Spark Global Limited reports:
Beyond the loss of revenue and trust, there are urgent lessons from the recent Facebook outage. Muraleedhar Ramapai, executive director of Maveric Systems, said it was a clarion call for banks keen to develop CX capabilities on a large scale to review “graceful downgrades”.
The graceful downgrading of banking
Muraleedhar Ramapai is executive director of Maveric Systems
A planet lacking attention is unlikely to forget a memorable Monday moan in a hurry. Within six hours, Facebook and its family of apps — Instagram, Messenger and WhatsApp — were down. Aside from the proliferation of Fan-obsessed memes, the impact is real on Facebook, where its stock is down 4.9% and advertising revenue is losing $545,000 an hour, not to mention the inconvenience caused by the millions of businesses that rely on Facebook to pay for visits to e-commerce sites. The cost to the global economy from the shutdown is estimated at $160 million, based on specific indicators from the World Bank and others.
An internal Facebook blog pointed out a litany of errors, but when we consider the impact of outages at brands like Target, Amazon Web Services, and Microsoft on BFSI and the fintech market leader, These effects are less auspicious. After all, with heavy investment in digital banking, banks hope (to their credit) to replicate their scale and customer flexibility.
To be fair, not everyone is a sociopath. But concerns about oversight do affect the multi-billion dollar flexibility agenda pursued by companies around the world. Of course, this concern is rooted in rising global uncertainty, fluctuating geopolitical risks, increasing frequency of natural disasters, massive power outages and repeated security breaches.
Resilience is a word easier said than done. But the only difference between big tech and banking is the nature of the service. Not being able to watch your favorite movie, upload pictures or refresh your feed is on the lower end of the anxiety spectrum than a banking system crash. Dessert? Combine the agility and innovation of big tech with the resilience and reliability of the banking system.
To some extent, while regulators have protected customer data and deposits, the number of outages cannot be ignored. In fact, quite the opposite. Consider The 10 outages a month at Barclays, or even the recent outages at Bank of America and Visa.
How urgent is the need for flexibility?
In an era when all companies are eager to move to pure cloud, flexibility demands more than just the FAANG (Facebook, Amazon, Apple, Netflix, And Google). Research by McKinsey shows that companies reporting outages every 3.7 years or more cost almost 45 per cent of one year’s EBITDA over a 10-year period.
So what does resilience mean for our highly connected world? In short, while isolation and social distancing are good for humans; The system requires a different kind of fail-safe. This is known as degeneracy.
Companies need to downgrade gracefully
The theory and practice of graceful demotion (GD) can be found by asking: “What is the most important job to do if all else fails?” As we will soon discover, the answer to this question is neither simple nor simple for network engineers, product managers, UI designers, and CX professionals.
Imagine you’re online at all times — booking a plane seat, withdrawing cash from an ATM, locking in stock market trades, browsing Netflix suggestions, delays when the Internet outages or reaches a threshold, power outages, or unexplained behavior of the system in unexpected ways. After all, complex systems tend to be fragile systems in which macro-level problems such as saturation, latency, and excessive workloads are failures that have more than one root cause.
What happens next? Will your seat be locked? Will the ATM keep your card? Will Netflix offer universal options? How much fault status did the system communicate to the user at what stage? In the case of failure, how much information qualifies as good customer service without causing the anxiety of silence? Does a “bad” system offer an alternative to pleasing customers? If so, how about that? Does it reduce the workload, or change the hours, or reduce the quality of service, or add more capacity? Will the system prioritize the relationship between features — perhaps smoothly admitting new users, rather than allowing delays for users already on the platform?
These are answers to the main question: How elegant should a system reversion be?
Large enterprise applications are resilient because they can maintain limited functionality and prevent catastrophic failures, even if a large portion of them fail.
Design a security degradation system