M-PESA's Mobile Money Revolution: Key Lessons for Banks

 Simon Taylor photo
Simon Taylor Co-founder 11:FS & CPO 11:FS Foundry
5min read

Lesley-Anne Vaughan, Director of MiLA Consulting and Co-Creator of M-PESA, Vodafone’s famed mobile money transfer service, has been dedicated to mobile money and mobile financial services in emerging markets since 2005. Here, she shares how M-PESA transformed from a microfinance product to a revolutionary service that’s changed Kenya’s infrastructure, increased financial inclusion, and earned the majority of the market for mobile money transfers.

The following Q&A comes from Simon Taylor’s interview with Lesley-Anne in episode 203 of FinTech Insider Interviews.

Simon: Well it’s great to have you here. I think the first question I have to ask is, what did you want to be as a kid?

Lesley-Anne: I was a kid in the 80s in Northern Ireland – not an engineer! I played with my dolls and everybody was a nurse or a doctor or a teacher, and probably the first interaction I had with an engineer was when I was about 17. My math teacher said go to this STEM event in Scotland. It was a week-long event for girls and I think that’s when the excitement grew around actually what an engineer is – it’s not mechanic! From there really that’s where my engineering focus came but as a kid, no, I was a proper girly-girl.

Simon: So how would you describe what an engineer is then, because it’s not a mechanic but I think to the layperson it can sound a lot like a mechanic?

Lesley-Anne: I guess in the day-to-day job I’ve been doing the past 20 years, it’s about making an idea real. It’s taking a ‘what if?’ and going, ‘can we actually do it?’ It’s playing with things. It’s ‘how would we?’ and the reason we have all the gadgets and gizmos we have in the world today are because engineers started to play with things and make things a reality.

My first job was kind of generic. It was full of physicists and mathematicians and engineers and they were hardware engineers and software engineers and there was cross-functional teams doing all kinds of cool stuff. I think it’s the cool stuff, that’s what I think engineers do.

The reason we have all the gadgets and gizmos we have in the world today are because engineers started to play with things and make things a reality.

Simon: There’s something about making something work. It’s a building some Lego can or it’s just getting a piece of software to say your name.

Lesley Anne: You have to play with it and try and try and try! An engineer doesn’t leave a problem. It takes a problem and you keep going until you figure out what to do next.

Simon: I love that. So talk to me a little bit about being an engineer at M-PESA. It might be worth explaining what M-PESA is.

Lesley Anne: So if you took Monzo, a mobile bank, and you said…you want to have a debit card, a current account, but you want to operate it via mobile phone, that’s effectively M-PESA in Kenya. It’s a way to transfer money from one person to another. We use a network, like the post office, [of] 70,000 stores in Kenya to put money in and out of the system, but effectively it’s a way to move money between accounts and use this cash agent network to do debits and credits.

The key is you don’t need a bank account, you only need to register for M-PESA and the KYC required for M-PESA is much lower than for a bank account, and the business model behind it is pay-as-you-go, back in 2005 it was anyway. The key for that was when you don’t use it, you don’t pay for it. I think that was the key differentiator to a bank account. People were a bit scared of bank accounts…but they still had to save money.

There’s a lot of migrant workers in Kenya, they work in Nairobi, they travel home with the money they send money with bus drivers. Maybe half the money would get home, so you end up with a situation where it’s risky, it’s expensive, it takes a long time to send money, and now you’ve got this way we can do it instantly – ‘fast. easy, secure, safe’ was the value proposition.

Simon: That’s very different to having cash, very different to being robbed and having your money stolen from you and it’s also very different to opening a bank account. I guess in 2005 you wouldn’t have a smartphone, especially not in Kenya, so it would have been a very different experience?

Lesley Anne: I didn’t have a smartphone in 2005. None of us did. There were no APIs, there were no smartphones. I was using a Nokia in 2005, but we created a digital service and we did it using very basic SMS-led services and they were encrypted end to end, triple DES, we had security experts who were doing proper bank-grade encryption stuff around the service. It was very secure in that respect but it wasn’t hard for the end users to use it – it was a very, very simple user experience and that was key.

We actually pivoted. In 2007 we launched M-PESA as a live service but in 2005 the original product was microfinance. We were solving a pain point of how do we get microfinance to be more effective, more efficient, more cheaply administered and we worked with the microfinance Institute and Safaricom on this.

It was all about repaying loans and disbursing loans and dealing with the group mentality of having to leave your business and do all kinds of things. The problem is far more complicated. We were actually giving people their phone in the meetings for the first time, we were training them how to use a Nokia and we saw then that the service is actually quite complex; what we designed had modelled the real life of field officers and moving money this, there and everywhere, but actually we trained people to use these phones and we put a couple of services on there that turned into M-PESA today. The P2P service and buy airtime went into the pilot alongside the microfinance portions of the business.

I didn't have a smartphone in 2005. None of us did. There were no APIs, there were no smartphones. I was using a Nokia in 2005, but we created a digital service and we did it using very basic SMS-led services.

Simon: Well that’s really quite powerful isn’t it, that you tested and learned and adjusted very quickly and then people took on and adopted something where previously they were using cash, migrating, and the cash was being stolen from them – there was a high percentage of crime at the time. Now they’ve got a real solution that – they can go to a store they can buy airtime, and they have to buy airtime anyway because they’re buying for their mobile phone to call each other, and they can send that airtime pretty much to each other as if it was the same as cash.

Lesley-Anne: It’s very like airtime but actually it wasn’t airtime. So there’s a completely separate engine that was able to move airtime around the system but our system was completely designed ground up, full stack and it was like trading one-to-one with money currency, completely separate to the airtime billing system. That was important because now we had a way to create a business model, whereas the airtime engine was happening completely empowered on the same company. That was being done for free. There was no business processes around it to make it secure, make it this, make it that, whereas now we had this effectively full stack banking system that was designed from the ground up that has nothing to do with the airtime but you could use the money in there to buy airtime.

Simon: Talk to me a little bit about that business case.

Lesley-Anne: It was pay-as-you-go and you didn’t pay to deposit – that’s all about viral growth. You get money into the system and it’s completely free. You pay to send, you pay to withdraw, and when you take the three transactions together, you see that you can make money but you assume that three transactions that happen in the P2P model. It’s grown since then we got Bill payments happening massively and there’s lots of different ways that you can use that.

In fact there’s all kinds of IoT (Internet of Things) stuff coming out now. There’s ways to buy your solar energy pay-as-you-go, all kinds of things happening and it’s a massive growth factor. A massive change has happened in that bill payment space and merchant payments are just starting to happen now, but there’s a real problem there with merchant payments and the business model in the sense that merchants aren’t used to paying like they are in the U.K., where merchants pay to accept card transactions. They are not used to that so it’s quite hard to get some acceptance and that’s still a model that isn’t quite settled yet in terms of where we’re going in the industry.

How do telcos make money? They can’t make money like a bank, they can’t offer interest, and they can’t loan the money out. The money must sit in the bank account and the interest is treated in a very special way because the regulations in play around the fact that we use the money transfer license, so it’s a lot of money in the bank it’s making a lot of interest.

The first places where we’ve seen some change in this is Tanzania. They’ve actually gone for a profit-sharing agreement and the regulators have agreed that they will send the money back to people who have deposited money. That might create some change, might create some different business models going on but really pay-as-you-go is heavily the mode in East Africa and I think that’s worked because it’s cheaper than the alternative. Possibly actually we need some business model innovation to see it working elsewhere. There’s no way M-PESA would work in the UK. We don’t pay for banking. You couldn’t put the same kind of models into different countries. And I think that’s quite important that you study your own markets and see what’s going on.

How do telcos make money? They can't make money like a bank, they can't offer interest, and they can't loan the money out. The money must sit in the bank account and the interest is treated in a very special way because of the regulations in play.

Simon: Well absolutely I think there’s something somebody once said that M-Pesa is a bit like the Galapagos. It’s this wonderful thing that has all these wonderful features and properties that you don’t see anywhere else in the world because of the unique things in that market. I read that M-PESA / Safaricom so far has 80% of the market as well?

Lesley-Anne: That helps a lot right?

Simon: Yeah really helps when you’ve got a monopoly of users.

Lesley Anne: You still need to do an awful lot of marketing and awareness campaigns and the like, and the big feature of M-PESA was viral growth. The regulators allowed us to be able to send money to anybody any phone number, without a registration, which meant that you as an early adopter can get your mum on the system. It costs a little bit more, so it’s actually an encouragement – you can help us get your mum on the system properly, you’ll pay less in the overall transaction, and that was really important in Kenya. We didn’t have lots of money to spend on marketing for Safaricom, we did do some but that viral growth piece was amazing.

Simon: So you have to go for a viral growth where you just send somebody money and if they’ve got a mobile phone they can pretty much use it straight away. I think the simplicity of that appears to be its brilliance.

Lesley-Anne: It was designed, though. We really thought hard about the onboarding process. You had to leave the agent’s store working and therefore we pushed quite hard on the KYC and when our AML officer came in and went ‘you can’t, you can’t, are you sure, how can we?’ I remember sitting in a room in Cambridge with our AML officer and our our product manager, Susie, and me and some of the tech guys going, ‘you have to photocopy the ID.’ It’s a rural shop in the middle of Kenya, they don’t have a photocopier! What can we do? OK let’s look closer at what FATF actually says. What can we do? Oh we can do quite a lot at less than a thousand euros a year! Oh, okay, let’s put in the concept to KYC. The fact that you can do quite a bit. As a poor person in Kenya, 1000 euros is quite a lot of money!

Simon: I think it’s worth just deconstructing that a little bit because there’s a lot of value in that for our listeners who might be facing challenges working in a bank or working in a FinTech trying to get something working, because KYC – “Know Your Customer” – being ‘I want to get a new financial product and for the first time I’m going to register for it.’ Traditionally that means you walk into a bank branch and you take your photo ID and you take a payment bill. Now what you have is in Kenya or in some rural area is somebody trying to walk into a small shop where they don’t have a photocopier. Now you’ve had to invent something that you guys called “tiered KYC” because of the FATF regulation. I think is FATF 16 which basically says ‘under a thousand euros or under a thousand dollars then certain levels of KYC are required, but over that, more KYC is required. Was that your biggest challenge or did you have other challenges that really came along and you thought ‘oh no we’re not going to get past this one’? and then how did you solve that?

Lesley-Anne: There’s all kinds of challenges, right? A big one, and anyone who’s worked in emerging markets with agents will definitely know it, is agents are the cornerstone of mobile money….A big part of that is, does the agent have any e-money to sell to you and does he have enough cash to take your e-money?…Going into a region, selling money virally into a region where there’s no agents that are willing to serve, they’re not well trained or they’re not this not that, it hits your brand quite hard, it hits the customer trust and actually that agent network needs to be managed really really well – it needs to be trained. It’s an analog problem. It’s not one that you can solve but it’s really important.

There’s ways you can help your agents a lot, though. Do they have access to data that helps them know where they can replenish their float? You could send in a motorbike with cash services. You could route plan if you can see the patterns in your data and all sorts of things. That’s the kind of clever stuff that’s coming out now. It’s not sexy front-end stuff that the customer sees, but there’s so much under the water that has to happen to make this thing really work well. That’s where I think some of the more modern kind of API-centric fintech can add a lot of value in our spaces, help and use that data, predict flows, get cash moving around the country, tell people where to go and what do. I can see some of that starting to happen.

Simon: What what does an API mean to you first and foremost and then how do you use them in M-PESA?

Lesley-Anne: OK. So in 2005 no APIs. Closed-silo system, but the first thing we did in 2007, probably our first integration happened…we offered cardless ATM transactions so you could go to an ATM and use a digital agent effectively to withdraw money without a card.

Our next integration was with Western Union to be able to send money from the U.K. into Kenya, so we needed an API for them to actually trigger that. Our bill payment solution is useless without APIs. If you can land the money in the bill payer’s account but they don’t know it, well they’re not going to update their systems. So when you’re talking to this equivalent of Sky, DSTV, and your Sky TV to turn back on, you need to get them the payment information. It’s standard nuts and bolts payments gateway stuff.

It’s all happened kind of in parallel, alternative payment infrastructure separately to the banking space, the card space but it’s all there is all happening in this space as well. So we’ve been using APIs for a long time.

Simon: What are the key lessons around APIs that you’ve learned? What should banks take from that what should they learn?

Lesley-Anne: It’s listening to your developer community.

The customers are effectively your developers, especially if we’re going to go for a ‘developer pays’ model…Develop as part of your sales process. If a developer likes your service they’re going to encourage their management teams to buy the one that they want.

I think the best thing you can do is have somebody who is technical be part of management, someone who can talk developer language to help do the part of management of design and what your API package should be.

Simon: What’s been the impact of M-PESA? Has it really aided financial inclusion, has it helped in some way?

Lesley-Anne: There was a paper brought out, I think in November or December last year, that explicitly said there is a direct correlation, that we’re actually improving a certain percentage of our families’ lives in Kenya with M-PESA, which is, just like, wow! Really quite amazing…It’s changed the infrastructure and kind of it’s changed the fabric of Kenya, and it’s got a verb to “M-PESA” someone.

It’s not even just about the saving – it’s about making money. An awful lot of people are traders…it changes the way that you can interact with the world if you can trade electronically. The distance I think that’s made the real difference, you can pay somebody remotely. You can do it with a bank account but it’s not the same – it’s hard! The bank accounts in Kenya are changing a lot. They are innovating massively because they now have this massive competitor, but back 10 years ago there was just no easy way to do it, you would actually use Western Union probably before you would use a bank to send money out-of-country. The bank accounts in Kenya are changing a lot.