Startups are a symptom, not Pakistan's solution

The graveyard of Pakistani startups

Muzamil opens by naming names. Airlift raised $85 million and still shut down. Swvl left Pakistan. Waada Cars, CarFirst, Trella, and TruckItIn are gone. Uber closed its Pakistan operations. Retailo, which had raised over $30 million locally, shut its Pakistan business and moved to Riyadh. Daastgyr and Paymob laid off more than half their workforce. Cheetay’s founders are reportedly weighing whether to close the company altogether.

Against that backdrop, the government has just announced a new startup fund: two billion rupees a year, no equity taken, up to $300,000 per company. It is not the state’s first attempt. Muzamil notes that Punjab and federal governments have poured billions of rupees into incubators and accelerators over the past decade without producing a single startup that scaled successfully out of them. The question he sets out to answer is whether this new fund changes anything, and whether startups can move the needle on Pakistan’s economy at all.

A decade of incubators, one real success story

Pakistani startups first appeared in the early 2000s, mostly out of LUMS, FAST, and GIK research circles, but the country’s digital infrastructure was not ready and none of it went mainstream. Umar Saif’s Plan9, launched in Punjab in the early 2010s, was the first serious incubator: office space, connectivity, networking, a path to market. PlanX followed for companies ready to scale, and once Umar Saif moved into the federal government, National Incubation Centers opened in Islamabad, Lahore, Karachi, Quetta, and Peshawar.

A decade and billions of rupees later, Muzamil says the honest tally is one real win. Patari, a music streaming service incubated at Plan9, tried to beat Spotify to the Pakistani market and did not survive. “The only name that has actually worked out is Bookme.pk,” he says, pointing to a company that later raised multiple funding rounds and kept operating successfully while everything else around it stalled before reaching the mainstream.

What Muzamil means by “startup”

Before judging whether Pakistan can produce one, Muzamil pins down what the word actually means. People message him constantly calling a T-shirt line or a naan shop run with a friend a “startup.” Technically, he says, that fits the textbook definition: a venture built to develop and validate a scalable business model. But that is not what investors, venture capitalists, or the media mean by the term.

The real distinction is the cost of scaling. Open a naan shop in Bahria Town for ten lakh rupees and it might return one lakh a month; opening ten shops needs roughly ten times the capital, and the ceiling rises in a straight line. A digital product like Spotify needs upfront investment once, and after that costs barely more to serve ten crore users than a hundred. That is why venture investors are not chasing a steady 20 to 30 percent annual return. They are chasing 40x, 50x, 100x multiples on an exit, and that appetite for an exponential outcome, not steady profit, is what actually defines a startup in the modern sense.

Cheap money built the game before Pakistan ever played it

Muzamil traces the mania back decades before it reached Pakistan. Apple and Microsoft were dismissed by Wall Street through the 1980s until personal computers proved themselves in the 1990s, which primed investors to chase the next “computer” story: the internet. That produced the dot-com bubble and its 2001 crash, when companies lost 80 percent of their value within weeks. After the 2008 financial crisis, the US Federal Reserve cut interest rates to zero, handing investors free capital and rebuilding their appetite for outsized bets, first in Google, Facebook, and Instagram, then across ride-hailing and grocery delivery through the 2010s.

In between those two crashes, Muzamil notes, the surviving giants consolidated rather than competed: Facebook bought Instagram and WhatsApp, Google bought YouTube, Microsoft bought LinkedIn. Each acquisition reinforced the same lesson for Wall Street, that early money into the right idea could return well beyond a normal business, and each retelling of that story widened the pool of capital chasing the next one.

Then came COVID. The US printed roughly $3 trillion in a single year; money supply went from $1.5 trillion in 2020 to $6.5 trillion by 2022. That capital had to park somewhere. It trickled from Wall Street to London, Dubai, and Singapore before landing in Pakistan as part of the roughly $350 million in announced 2021 startup funding, a figure that included Airlift’s $85 million round for an app delivering groceries in ten minutes on a bike. Muzamil calls this whole stretch, from the ride-hailing boom through the pandemic, an exercise in “quack economics”: capital chasing growth numbers and user counts with no real interest in whether the underlying business could ever turn a profit. India’s Byju’s is his outside example of where that road ends, a company once held up as a generational success story that took on heavy debt and had to restructure once the free-money era closed.

Careem’s exit sold Pakistan a fantasy

The single biggest validator of the “Pakistani startup” idea, in Muzamil’s telling, was Careem. Its engineering ran out of Karachi, its middle management was largely Pakistani, and its growth into the largest startup in the Middle East built a national narrative that a Pakistani-led company really could scale. When Uber acquired Careem for $3 billion in January 2020, employees who had joined straight out of university on modest salaries and stock options woke up as millionaires overnight.

Muzamil is careful not to single out individuals caught up in that moment, but the effect on the wider market was structural. A generation of “Careemers,” and the investors watching them, concluded that if Dubai-based Pakistanis could build and flip a company that fast, the same playbook could run inside Pakistan itself. It set off a wave of copycat founders and family money chasing the same exit, with the same disregard for whether the underlying business made money.

A billion-dollar valuation is not a billion dollars in Pakistan

This is where Muzamil turns from narrative to arithmetic. When a $10 million round is announced, it does not arrive in Pakistan. It lands in a holding company in Singapore or Dubai, and only a fraction trickles down to pay a lean local team of twenty to forty people. The bulk goes to marketing and to foreign vendors, Meta ads and AWS hosting among them. It never touches the local economy at all.

“That doesn’t mean it’s bringing a billion dollars into Pakistan every year,” he says of a headline valuation, which is really one investor’s estimate of worth, not cash flow. When that company eventually sells, the money moves from one investor to another outside the country. Startups, he argues, are fundamentally part of the consumption economy: they optimize spending patterns, the way Careem and Uber turned a taxi’s idle return trip into a paid one. That is a legitimate business in an economy built on consumption, like the United States. It is a much harder case to make in a country losing $2 to $3 billion a year in dollarized terms and financing the gap with debt.

He points to grocery delivery as the clearest case of money substituting for a real business model. Startups were paying a rider’s fuel and wages to deliver a drink that cost less to buy off the shelf a few doors down, selling it to the customer for less than retail while absorbing the loss themselves, all to keep the user and revenue numbers climbing for the next funding round. That subsidy, Muzamil says, was never really a product; it was investor money being spent to manufacture growth metrics, and it evaporated everywhere in the world, not just Pakistan, once free capital dried up.

Consumption economies need manufacturing first

Muzamil’s structural argument is that developing economies climb a ladder: hard industry, then soft industry, then services, and only then a consumption layer. Pakistan, he says, keeps trying to skip straight to the top rung. He points to the everyday example of trying to assemble cars locally: pull the thread from parts to alloys to raw materials, and the supply chain simply does not exist, so manufacturers default to importing everything instead.

India built roughly $150 billion in annual IT exports before its unicorns arrived, creating millions of middle-class jobs and a consuming public underneath the startup boom. Indonesia did something similar, building toward roughly $320 billion in exports over two decades of reform. Pakistan’s oft-cited statistic that 65 percent of its population is young is not the same as a middle-income population with disposable cash to spend on new digital services. A company chasing a billion-dollar valuation, Muzamil says, needs something close to $100 million a year in revenue to justify it, and Pakistan’s consumer market simply does not hold that much spending power yet.

“Startups are not a solution,” Muzamil says. “They’re just a symptom, an outcome of a good, thriving economy.” Pakistan, in his view, has not yet built the economy that symptom depends on. He is careful to add that India’s own sequencing was not perfect either: its heavy tilt toward services over manufacturing has left it with real joblessness and inequality, since not every worker in a densely populated economy can be absorbed into IT jobs. A country Pakistan’s size, he argues, needs the factory floor as much as it needs the services layer, so that a person without a university degree still has a path into a stable, consuming middle class.

The seth problem: family capital that won’t do the work

Muzamil reserves some of his sharpest criticism for Pakistan’s traditional business families, the seths, who he watched pivot from sugar mills and conventional industry straight into startup investing without understanding the sector, chasing the same flip that Careem’s exit had validated. “Our seth, who I keep calling a rent-seeker because he doesn’t want to do the work, is like a degenerate gambler,” he says, contrasting that instinct with how Indian industrial money built Tata Consultancy Services and Infosys: real companies that now employ hundreds of thousands of people in stable, exportable IT services.

His prescription is a partnership Pakistan has rarely tried: businessmen who understand scale and capital allocation working alongside engineers who understand the technology, instead of both sides insisting on full ownership and refusing to trust each other. He calls the country’s failure to build a serious IT services industry, despite having every necessary ingredient, a problem of trust rather than of talent or opportunity.

Startups are a symptom, not a solution

Muzamil closes by resetting expectations rather than dismissing the sector. Startups are good, he says, and will keep emerging in Pakistan as they do everywhere. The problem is the timeline and the promises attached to them. Anyone building one honestly should expect ten to fifteen years to reach real scale, not the two-to-three-year multiple that boom-era investors promised.

He is blunt that the government’s new fund misreads the moment. “This isn’t helping Pakistan’s economy,” he says. “It’s helping some private investors, definitely, but the economy at large isn’t benefiting from it.” Two billion rupees a year spent building actual technical talent, he argues, would do more for the country than another equity-free bet on the next Careem, since Pakistan’s IT industry is limited far more by a shortage of trained people than by a shortage of ambition.

His advice to a young graduate who genuinely wants to build something is to go ahead, but to build for the value rather than the race: serve two customers or two million with the same patience, and stop measuring progress against a two-year payoff that the market’s current variables simply do not support. His closing appeal is directed at the startup community itself: stop treating criticism as betrayal, the way Pakistani cinema once demanded blind loyalty regardless of quality, and start treating honest critique as the only way the same mistakes stop repeating.