Why Does Every Government Website in India Crash When We Need It Most?

Every Indian has lived this moment. The 10th or 12th result day arrives. A student who studied for two years sits in front of a screen, types the roll number, and waits. The page spins. It times out. It throws an error. The family refreshes for three hours. A railway booking opens at 8 in the morning and the site freezes at the exact second tickets go live. The NEET form refuses to upload a photo. The GST portal slows to a crawl on the filing deadline.

This is not bad luck. It happens every single time, under every single government. The party in power changes. The website still crashes. So the problem is not political. It is deeper than that.

Let us be honest about what is really going on.

The excuse we are always given

The standard explanation is “too much traffic.” Lakhs of people hit the site at the same time, so it slows down. This sounds reasonable. It is also an excuse, not a reason.

Think about what else handles huge traffic without breaking. When India played a World Cup final, millions watched it stream live without the screen freezing for everyone. When a big sale opens on a shopping app, crores of people log in at midnight and the app holds. When your salary is credited, lakhs of bank transactions clear in seconds.

So the technology to handle a crowd already exists. It is mature. It is affordable. The question is not whether it is possible. The question is why the government does not use it.

What modern systems actually do

There is a simple idea in computing called scaling. When more people arrive, the system adds more machines to share the load. When the crowd leaves, it removes those extra machines so no money is wasted. This is automatic. It happens in seconds. It is the basic feature of every serious online service today.

A result website knows exactly when results will be declared. The date is announced days in advance. There is no surprise. A competent team would prepare for the rush, test the site with simulated traffic, and arrange extra capacity for that one day. None of this is secret knowledge. It is ordinary engineering.

When a government site still crashes on a known, scheduled, predictable day, it tells us the preparation never happened.

Look at our own banks

We do not need to look abroad for proof. Look at ICICI, HDFC, and Axis. These banks run apps and websites used by crores of people every day, moving real money where a single error has real consequences. They mostly work. And when they do need to shut down for maintenance, they send you a message a week earlier. “Our services will be unavailable on Sunday from 1 AM to 4 AM.” They plan. They warn. They respect your time.

The same government that regulates these banks, that sets the rules they must follow, cannot run a website that shows a student his marks. The institutions under the government perform better than the government itself. That gap is the whole story.

The contracting problem nobody talks about

Here is the part that goes deeper than technology.

Government websites are usually built by the lowest bidder. The tender goes out, and the contract goes to whoever quotes the cheapest price, not whoever builds the best system. A cheap quote means a cheap build. No load testing. No backup plan for the rush hour. No real ownership once the site goes live. The vendor delivers something that works on a quiet day and disappears.

Private companies build differently because their survival depends on it. If a shopping app crashes on sale day, it loses crores and its competitor wins. That fear forces quality. A government department feels no such fear. The result page can crash and no official loses anything. There is no penalty for failure, so failure repeats.

This is why changing the ruling party never fixes it. The party changes. The contracting system does not. The incentives do not.

The bigger shadow: we do not own the ground we stand on

There is a second, quieter problem hiding behind all of this.

Most of India’s serious digital systems run on cloud infrastructure owned by foreign companies. Amazon Web Services, based in the United States. Microsoft Azure, the United States. Google Cloud, the United States. Alibaba Cloud, China. A handful of giants in two countries hold the machines that run much of the modern world, including large parts of India.

India does have some homegrown players. Yotta, CtrlS, ESDS, and the government’s own National Cloud run by NIC. But these are small compared to the foreign giants, and our heaviest, most important systems still lean on outside infrastructure.

This is what people mean by cloud sovereignty. Just as we now worry about depending on foreign companies for artificial intelligence, we should worry about depending on them for the very ground our digital life stands on. If a foreign government tightens export rules, changes a policy, or applies pressure during a tense moment, the machines that run our services sit outside our control. We would be negotiating for access to our own digital backbone.

This does not mean foreign cloud is evil or that we should rip it out tomorrow. It means a country of our size should be building real capacity of its own, the way we built our own space program and our own payment system. Depending entirely on others for something this critical is a strategic weakness, not a convenience.

We keep failing the same exam

The most frustrating part is that we never learn. The NEET-UG paper was compromised, the exam was cancelled, and a re-examination had to be scheduled. And what was the government’s headline response ahead of that re-exam? It temporarily blocked Telegram across India until June 22, 2026, and ordered the app to switch off its message-editing feature till the end of the month, because cheating rackets were using it to sell fake papers and fabricate leak “proof.”

Think about what that really says. The fix for a broken examination system was to switch off a messaging app for a few days. That is not a solution. That is treating the symptom while the disease sits untouched. The rackets will move to the next app. The forms will still fail to download. The portals will still crash on result day. And next year, we will be here again, blaming the next platform.

The disease is simple. We do not treat public digital systems as serious infrastructure. We build them cheaply, run them carelessly, and prepare for nothing. A road that collapses every monsoon would cause an uproar. A website that collapses every result day is somehow accepted as normal. An exam that leaks three years running is met by banning an app for a week.

It should not be normal. A student waiting for his result is not asking for a miracle. He is asking for a webpage to load. In 2026, in a country that calls itself a digital leader, that is the lowest possible bar. And we keep walking under it.

The technology is ready. The money is available. The only thing missing is the will to treat the citizen’s time as something worth protecting.

Until that changes, the page will keep spinning. And we will keep refreshing, year after year, waiting for a country that can build rockets to finally build a website that works.

Comments

comments