(The Weekend Insight) - Have India’s Unicorns Made India More Productive?

In today’s deep dive, we will look at whether India’s unicorns have actually made the country more productive, or mostly made consumption faster and easier. We will examine where startup capital has gone, which sectors have improved output per worker, machine and rupee, and where convenience has been mistaken for productivity.

Agriculture employed 46.1 percent of India’s workforce in 2023-24 and contributed about 17.8 percent of GDP. Almost half the country’s workers were concentrated in a sector producing less than one-fifth of its economic output.

This is where India’s productivity problem begins. It continues through small factories running ageing machines, transport operators managing fleets through phone calls, construction sites with poor control over labour and materials, hospitals buried in paperwork and retailers whose money remains trapped in inventory and unpaid bills.

India’s startup boom unfolded alongside this economy. Over the past decade, the country produced more than 100 unicorns and attracted tens of billions of dollars in venture capital. Startups changed payments, shopping, food delivery, transport, lending, investing and software services. A large part of urban India can now order groceries in minutes, transfer money instantly, compare insurance policies, buy mutual funds and book a cab without speaking to anyone.

The change is easy to see. The productivity gain is harder to establish.

Productivity has a fairly plain meaning. A worker should be able to produce more in the same number of hours. A factory should get more output from the same machinery. A truck should spend less time empty or waiting. A farmer should earn more from the same land without a matching increase in water, fertiliser and labour. A hospital should treat more patients without weakening the quality of care.

Many Indian startups have improved convenience, access and distribution. A smaller group has improved the use of labour, capital and physical assets. The two categories often get mixed together because both use technology, raise venture funding and report rapid growth.

The difference becomes clearer when the discussion moves away from app downloads and order volumes. A grocery order reaching a home in ten minutes says very little about the total labour, real estate, inventory and capital used to complete that order. A payment processed in seconds may represent a genuine reduction in transaction costs. A software product that cuts three days of testing to a few hours has a direct effect on output per employee.

India’s biggest startup successes sit across all three examples. They should not be judged by the same standard.

Capital went where growth was easiest to show

India received about $13.7 billion in venture funding in 2024. According to Bain, consumer technology alone attracted roughly $5.4 billion. Consumer technology, software and SaaS, and fintech together accounted for around 62 percent of total funding. Venture and growth investment increased to about $16 billion in 2025, with the same broad categories remaining dominant.

There was a commercial reason for this concentration.

A consumer startup can launch an app, offer a discount and know within days whether people are willing to use it. Orders, customer acquisition costs, repeat rates and city-level growth are visible on a dashboard. A company that increases transactions quickly can use those numbers to raise another round before the underlying economics are settled.

Selling to factories is slower. The customer may take months to approve a pilot. The software has to work with machinery bought from different vendors over several decades. Plant managers may resist changing established processes. Payment cycles are long. A mistake can stop production rather than merely inconvenience a user.

Agriculture is harder still. Farmers are spread across thousands of villages, incomes are seasonal and the ability to pay for software is limited. Weather, commodity prices and government policy can affect the outcome even when the product works. A tool that improves yield may create more value for the farmer than the startup can capture as revenue.

Consumer demand offered shorter sales cycles and clearer numbers. India’s most serious productivity problems came with long implementation periods, fragmented customers and uncertain pricing.

Venture capital followed the easier customer.

The result was an economy where startups became very good at understanding demand. They know which neighbourhood buys premium milk, what discount makes a customer order again, which advertisement improves conversion and what time people buy snacks. Far less is known about why a machine keeps breaking down, why a truck returns empty, why a construction project is late or why a small manufacturer holds three months of excess inventory.

Demand data leads directly to sales, advertising and credit. Operating data usually creates value through lower waste, fewer defects or better use of assets. Those gains take longer to prove and are often harder to convert into startup revenue.

Software has the strongest case

India’s enterprise software companies come closest to the conventional meaning of productivity.

BrowserStack allows developers to test websites and applications across thousands of browsers and devices without maintaining their own testing laboratories. More than 50,000 teams use the platform, which processes millions of tests every day.

The benefit is measurable. Developers spend less time setting up devices and testing environments. Companies release products faster. Bugs are found earlier. The customer avoids buying and maintaining large collections of phones, tablets and computers.

BrowserStack replaces equipment and manual work with software. There is little ambiguity about the productivity gain.

Freshworks operates in a similar part of the economy. Its products manage customer support, IT services and employee workflows. The company generated $838.8 million in revenue in 2025 and reported positive operating income for the full year.

In May 2026, Freshworks announced an 11 percent reduction in its workforce. Its chief executive said artificial intelligence was writing more than half the company’s code and handling a growing amount of routine work.

The layoffs were painful for the employees affected, but they also exposed a weakness in the usual startup debate. Headcount growth and productivity growth are treated as though they are the same outcome.

A delivery company can add 20,000 workers and still produce relatively little value per worker. A software company can reduce its workforce and produce more software than before. The first company looks better in a job-creation presentation. The second may be more productive in economic terms.

Zoho offers another useful comparison. It generated more than Rs 12,000 crore in operating revenue and over Rs 3,000 crore in profit in FY25. It built software for accounting, customer management, collaboration and business operations without relying on repeated venture rounds.

Zoho created intellectual property, export revenue and a profitable product business. It also did so with far less capital than many consumer unicorns used to build delivery and discount networks.

There is a limitation to the Indian SaaS story. Much of the productivity created by these companies benefits businesses outside India.

An engineer in Chennai may build software that reduces customer-support costs for a company in the United States. A product developed in Bengaluru may help a European business automate internal IT work. India receives export income, engineering jobs and product-building experience. The local manufacturer still operating through paper records may see no improvement.

Indian software has become good at raising productivity abroad. Bringing similar tools into India’s own fragmented business base has proved slower because local customers are smaller, price-sensitive and harder to serve.

Logistics produces clearer physical gains

Logistics is one of the few startup sectors where technology can improve the use of large physical networks.

Delhivery built a common system for parcel delivery, part-truckload freight, warehousing and supply-chain services. Thousands of businesses can use its sorting centres, vehicles, warehouses and software instead of building separate networks.

The efficiency comes from consolidation. Several customers can share one truck, one sorting centre and one warehouse. Better routing can reduce failed deliveries and idle time. A common network can carry more freight without requiring every seller or manufacturer to create its own logistics operation.

Delhivery reported its first full-year profit in FY25. The path was expensive and took years, but the underlying productivity argument remains sound.

BlackBuck works at the level of the truck owner. Its products cover toll payments, vehicle tracking, fuel management, financing and freight discovery. The company generated about Rs 652 crore in operating revenue in FY26, and average monthly transacting truck operators reached roughly 8.66 lakh in the March quarter.

The app itself is not the useful part. The useful part is what happens to the truck.

A small fleet owner should be able to reduce fuel leakage, locate a vehicle, find a return load and spend less time chasing payments. A truck earning revenue for a larger share of its journey is more productive than one returning empty.

Rivigo showed that solving a genuine logistics problem does not guarantee a sound business. Its relay trucking system was designed to reduce transit time and allow drivers to return home more often. The operating idea had merit. The company still consumed large amounts of capital and eventually sold its core business at a small fraction of its earlier valuation.

Rivigo may have improved parts of freight movement. It did not create enough durable value from the capital invested.

This distinction is often missing from startup coverage. A company can improve its customer’s productivity and remain financially inefficient itself. Subsidised prices, excessive expansion and weak margins cannot be excused simply because the underlying problem was worth solving.

Manufacturing platforms need to be judged as industrial businesses

Zetwerk, OfBusiness and Infra.Market are often grouped under manufacturing technology or B2B commerce. Their contribution is real, though the word technology can make the businesses appear lighter and more profitable than they are.

Zetwerk coordinates manufacturing across electronics, aerospace, defence, energy and industrial categories. It works with suppliers and also operates its own facilities. The company can help customers find capacity, manage production and monitor quality. Smaller manufacturers may gain access to orders they could not win independently.

The work depends on factories, engineers, inspections, procurement, working capital and project execution. Software helps hold the process together. It does not remove the physical risks.

OfBusiness solves a different problem. Small manufacturers often struggle to buy raw materials at competitive prices and obtain working capital. OfBusiness combines procurement with financing across industrial inputs.

The company generated Rs 20,645 crore in revenue and Rs 724 crore in profit in FY26. Revenue fell after it exited lower-return activities, while profit increased. That decision suggests greater operating discipline than simply chasing turnover.

OfBusiness remains an industrial procurement, trading and lending company with technology inside the operation. Its value comes from sourcing, credit underwriting, supplier relationships and control over working capital.

Infra.Market has expanded across concrete, tiles, paints, blocks and other building materials. Its FY25 revenue rose 27 percent to Rs 18,472 crore, though net profit fell 42 percent. The group has built and acquired manufacturing capacity and created a large retail and distribution network.

Standardising products, improving supply reliability and using plants more effectively can raise productivity in construction materials. The company is still closer to an industrial group than a software marketplace.

There is no reason to treat that as a weakness. India needs companies willing to build manufacturing capacity. The problem appears when factory ownership, inventory and credit exposure are wrapped in software language and valued as though the underlying capital requirements have disappeared.

The useful measures are factory utilisation, rejection rates, inventory turns, payment cycles, gross margins and return on capital. Gross merchandise value mainly shows how much material moved through the system.

Fintech inherited strong public rails

Fintech is frequently cited as proof that startups transformed Indian productivity. The claim needs some separation between the payment infrastructure and the private companies built on top of it.

UPI reduced payment time, cash handling and merchant costs across the economy. It now accounts for roughly 85 percent of India’s digital-payment volume. India represented about 49 percent of global real-time payment transactions in 2025.

These are large productivity gains. A small merchant can accept payment without a card machine. A customer can transfer money instantly. Reconciliation is easier, and cash does not have to be counted, stored or transported.

UPI was developed by the National Payments Corporation of India as an interoperable payment system linked to banks. Private companies such as PhonePe, Paytm and Google Pay helped millions of people use it. They built interfaces, acquired customers and onboarded merchants. The core rail came from a regulated, shared infrastructure.

A similar pattern runs through Aadhaar, DigiLocker, GST systems, FASTag and Account Aggregator. Public infrastructure handled much of the difficult work of creating common systems. Startups built lending, payments and commercial services on top.

Private innovation was still important. Payment gateways, automated reconciliation and financial-data products from companies such as Razorpay, Cashfree and Perfios reduced manual work for businesses.

The productivity case becomes weaker when fintech moves towards personal loans, trading and consumer credit.

A loan used by a workshop to buy a machine expands productive capacity. A loan used to buy an expensive phone increases present consumption. Both transactions appear in digital lending data.

India became highly efficient at moving money. The businesses and households receiving that money did not necessarily become more efficient at producing income.

Quick commerce shows where the argument breaks down

Quick commerce saves time for the customer. A household can avoid a trip to the market. Working parents can order an item they forgot. Elderly customers can receive basic products at home.

The service depends on dark stores close to residential areas, inventory duplicated across neighbourhoods, pickers waiting inside warehouses and riders available to complete small deliveries.

The order reaches the customer faster because labour, inventory and real estate have been placed nearby. That says little about the total resources used.

BigBasket’s consumer business generated Rs 8,223 crore in revenue in FY26 and lost Rs 3,073 crore. Revenue grew 7.7 percent, while the loss increased sharply.

The company’s older scheduled-delivery model had a clearer efficiency logic. Customers placed larger orders. Warehouses could be centralised. Delivery routes could be planned in advance. Quick commerce requires more local inventory and many more small trips.

BigBasket has had to support its existing operation while spending heavily on dark stores and faster delivery. The investment may be necessary to remain relevant. The numbers do not suggest that the company is buying clear market leadership.

Instamart has improved its contribution margins while expanding beyond 1,000 stores. Blinkit has built strong order density in major cities. These businesses may become profitable through delivery fees, commissions, advertising and better purchasing terms.

Profitability would show that quick commerce can work as a retail model. It would not settle the productivity question.

A dark store can become very efficient at picking orders. The wider system may still encourage smaller baskets, more frequent purchases and repeated delivery trips. The household saves time. The economy uses commercial labour to perform a task that consumers previously completed themselves.

That can be a worthwhile exchange. It should be described as convenience rather than evidence that India’s productive capacity has improved.

Cheap labour made the model possible

India’s consumer internet companies grew in a market with a large supply of relatively low-cost labour. Startups could create a technology experience by putting more workers behind the screen.

A grocery order arrives quickly because pickers and riders are waiting. A food-delivery platform connects a restaurant to the customer through a large labour network. A support problem is handled by adding agents. A warehouse can increase throughput by hiring more people.

These jobs provide income and can be better than the alternatives available to many workers. That benefit is real.

Low labour costs also reduce the pressure to redesign work. A company may hire more pickers instead of investing in automation. A construction contractor may add supervisors instead of using better project controls. A customer-support operation may increase headcount rather than fix the process producing complaints.

The country can become more convenient without seeing a large rise in output per worker.

The long-term issue is the quality of these jobs. A productive economy should allow wages to rise because workers are equipped with better tools, skills and machinery. Platform work that depends permanently on low-value trips and thin margins offers limited room for that improvement.

India needs employment. It also needs jobs where earnings can rise without workers simply putting in more hours.

The difficult sectors remain underfunded

Agriculture should be one of India’s largest markets for productivity tools because it employs such a large share of the workforce.

DeHaat, Ninjacart, CropIn, Fasal, AgroStar and Ecozen work across farm inputs, crop intelligence, irrigation, storage, equipment and market access. Their products address genuine problems. The business models remain difficult because farmers often cannot pay enough for the value created.

Agritech startups frequently move into input sales, lending or produce trading because software alone does not generate sufficient revenue. These additions may help the company survive, though they also make it harder to know whether the business is improving farm productivity or mainly operating another distribution chain.

Healthcare shows a similar pattern. Medicine delivery and doctor discovery scaled faster than hospital workflow, diagnostic systems and clinical tools.

Education startups found it easier to sell courses and test preparation than to improve teacher productivity inside schools. Construction remains one of India’s largest employers, yet material waste, labour planning and equipment use are still poorly tracked across much of the sector.

India spends only about 0.64 percent of GDP on research and development. South Korea and Israel spend several times more. Their private sectors support long research cycles in electronics, semiconductors, defence, biotechnology and industrial equipment.

Indian venture capital has generally preferred a business where the main risk is execution. Scientific risk is harder to price. A company developing a new material, medical device or manufacturing process may require years of testing before earning meaningful revenue. A consumer marketplace can show transaction growth within months.

More money is now entering space technology, defence, robotics, climate technology and advanced manufacturing. The amounts remain small compared with the capital available for commerce and financial services.

The numbers used to judge startups are part of the problem

Funding, valuation, users, orders and gross merchandise value are easy to report. They are less useful for judging productivity.

A manufacturing startup should show changes in output per shift, machine downtime, rejection rates and energy consumption. A logistics company should disclose vehicle utilisation, empty kilometres and warehouse waiting time. An agritech company should measure yield, water use, input costs and farmer income.

Lending companies should separate credit used to buy productive assets from credit used for discretionary consumption. Commerce platforms should show what sellers earn after commissions, advertising, logistics and returns.

The startup itself should be judged on the revenue, cash flow and enterprise value created from the capital raised.

These figures are harder to turn into funding announcements. They also reveal far more about what the business has changed.

India does not need to stop funding consumer companies. People will continue to pay for convenience, choice and faster service. Food delivery, beauty retail, travel, entertainment and quick commerce are legitimate businesses.

Their contribution should be stated accurately.

India’s clearest startup-led productivity gains have come from enterprise software, payment infrastructure, B2B logistics and parts of industrial technology. Even these companies need to be judged carefully. Trading businesses cannot be treated as software companies. Public rails should receive credit for the infrastructure they created. Physical businesses have to earn returns on the capital they use.

The next useful test for India’s startup market sits inside ordinary operations. Whether a factory can produce more without adding another line. Whether a truck can earn more without travelling farther. Whether a farmer can raise income without using more land and water. Whether a hospital can treat more patients without lowering care. Whether better tools can increase what an Indian worker earns from an hour of work.

Until those numbers improve at scale, India’s unicorn boom will remain a larger success in consumption and distribution than in productivity.