The Hard Things About Subscription Payments in Africa

Reading Time: 15 minutes

Over the last few years, there have been many developments in the consumer payments space in Africa, led by the explosive growth of mobile money. It doesn’t seem to be slowing down ever since its introduction over fifteen years ago.

Despite this, online subscriptions still seem incredibly hard to crack for most countries.

South Africa and Mauritius are notable exceptions due to their healthy rates of card ownership.

In some countries, specific payment methods have emerged in recent years, such as Pay By Bank in South Africa and virtual cards in Nigeria. All of these payment methods have been integrated with local e-commerce platforms in each country, and are strong drivers of online purchases that are still growing today.

For the rest of the continent, it’s quite clear that people just don’t have as many cards as they do have accounts with different types of financial institutions. And why would they, when they can make instant payments and withdraw cash at both ATMs and human agents using their mobile phones?

The payment method mismatch isn’t even the biggest issue. Most global subscription services price and charge in major world currencies like the US Dollar, so even users who have cards in Africa aren’t always able to use them for purchases on foreign websites.

This is because many African countries have regulated foreign exchange controls on their currencies. A few of them are even struggling with forex shortages, leading to some drastic actions in recent years. In March 2022, Nigerian banks started restricting foreign currency transactions and in October 2023, Egyptian banks did the same. These are two of the top 3 largest economies by GDP in Africa, where many of the ideal customers for these services are.

Beyond the core infrastructure challenges, there are nuances that exist in the behavioural economics of people in Africa that have largely been influenced by the prepaid mechanisms surrounding most services. This, coupled with an absence of easily accessible consumer credit, has led to people being used to settling their recurring expenses in very specific ways.

In order to better understand all of the different ways all of these challenges interact to create this environment, it’s helpful to break them down into layers:

Product: Why Cards Work Well For Recurring Payments

There’s a certain magic to how card payments generally work online. You enter the card number and CCV/CVV into an online form just once, and whoever you’re paying can keep charging you. This is done through a process called tokenisation.

Let’s start with some definitions:

Merchant: the business entity offering a product or service available for sale via a digital payment method.

Issuer (Issuing Bank): the financial institution that issues debit or credit cards to consumers on behalf of the card networks, such as Visa and Mastercard.

Acquirer (Acquiring Bank): the financial institution that processes credit and debit card transactions on behalf of a merchant.

Tokenisation in card payments, especially for online recurring services, works by replacing sensitive card details with a unique identifier or “token.” This process involves several steps:

  1. Initial Transaction: When a customer first enters their card details for a recurring service, these details are sent to a tokenisation service provided by a payment gateway or acquirer.
  2. Token Generation: The tokenisation service replaces the actual card details (like the card number) with a randomly generated token. This token is unique to the customer and the merchant but does not carry any meaningful card information itself.
  3. Storing Tokens: The merchant stores this token instead of the actual card details. This way, they can initiate future transactions without needing to store sensitive card information, reducing the risk of data breaches.
  4. Recurring Payments: For subsequent transactions, the merchant uses the stored token to process payments. The payment gateway recognises the token, maps it back to the original card details securely stored on its end, and completes the transaction.

Because of this buttery smooth experience for both the customer and the merchant, it’s relatively worthwhile for merchants to invest in signing up customers that they can easily monetise on a recurring basis.

The basic principle is that with cards, the customer only has to say “yes” once to the merchant until they are ready to say “no”.

With mobile money, this concept doesn’t exist in quite the same way. Typically, a mobile money user has to say “yes” to authorise every transaction each time, with either their PIN code, password or biometric. For merchants, this isn’t as attractive as card users, as the merchant has to make a concerted effort to convince the user to go through this authorisation flow every time.

Online merchants tend to use two key financial metrics to determine how profitable signing up a user can be. These are customer acquisition cost (CAC) and lifetime value (LTV):

💡 CAC (Customer Acquisition Cost): The cost associated with acquiring a new customer. It includes all marketing and sales expenses over a specified period, divided by the number of new customers acquired in that period.

💡LTV (Lifetime Value): The total revenue a business expects to earn from a customer throughout their relationship with the company. It’s a prediction of the net profit attributed to the entire future relationship with a customer.

In the eyes of online service merchants, the LTV of mobile money customers will seem to be a lot lower than that of card customers, since they can cancel much more easily by not authorising their next payment. The CAC of mobile money customers may also be higher, because of the increased marketing effort to keep them coming back.

Method Authorisation Factor Recurring Payment UX
Card Card Number Default is opt-in at first payment, user has to opt-out
Mobile Money PIN code Default is opt-out, user has to opt-in for each payment

So why don’t mobile money operators just build out the technology that enables the same flow as cards? That seems like the obvious solution, right?

Turns out it’s not so simple. Let’s dive in to see why.

One of the key differences between mobile money and cards is the speed of settlement.

💡 Settlement refers to the process where the funds from a transaction are transferred from the customer’s account to the merchant’s account.

With card transactions, the customer’s payment may be authorised and deducted almost immediately but the actual settlement to the merchant’s account can take several days. This delay is due to the processing time required by intermediaries and banks to verify, clear, and transfer the funds. Thus, the time the customer pays and the time the merchant receives the money are not the same in card transactions.

With mobile money, this is a completely different story as the funds are settled to the merchant’s mobile money account instantly.

This is great for the merchant, and I’m sure card accepting merchants would love to have this as a feature – some acquirers and payment gateways do offer it today.

It’s not so great for the folks who actually give you the payment instrument though (issuers), because of fraud.

According to Juniper Research, fraud cost the global e-commerce industry over $41 billion in 2022. A lot of this fraud takes place on top of card rails, despite there being an opportunity for compliance teams within financial institutions to catch the fraud before the money leaves the system due to the settlement time.

With mobile money systems, the risk of fraud is even harder to manage because the funds settle instantly, and can be converted to cash quite quickly due to the prevalence of mobile money agents and ATMs for easy withdrawals.

This creates a potential nightmare for mobile money operators, who already dealt with over $1bn of fraud cases in Africa alone in the year 2021.

Introducing recurring payments on top of mobile money in the way that it works for cards would be a big risk for any provider. This is especially true given that according to the GSMA State of the Industry Report on Mobile Money for 2023, 31% of mobile money account holders in Sub-Saharan Africa cannot use their account without help.

Imagine the sort of crafty schemes mobile money fraudsters could come up with to take advantage of innocent people who may not fully understand what they are opting into, especially as they have no reference point anywhere else for recurring payments.

Couple this with the abundant liquidity of cash with mobile money agents, and criminals could easily exploit such systems and cash out before anyone was the wiser.

Apart from this, mobile money operators have to consider the cost from their infrastructure providers of developing and maintaining their systems. For example, while Safaricom owns and operates M-PESA in Kenya, the technology it runs on is actually a platform developed by Huawei. Similarly, MTN’s MoMo platform runs on top of Ericsson’s Converged Wallet platform. Both Airtel Money and Orange Money run on top of Comviva’s Mobiquity platform. The economics of these deals are uniquely negotiated, but there almost always is a component of cost for the operator to develop and implement new functionality on the system.

This makes the idea of tokenising mobile money payments in the way that cards can be tokenised a risky endeavour.

All the same, suppose a mobile money operator is able to mitigate these challenges and proceeds to develop recurring payments for their product, what are the other concerns?

One of the biggest ones is the overhead cost of marketing the new feature and educating their customers on how to use it safely. Soon after launch, the mobile money operator can also expect a significant increase in support cases as customers learn how to use the new feature. All of these are factors that have to be considered carefully before new functionality is availed.

What Recurring Payments Look Like on Mobile Money Today

The industry in Africa has made tremendous strides in recent years towards enabling mobile money users to experience recurring payments. Since 2020, MTN MoMo has supported a form of recurring payments through their Scheduled Transfer feature. In October 2023, Safaricom announced a similar offering through their feature called M-Subscriptions.

While these features achieve the same goal of making recurring payments to a merchant, they don’t quite work the same way as recurring payments on cards.

Below is an illustration of how recurring card flows work:

With mobile money, the process is different:

You will notice that it seems that the mobile money flow looks a lot simpler than the card flow. This is because mobile money payments for recurring orders are almost always within the same mobile money network. This makes the logistics of moving money from one account to another account very easy, as it’s all housed within the same institution. It is possible to move money across different mobile financial institutions, something called mobile money interoperability, but it remains to be seen how well this will work for merchant payments with different providers at scale.

Another key difference between card recurring payments and mobile money recurring payments is that with card, the merchant pulls the payment from the bank account of the cardholder using the token. With mobile money, the mobile money user pushes the payment to the merchant instead.

This introduces challenges in maintaining the subscription, for example if the price changes (either lower or higher), the card payer doesn’t need to do anything to get the new price, whereas the mobile money user has to cancel their existing recurring order and create a new one with the new price.

In recent years, over the top (OTT) mobile money platforms – those that aren’t provided by mobile telephone operators – have come up with creative approaches to recurring payments. Specifically, in Nigeria, platforms such as Paga (20m users), and Chipper Cash (5m users) both support recurring payments in a similar flow to the way cards do.

All of the product challenges around recurring payments can and likely will be solved for in very creative ways over the next few years, but that won’t automatically means subscriptions will magically take off in Africa.

Process: Aligning Subscriptions with the Prepaid Economy

In Africa, most consumers show a marked preference for prepaid payment models over postpaid options, a trend that is deeply rooted in both cultural and practical factors. Prepaid models resonate well with the African consumer psyche which prioritises direct control over expenses and a clear understanding of cost structures. This preference is partly driven by a desire to avoid the unpredictability and potential debt associated with postpaid models.

The reluctance towards postpaid models can also be attributed to a lack of widespread understanding of how these systems operate. In many African countries, financial literacy levels are varied, and the complexities of postpaid billing — with its variable charges, potential for hidden fees, and sometimes opaque terms and conditions — can be daunting for consumers who are more comfortable with the straightforward nature of prepaid services.

In prepaid models, consumers pay upfront and consume exactly what they have paid for, which aligns well with a general inclination towards financial caution and risk aversion.

Moreover, the economic landscape in many African countries, characterised by inconsistent income streams for a large portion of the population, makes prepaid models more practical and attractive. They offer a level of flexibility and control that postpaid services often cannot match. As a result, while postpaid models are common in many parts of the world for services like mobile phones, internet, and utilities, in Africa, the prepaid model continues to be the preferred choice, reflecting both a cultural norm and a practical necessity.

Because of this reality, the idea of giving a 3rd party direct access to charge your account without your explicit consent each time can be incredibly uncomfortable for many consumers.

The typical African consumer wants to be able to pay upfront for only what they need in consumables and services. You see this play out in very clear ways in most of Africa’s major urban centres: people buy only the fuel, mobile data, food and electricity that they need until their next pay day.

Some global merchants have understood this reality very well and have adapted their offerings accordingly. For example, in Kenya, Spotify offers users the ability to prepay for specific time periods without automatic recurrence using M-PESA mobile money.

The Spotify prepaid plans are priced the same as, and work very similarly to card payments.

You pay upfront for a period of service, but the difference is that you aren’t going to be charged automatically at the end of the period. I imagine that global merchants like Spotify who are already working with local payment methods in Africa will be among the first to enable them for recurring payments as soon as the functionality is availed.

There are some interesting tactics deployed in the market to incentivize non-card users to keep their payments going for services. For example, Multichoice, the largest pay TV provider in Africa charges a reconnection fee to customers who pay for their DSTV subscription after the due date.

Similarly, in South Africa, Netflix offers a mobile carrier billing option for Vodacom South Africa customers. This lets users pay with their airtime balance or add the Netflix charge to their postpaid bill. This method also supports payments on a recurring basis. As of the time of this writing, this is service is only available on Vodacom – South Africa’s largest telecom by mobile subscribers at over 44 million.

This is probably the best option for most subscription services today, as airtime is an abundantly available form of value all across the continent. The main drawback to using this is the very high cost to the merchant of acceptance, which can be many times higher than the cost of accepting either cards or mobile money.

Earlier this month, some research from Omdia was made public that claims to have an indication of Netflix’s customer base in Africa as of 2022:

In my view, this data is likely at least directionally correct, as South Africa is the only market where I’ve seen Netflix invest heavily in supporting alternative payment methods. Apart from the direct carrier billing, Netflix also supports gift cards in South Africa.

Gift cards for several major global services are commonplace in most major retailers in South Africa, as it seems to be the largest market for these merchants.

Despite not being the largest economy in Africa, South Africa boasts one of the highest per-capita GDPs in the region, at about $7,100. This is comparable to Brazil in terms of consumer spending power.

Ultimately, the market opportunity would be the primary determinant in whether or not a global merchant will put in the effort to support the myriad of payment methods that exist in each African country.

This brings us to the final layer.

People: How Much They Can Really Spend

Likely the most significant consideration that global merchants make is whether investing in supporting mobile money and other alternative payment methods is commercially worthwhile in the end.

Africa is a region where copyright infringements and piracy violations are notoriously difficult to enforce, and as a result a lot of media and software is illegally distributed with creators never seeing their earnings.

Where piracy isn’t an option, people still love a good deal.

The relatively small size of the middle class in individual African countries and even entire regions is often unattractive when compared to single countries in other parts of the world that could be much larger. Renowned fintech investor Jake Kendall, once made the point that the GDP of Nigeria – Africa’s largest economy, is only slightly smaller than that of Boston, MA.

While very many consumers on the continent might be more willing to consume more subscription services if their payment method of choice is available, it doesn’t always make business sense for the merchants to make them a priority.

On the merchant side, they have to consider the technical work required to alter their UI to support the payment methods. Oftentimes, the technology is centralised and it’s not always easy to show users different experiences based on their country.

Additionally, merchants must also consider the effort of accepting and reconciling payments in multiple “exotic” foreign currencies with volatile exchange rates. If you don’t have enough addressable customers, it can be quite challenging to present a credible business case.

The Africa subscriptions opportunity starts to look a lot more attractive to merchants if they can combine multiple markets to address parts or all of the continent as a whole.

This is precisely the reason that global payment aggregators, like Ebanx, where I work exist.

But I’m not here to sell you anything.

The point of this piece was to illustrate some of the complexities associated with subscription payments in Africa, and to hopefully encourage the continued debate on how the industry can work more closely together to tackle them with innovative solutions.

What does the future of subscription payments in Africa look like to you? What might I be missing here? What am I wrong about? What am I right about?

I hope you enjoyed reading this as much as I enjoyed writing it, and I’m always happy to chat and receive your feedback on Twitter/X or on email via

Finally, please do consider sharing this article if you found it helpful. Every view counts!