Stop worshipping innovation. Start worshipping reliability
The Indian economy of the next twenty years will be defined less by how fast it can innovate and more by whether the foundations it has built can carry the weight it intends to put on them.
It is 8:47 on a Tuesday evening in Indiranagar. A man holds his phone over a QR code at a tea stall. The screen turns. He waits. The waiting becomes long. He tries again. The payment fails. He apologises to the vendor, fumbles for cash he does not have, and walks away. Behind him, a queue of three customers is already starting the same dance. The vendor, who has not handled paper currency in months, watches forty rupees disappear into the digital ether.
This small failure, multiplied by millions of similar failures across the country at the same hour, is not a story about technology. It is a story about how we built it.
We are living in the age of innovation worship. Every quarter brings a new feature, a faster app, a smarter assistant, a freshly raised round. The conversation in boardrooms is about acceleration. The conversation in news cycles is about disruption. And yet, beneath this confident vocabulary, the systems we depend on most are breaking more often, in more visible ways, than they used to. The phone that does everything cannot reliably do one thing. The bank with the most modern interface goes dark for three hours on a Saturday morning and pretends it did not. The food delivery app that promised dinner in twenty minutes shows a spinning circle and then forgets the order existed.
The conventional explanation is that we have not innovated enough. The truer explanation is the opposite. We have over-optimised.
Modern systems are fragile not because they lack technology, but because they have been engineered, decade after decade, for maximum efficiency and minimum slack. Buffers were removed. Backup systems were retired. Redundancy was sold off as waste. A payment that once moved through three independent networks now flows through one. A service that once had its own infrastructure now rents from a cloud provider that rents from another cloud provider. Inventory was squeezed out of supply chains. Spare capacity was eliminated. The result looks beautiful on a balance sheet. It is also, in a way that has only recently become visible, fragile in a manner the previous era was not.
This fragility has a specific shape. It does not announce itself with a single dramatic failure. It cascades. A small error in one service triggers a slowdown in another. The slowdown causes a queue. The queue trips a circuit breaker. The circuit breaker disables a third service that depended on the second. By the time anyone understands what is happening, an outage that started with a configuration error in one company is grounding flights at airports in another country.
The world saw this clearly on the 19th of July 2024, when a faulty software update from a single American cybersecurity vendor knocked out millions of Windows machines in a few hours. Hospitals could not access patient records. Airlines could not check passengers in. Television stations went off the air. The cause was not a cyberattack. It was a routine update, deployed slightly wrong, into a system that had been quietly arranged so that a single point of failure could affect a great deal at once. The technology had not failed. The architecture had.
Efficiency made systems faster. It also made them easier to break.
This is the part of the modern economy that India is now living more intensely than almost any other country. India does nothing on a small scale. UPI moves more transactions in a single month than most national payment systems handle in a year. The country's identity infrastructure serves more than a billion people. Telecom networks carry data at densities most engineering textbooks do not anticipate. The achievement is real, and it is the envy of much of the world. The vulnerability is also real. When something fails in India, it does not fail for thousands. It fails for tens of millions, simultaneously, often during peak hours, often without warning.
A UPI outage on a Saturday evening is not a technology story. It is a story about a tea vendor who cannot accept payment, a daily wage worker who cannot send money home, a small clinic that cannot collect a fee, and a thousand minor commercial relationships that quietly do not happen. None of these failures appear in quarterly results. All of them, added up, shape what citizens come to believe about whether the system works. India's challenge in this decade is not building new systems quickly. It is keeping the ones it has built standing up under the weight of their own success.
This is where the leadership blind spot lies. Resilience does not photograph well. It does not announce itself in a press release. A system that stays up through a stress event produces no headline; the headline arrives only when it does not stay up. The internal incentive structures of most organisations, public and private, reward visible novelty over invisible reliability. New features ship on time. Backup systems are quietly deprioritised. Engineers who build resilient infrastructure are paid less, and promoted slower, than those who launch the new. This is not a moral failure. It is an accounting failure. We have not yet learned to value what we cannot easily see.
The most interesting commercial development of the next decade may be the slow correction of this accounting error. The companies that win in markets like India will not be the ones with the most features. They will be the ones that work on the day when nothing else does. The bank whose app stays open during a regional outage will earn a kind of trust that no advertising budget can purchase. The payment network that handles a festival surge without breaking will be the one merchants quietly default to. The logistics platform that delivers when a competitor's system is offline will keep the customer for a decade. Reliability, in saturated markets, is becoming a category of luxury.
The technical idea behind this is older than the markets have noticed. It is called graceful degradation. A well-designed system, when it cannot do everything, decides what to do less of. A poorly designed system, when it cannot do everything, does nothing at all. The difference between these two responses is invisible on a normal day. On a bad day, it is the difference between an inconvenience and a crisis. The organisations that learn to design for the bad day will, paradoxically, have better good days, because their users will trust them in a way users do not trust services that occasionally vanish without explanation.
There is a deeper lesson here, and it goes beyond technology. The Indian economy of the next twenty years will be defined less by how fast it can innovate and more by whether the foundations it has built can carry the weight it intends to put on them. Fast capital, ambitious policy, and a young workforce can produce remarkable systems. They cannot, on their own, produce systems that survive their own success. Surviving success is a different discipline. It is patient. It is unglamorous. It pays attention to the boring middle of the technology stack rather than the shiny edges. It assumes things will go wrong and asks, calmly, what happens next.
The man at the tea stall in Indiranagar does not care which company wrote which line of code, or which board approved which strategy. He cares whether his payment goes through. The country that figures out how to give him that simple thing, reliably, on a Tuesday evening when half the network is under stress, will have built something more valuable than any number of new apps. The next competition is not between the fastest builders. It is between those who can keep working when everything else stops.
(Aditya Vikram Kashyap is currently Vice President at Morgan Stanley, New York. Kashyap is an award-winning technology leader. His core competencies focus on enterprise-scale AI, digital transformation, and building ethical innovation cultures. Views expressed are strictly his own and do not reflect any entity or affiliations, past or present.)
It is 8:47 on a Tuesday evening in Indiranagar. A man holds his phone over a QR code at a tea stall. The screen turns. He waits. The waiting becomes long. He tries again. The payment fails. He apologises to the vendor, fumbles for cash he does not have, and walks away. Behind him, a queue of three customers is already starting the same dance. The vendor, who has not handled paper currency in months, watches forty rupees disappear into the digital ether.
This small failure, multiplied by millions of similar failures across the country at the same hour, is not a story about technology. It is a story about how we built it.
We are living in the age of innovation worship. Every quarter brings a new feature, a faster app, a smarter assistant, a freshly raised round. The conversation in boardrooms is about acceleration. The conversation in news cycles is about disruption. And yet, beneath this confident vocabulary, the systems we depend on most are breaking more often, in more visible ways, than they used to. The phone that does everything cannot reliably do one thing. The bank with the most modern interface goes dark for three hours on a Saturday morning and pretends it did not. The food delivery app that promised dinner in twenty minutes shows a spinning circle and then forgets the order existed.
The conventional explanation is that we have not innovated enough. The truer explanation is the opposite. We have over-optimised.
Modern systems are fragile not because they lack technology, but because they have been engineered, decade after decade, for maximum efficiency and minimum slack. Buffers were removed. Backup systems were retired. Redundancy was sold off as waste. A payment that once moved through three independent networks now flows through one. A service that once had its own infrastructure now rents from a cloud provider that rents from another cloud provider. Inventory was squeezed out of supply chains. Spare capacity was eliminated. The result looks beautiful on a balance sheet. It is also, in a way that has only recently become visible, fragile in a manner the previous era was not.
This fragility has a specific shape. It does not announce itself with a single dramatic failure. It cascades. A small error in one service triggers a slowdown in another. The slowdown causes a queue. The queue trips a circuit breaker. The circuit breaker disables a third service that depended on the second. By the time anyone understands what is happening, an outage that started with a configuration error in one company is grounding flights at airports in another country.
The world saw this clearly on the 19th of July 2024, when a faulty software update from a single American cybersecurity vendor knocked out millions of Windows machines in a few hours. Hospitals could not access patient records. Airlines could not check passengers in. Television stations went off the air. The cause was not a cyberattack. It was a routine update, deployed slightly wrong, into a system that had been quietly arranged so that a single point of failure could affect a great deal at once. The technology had not failed. The architecture had.
Efficiency made systems faster. It also made them easier to break.
This is the part of the modern economy that India is now living more intensely than almost any other country. India does nothing on a small scale. UPI moves more transactions in a single month than most national payment systems handle in a year. The country's identity infrastructure serves more than a billion people. Telecom networks carry data at densities most engineering textbooks do not anticipate. The achievement is real, and it is the envy of much of the world. The vulnerability is also real. When something fails in India, it does not fail for thousands. It fails for tens of millions, simultaneously, often during peak hours, often without warning.
A UPI outage on a Saturday evening is not a technology story. It is a story about a tea vendor who cannot accept payment, a daily wage worker who cannot send money home, a small clinic that cannot collect a fee, and a thousand minor commercial relationships that quietly do not happen. None of these failures appear in quarterly results. All of them, added up, shape what citizens come to believe about whether the system works. India's challenge in this decade is not building new systems quickly. It is keeping the ones it has built standing up under the weight of their own success.
This is where the leadership blind spot lies. Resilience does not photograph well. It does not announce itself in a press release. A system that stays up through a stress event produces no headline; the headline arrives only when it does not stay up. The internal incentive structures of most organisations, public and private, reward visible novelty over invisible reliability. New features ship on time. Backup systems are quietly deprioritised. Engineers who build resilient infrastructure are paid less, and promoted slower, than those who launch the new. This is not a moral failure. It is an accounting failure. We have not yet learned to value what we cannot easily see.
The most interesting commercial development of the next decade may be the slow correction of this accounting error. The companies that win in markets like India will not be the ones with the most features. They will be the ones that work on the day when nothing else does. The bank whose app stays open during a regional outage will earn a kind of trust that no advertising budget can purchase. The payment network that handles a festival surge without breaking will be the one merchants quietly default to. The logistics platform that delivers when a competitor's system is offline will keep the customer for a decade. Reliability, in saturated markets, is becoming a category of luxury.
The technical idea behind this is older than the markets have noticed. It is called graceful degradation. A well-designed system, when it cannot do everything, decides what to do less of. A poorly designed system, when it cannot do everything, does nothing at all. The difference between these two responses is invisible on a normal day. On a bad day, it is the difference between an inconvenience and a crisis. The organisations that learn to design for the bad day will, paradoxically, have better good days, because their users will trust them in a way users do not trust services that occasionally vanish without explanation.
There is a deeper lesson here, and it goes beyond technology. The Indian economy of the next twenty years will be defined less by how fast it can innovate and more by whether the foundations it has built can carry the weight it intends to put on them. Fast capital, ambitious policy, and a young workforce can produce remarkable systems. They cannot, on their own, produce systems that survive their own success. Surviving success is a different discipline. It is patient. It is unglamorous. It pays attention to the boring middle of the technology stack rather than the shiny edges. It assumes things will go wrong and asks, calmly, what happens next.
The man at the tea stall in Indiranagar does not care which company wrote which line of code, or which board approved which strategy. He cares whether his payment goes through. The country that figures out how to give him that simple thing, reliably, on a Tuesday evening when half the network is under stress, will have built something more valuable than any number of new apps. The next competition is not between the fastest builders. It is between those who can keep working when everything else stops.
(Aditya Vikram Kashyap is currently Vice President at Morgan Stanley, New York. Kashyap is an award-winning technology leader. His core competencies focus on enterprise-scale AI, digital transformation, and building ethical innovation cultures. Views expressed are strictly his own and do not reflect any entity or affiliations, past or present.)
