service allowed learning institutions to receive school fees via M-PESA.
2010
- M-Ticketing: Customers could now book and pay for tickets to concerts, galas, and even sporting events with M-PESA.
- Supermarket service: In a milestone that re-defined the Kenyan shopping experience, customers could pay for groceries using M-PESA.
- M-KESHO Bank Account at Equity: Safaricom and Equity Bank partnered to offer a new banking system that allows customers to manage their bank accounts via mobile.
2007 – 2009
- Paying for fresh water: Safaricom partnered with Grundfos LIFELINK (a pump manufacturer that aides rural access to fresh water) to allow these residents to pay for their water through M-PESA.
- Small and Micro Enterprise Program (SMEP): SMEP allowed its customers to make their monthly loan repayments and savings contributions through M-PESA.
- ATM withdrawal without ATM cards.
- Acquired 19,671 users two weeks after launch, Mar 31st, 2007.
|
