M-Pesa is, arguably, the most developed mobile money transaction platform in the world. It’s adoption, and impact,  in Kenya has been nothing short of phenomenal. Outside Kenya, it has been implemented in a number of countries to some level of success.  There’s even a school of thought that holds that M-Pesa is the ideal product when it comes to mobile money.

Still, big part of customers consider it an “at least we have this” product. Its innovative basis has not been able to extend to pre-empting, or evolving with,  user preferences and technology trends. It’s own CEO admits to that. Bob Collymore appeared to scoop my then two weeks old research into this piece when he called M-Pesa a clumsy product that should be innovative and collaborative or risk dying .

The current position of M-Pesa means infinite possibilities for anyone working to re-imagine mobile money. A quick survey points  to general pain points, that can also be interpreted as potential ways of eclipsing it.


M-Pesa is a payments platform. But the payments experience is horrible.  In payments , two steps is one step too many. With M-pesa, it takes up to five steps, or more for iPhone users. Several people say the only reason they use M-Pesa payments is because of lack of an equally ubiquitous alternative.

This experience is exacerbated by the cost of transacting on this platform.  Few people bother to find out the cost of transacting on phones beyond the cost of transfer and withdrawal. However, this has changed and is expected to improve further as banks and telcos move to comply with the directive from the Competition Authority of Kenya requiring them to disclose to customers all fees incurred during mobile money transactions.

Shift in User Interface for Mobile Money Applications

The user interface technology for accessing mobile money won’t be the  SIM toolkit for long, with significant implications on M-Pesa for two reasons.


Everything that was done, or is being done, on the SIM toolkit has moved, or is moving, to the web. Even the SIM toolkit itself is morphing into a software based technology. The  default inclinations of users are becoming more web oriented.  Millennials and Generation Y’ers prefer conducting most of their activities within web based apps, leading to a greater push out of the SIM toolkit.

It is a matter of time before mobile money moves fully to over the web. It already is in some cases. The likes of Square have peer to peer mobile money apps. Yet, there is no standalone transactional app for M-Pesa.

2.Embedded SIMS

It eliminates the need for a connector, basically the telco/MVNO, a role that is critical in Safaricom’s continued grip on the market. As it is structured now, M-Pesa is hosted on the Safaricom SIM toolkit. If you need to transact anything on M-Pesa, at enterprise level, Safaricom has to give space on their SIM toolkit.

With a pre-embedded chip, the SIM will belong to the customer who will choose how to connect to what MVNO, diminishing the position of Safaricom as gatekeeper.

GSMA has been discussing the possibility of a software based SIM toolkit since 2010. Apple has been at war with the toolkit for sometime now and released gadgets with an embedded chip before. Samsung and Apple’s talks with GSMA to embed a standardized  e-SIM have been, for some time now,  described as advanced. Basically, the e-SIM is here with us, and we are out of the SIM toolkit, to the web.

M-Pesa API (Is Not Self Service)

Safaricom has been cagey with the API, at first taking a while to release it, then proceeded to release it in bits without documentation. Those who have  gotten around it say the functionality is good. However, a lot of developers say it takes significant effort to work with it. It is, basically granting the much awaited access but in a form that discourages its use. There are fewer ways to make interested parties less interested.

This is but a case of lost business and a chance for a product to evolve. It is also a limiting of possibilities in an environment where an easy to access and  use mobile payments API is highly applicable due to the pervasiveness of mobile money.


It is possible future MVNOs wouldn’t need masts. At the GSMA conference held in Dar es Salaam earlier this year, people were already discussing a scenario where masts would be replaced by, possibly, fibre. There have been tests conducted on that, and it is feasible in ten years.

In such a scenario, entities with significant muscle can set up their own network. Much like it became easier to set TV stations after the migration from the analogue to digital platforms. Mobile carriers, at least as we know them, will be redundant. M-Pesa, as we know it couldn’t retain its competitive edge in such a scenario as Safaricom won’t be the controlling carrier.

Digital Currency, & Cryptocurrency

The challenge with value has always been its transfer. The storage of value has been addressed by banks for a long time. M-Pesa started as a way of transferring value before it moved to incorporate its storage. Its appeal is the less risk it carries when transferring value as it does this in digital  form. Hower, its other overriding advantage was the ease with which one can access this money; getting them off the digital form into hard cash in one’s hand. The two made the strongest business case for the platform.

However, there is a rise in the use, and types of,  digital currencies; to take immediate examples, Bonga points now have transactional value, so do KPLC tokens and cash vouchers. All these are currencies used in the place of cash, negating the need to translate digital currencies to hard cash.  If the negation becomes widespread, then M-Pesa’s  much touted advantage of agents’ network, which is relevant because of the need to remove the cash from its digital form, becomes less significant.

Still, even if the need to translate digital currency into hard cash persists, and while still in the current telcos scenario, an option of transferring value independent of a telco as a gatekeeper is feasible. The University of Cambridge and Strathmore University are working on a mobile electronic purse system called DigiTally  whose design and implementation would be put in public domain to be accessed by anyone.

So, what does this context mean for the user?

In this scenario, the issue becomes what it costs to access the service and knowledge on how to use it. The elimination of the USSD and SIM toolkit interface means the issue is the cost of data, which will be negligible. A Facebook page for example, is, on average  800KB. An SMS is 1KB. Those are more than 800 M-Pesa transactions. A Youtube video is the equivalent of 10000 M-Pesa transactions. In such a scenario, there won’t be a need even for M-Pesa to charge transaction fees, a plus to financial inclusion efforts.  They would find their value elsewhere, like the data accrued from all these transactions, which they can use to upsell people on products.

However, M-Pesa is not over the web yet, and all this is not possible for them in their current state. Which leaves them at a huge disadvantage.


M-Pesa wasn’t probably built with the mindset of creating a global platform. This might have hurt its chance of becoming a global product. That, and their current style of operation, could mean  their  chance to continue staying at the top of mobile money may have passed them.

We are starting to get beyond mobile money, as we know it. What would probably happen is there will be developed a mobile money app that will be ubiquitous, either by M-Pesa, or someone else. There is also a big chance that  future mobile money platforms will be developed independent of the M-Pesa API. What is almost certain, though, is  whoever develops an efficient web based application for mobile money , even with a starting a user base of three, stands a good chance of staking the future of mobile money, because, not only will it operate over the web, but will likely be open to several currencies as well, leading to a truly global mobile money platform.

Leo Korir, CEO of Mobidev Kenya Ltd provided significant guidance on how mobile platforms, and mobile money platforms work.

I also spoke to Michael Kimani of Umati Blockchain Ltd, at length, about digital currencies.

Featured image by Vodafone

1 Comment

  1. Mpesa needs NFC.

Leave a Reply