Can Mobile Money Agents act as battery rental outlets, and generate new revenues?

Mobile Money Agents (MMAs) are the end-points in Mobile Money (MM) distribution networks that have density in urban areas, and which become less dense in peri-urban areas, and can be very sparse (or absent) in rural areas.

The commercial viability of MMAs can have an important impact on the end-customers who rely on their services, as well as the wider digital financial services market system. MMAs with weak commercial models are likely to provide a limited range of services to their end-customers, and limit the overall growth of digital financial services.

The Idea

Given 4R Digital’s heritage in digital financial services (our CEO Nick Hughes was behind the development of M-PESA, and co-founded M-KOPA) our hypothesis is that there are many opportunities to expand the range of services and products that MMAs can offer to their customers. Leveraging innovative approaches to financing, it’s possible to offer services to end-customers that meet their needs and provide good value for money, whilst also increasing the revenues of the MMAs.

With support from the Frontier Technologies programme, and the support and collaboration of the FCDO Joint Sahel Department we set out to test this higher-level hypotheses in our first Sprint through a series of experiments — all in Senegal.

The product concept that we wanted to test is of “daily rentable batteries”. This involves batteries which customers can use to light their homes and charge their phones, returning to the MMA each day to replace the battery for a freshly charged unit. The concept of battery rental is not new, and there are several businesses that are doing very well expanding their dedicated kiosk-based businesses. However, this model focuses on leveraging the MMA network to add an additional value-adding service alongside mobile money, and targeting the needs of MMAs and their end customers.

It’s worth noting that starting this endeavour as Covid-19 slowed the world down presented unique challenges, but through the use of technology, and great in-market partners we have managed to complete these experiments and are well set for the next stages (more to follow in subsequent blogs).

The rationale for supporting MMAs

The rationale for this pilot is underpinned by the findings of organisations, such as CGAP and the Bill and Melinda Gates Foundation (BMGF) that are collaborating on a global effort to: “distil recommendations on how networks of convenient, affordable, and trusted CICO (cash-in-cash-out) service points linked to digital finance can be provided to low-income customers in rural areas where financially underserved or excluded people tend to live. It reveals six principles that lead to CICO networks with greater reach and quality[1].” Principle 1 is “Enable Rural CICO Agents to Generate More Revenue Streams: in rural contexts, look for strategies that are conducive to agent non-exclusivity or non-dedication (i.e., offer more than CICO services) because these types of strategies allow agents to maximize revenue streams given fewer rural customers (Unnikrishnan et al. 2019).”

Research supported by the BMGF and undertaken by BCG suggests that “Financial Services providers should [amongst other actions] pursue innovations in business models, operations, and technology to help agents diversify their revenue….provide rural agents with the tools to launch ancillary businesses. Providers could also seek partnerships with companies in other industries, including agricultural suppliers, fuel companies, and fast-moving consumer goods players.”[2]

Choosing the location to test the hypothesis and partnerships

As a business, we have very strong links in East Africa, but the opportunity to test this concept in West Africa was very appealing. The need for various services amongst rural communities is great, and the rapidly expanding MMA networks hold great potential. However, it was agreed that some of the possible markets were too challenging for this type of early stage test for this product, and so we all agreed that Senegal would be the right market to test this solution.

However, the first step was to seek collaboration with mobile money providers and MMA network Aggregators. Of course, these businesses were all very busy reacting to the challenges of Covid-19, but we were delighted when InTouch (an Agent Aggregator with operations throughout West Africa and beyond) agreed to provide access to their agents. InTouch manages these MMAs on behalf of Mobile Money providers in several African markets.

Designing the experiments

A key challenge is that this products concept has several intertwined elements that must all have a strong rationale and solution:

• MMAs must have the necessary incentives to adopt and operate a new service offering. Some schemes (not related to battery rental) push unwanted costs and procedures onto micro-retailers. This inevitably results in low adoption or use of the service.
• End-customers must equally have the necessary incentives (cost, utility, convenience) to adopt a new service that they access from their local MMAs.
• The hardware must be robust and reliable, have the necessary energy capacity, and be enabled with appropriate remote locking and data communications.
• Underpinning all of this, the necessary technology platform being developed by 4R Digital must hold the entire model together, with the ability to function across the entire market.

The hypothetical steps that must be followed by both MMA and end-customer are illustrated here.

As a result, we first undertook background desk research into the market context in relation to mobile money and energy access in Senegal. This showed us that the data on both subjects are reasonably sparse, and do not provide the level of granularity to support product development of this nature. So market research, and market testing would be central to this pilot.

The image below illustrates the hypothetical processes of the interaction between MMA and their customers for this solution.

Testing the hypotheses

We set out to test the relevant hypotheses in three experiments:

• Testing MMAs need, willingness to pay, and ability to pay through in-person interviews (conducted by our in-market research partners CRDES).
• Testing the same factors through telephone surveys with MMA end-customers.
• Exploring (at a high level) the business model of battery rental operators and suppliers in other markets. Also informing our own business model by undertaking early-stage development of the platform modules, and design of the hardware prototypes.

Learnings

Mobile Money — well used but not ubiquitous

Granular data on mobile money usage in Senegal are not available, but the data that are available show that many users visit their MMAs at least once a week. Our own surveys with MMAs showed that they would expect to see their customers on average at least twice a week, more if their business also sold other non-MM products. In addition, MMAs are undertaking a significant number of over-the-counter (OTC) activities where the MMA undertakes transactions (such as paying bills) on behalf of their customers. This is a favourable condition for our proposition — as customers are used to visiting the MMA to make payments, although it does not necessarily encourage the wider acceleration of truly digital mobile money. We also learned that customers are used to visiting their local market and shops frequently (to buy food or other goods).

All of these conditions suggest that a battery rental service based at the MMAs’ location would fit with the behaviours of their existing customers.

Energy services — available but not yet reliable

Both MMAs and their customers appear to be reasonably well served by either grid services (~60% of both groups) or a pay-as-you-go solar home system (PAYG SHS) (~30 of both groups). Surprisingly, the majority of all customers indicated that they would be interested in a daily battery rental service from their MMAs. Discussions with battery rental providers in other markets suggest there may be a few factors behind this initial willingness:

• In markets with presence of PAYG SHS, many customers have some experience with battery failure that takes time and cost to resolve. Since the customers have paid a deposit for the SHS, they consider that they are carrying the burden of the battery failure. With a battery rental they have no perceived liability for the battery and receive a fresh battery each day.
• Even where PAYG SHS offer a good solution (in terms of reliability), there are customers who cannot afford (or raise) the deposit — even if they can afford the daily repayment.
• Our own experience is that grid presence does not equate to grid reliability and that, because of the cost of grid electricity, many customers prefer to pay daily for an off-grid solution that has a known output and daily cost.
• In general, the expansion of the innovative PAYG SHS market has served customers well. However, there has been a general movement of providers toward urban/peri-urban areas, or toward customers with slightly higher incomes, leaving an segment in the market still seeking a solution.

Hardware and software development

Developing the necessary hardware devices

Although the supply side of the market has expanded rapidly over recent years, responding to the growth of the PAYG SHS market, we found few if any suppliers that we could work with to explore this opportunity. Existing operators have developed their models well around slightly larger systems, both in terms of the charging hubs, and the batteries that are rented to customers. the solutions they have arrived at were larger and more costly than we require for this model. Understandably, these providers are focused on meeting the needs of their own businesses and do not currently have capacity to undertake early stage development of a new product prototype — though the opportunity may arise in the future.

As a result, we are working with a product development partner to prototype the charging stations and use refurbished batteries and charge controls to provide to customers in Sprint 2.

It has been interesting to observe the direction of travel of existing providers who appear to be heading toward a future model providing a range of charging services as well as points of distribution for non energy products and services, which could well serve the needs of their customers beyond basic household energy.

Developing the software models to meet the unique market circumstances

4R Digital’s platform has been developed with a number of use cases in mind. Our Dev team has been working hard to develop the modules for this specific solution. This is more a matter of taking into account the specific circumstances in which the system will be working than a complex software development task. Factors that must be taken into account include factors such as the:

• type, frequency, and modality of digital payments,
• availability and reliability of GSM networks, and
• the type, volume, and cost of data to be recorded at the level of customers household and MMA.

Our Dev team has made great progress and we’re in a good position as we enter Sprint 2.

Proof of the prototype is in the testing

We have been somewhat surprised by the expressed appetite for and the willingness to pay for the battery rental service. However, we know from long experience that the most reliable evidence will be customers real parting with their hard earned monies for the service. This is why we are really keen to move to the next Sprint in which we will be able to tests some devices both with MMAs and their end customers.

We believe this offering will be attractive to a significant percentage of MMAs and can provide a valuable service to their customers that do not yet have access to affordable reliable energy services. If successful, this could be the first of a number of additional services provided by MMAs to their customers that will help to increase and diversify the income streams. The wider potential benefits will be to encourage customers to make greater use of digital financial services, and to build their track record in a way that enables the provision of more and more sophisticated services.

More broadly there are a number of other applications of the battery rental system outside of this market setting. For example, the combination of the energy service with a digital payment account would enable direct subsidies to be provided to customers where necessary (for example within communities of displaced peoples).

Next steps…

In the next sprint we will move on to develop the hardware devices, the platform modules, and test the whole system both technically and in the market with MMAs and customers. We are excited to learn much more in this next stage, and to providing updates on progress in subsequent blogs.

This is very much an open-loop learning exercise, and we welcome contacts, comments and suggestions from product providers, MNOs, research organisations, we are always happy to learn and collaborate!

[1] “Agent Networks at the Last Mile: A Guide for Digital Finance to Reach Rural Customers”. CGAP/World Bank 2019.

[2] How Mobile Money Agents can expand Financial Inclusion”. BCG 2019.