Payment resilience has become a boardroom issue

In the Middle East and Africa, the drive to have payments happening constantly, in real time, has completely removed any leeway for systems to go down. Payments are a continuous flow. Breakdowns don't mean transactions line up waiting to go through - they simply halt, and the knock-on effect is felt throughout retail, travel, collecting for the government, and international trade. 

At the same time, a relatively small number of very large companies do the bulk of the work in these payment systems, managing card issuing, handling payments to businesses, and digital shopping, all on a national scale. Add in how much they rely on big cloud computing companies and payment networks around the world, and you can easily see that a single point of failure could cause huge problems in many countries.

Because of this, planning for resilience in this part of the world isn't just a thought experiment anymore. Banks and other institutions are planning for things that are extreme, but could happen: trouble in a region, power cuts, problems with connections between countries - and still being able to process all payments for the country without a hiccup.

Regulators have responded to this. Central banks aren’t just offering advice anymore; they're saying what must happen. Institutions have to prove they are resilient - with processing happening at multiple locations, almost instant recovery from problems, and the ability to continue working during a disruption. The focus is changing from getting things going after a problem to keeping them going all the time.

This is causing a major change in how these systems are built. The best platforms are going beyond the usual ‘disaster recovery’ to ‘active-active’ systems. This means multiple locations that can each manage all payment activity independently, with data copied between them in almost real time, and a smooth switch to another location if one fails.

The principles of good design are changing at the same time. Resilience isn't about just having a spare copy of something anymore, it’s about completely removing anything that both systems would rely on and could cause both to fail. That means having systems in different places, using multiple internet service providers, using a mix of cloud and in-house computing, and sometimes, having extra processing power in a different country from the main one.

And importantly, resilience now has to cover the entire process. Even if the main payment systems themselves stay up, problems with things like setting up new customers, preventing fraud, or how customers interact with the system, can still ruin the experience. Institutions are now building resilience into every stage of a payment.

How well things are done is just as important as the design of the system. When dealing with large numbers of transactions, resilience is about how quickly problems are found, someone takes responsibility for them, everyone is told about them, and they get fixed. Real time tracking, clearly defined responsibilities, and having a plan that has been practiced are what make the difference between a small problem and a complete system failure.

In the end, being online and working is how people trust you. In an economy where everything is done digitally, institutions that continue to operate when something goes wrong strengthen their relationship with customers, partners and the regulators. Those that can't will have problems right away that last a long time.

Resilience in payments has moved from being something the tech team worry about, to being something the board of directors and the government think about. And in places like the Middle East and Africa - where more and more people are using digital payments and everything is linked together - those who see resilience as a key part of their strategy, rather than just something to do to meet a rule, will be the ones who lead the next stage of the industry.