Maharashtra logged 1.38 lakh platform workers on the central e-Shram portal as of March 10, 2026, the highest among states, according to NDTV. The figure comes as the Union government recalibrates household labour surveys that have consistently failed to capture platform work, with official projections placing India's gig workforce above 2.3 crore by 2030.
The number matters less for its scale than for what it reveals about measurement. A single state accounting for the largest platform-worker registrations on a national database signals how aggregator activity clusters around logistics corridors, urban consumption nodes, and the informal labour pools that serve them. It also exposes the limits of registration counts alone; Maharashtra's 1.38 lakh is substantial but almost certainly understates the actual number of delivery riders, taxi drivers, and on-demand service workers operating in Mumbai, Pune, and Nagpur.
Maharashtra's lead on the e-Shram portal is not incidental. The state holds the country's largest urban labour markets outside the national capital region, a dense network of logistics and quick-commerce firms, and monsoon-sensitive demand cycles that force both platforms and workers to renegotiate time, distance, and income. When a state with that profile accounts for the largest number of platform-worker registrations, it demonstrates that the gig economy is becoming the default entry point into urban employment for young men and women who are neither in manufacturing nor in formal services. The survey revision must capture that specificity, not just the aggregate.
A Household Survey That Cannot See the App Worker
Traditional employment surveys in India are built around a stable employer-employee relationship: a respondent names a workplace, a supervisor, a wage, a tenure. Platform work dissolves each of those anchors. A delivery rider may switch between three apps in a day, hold no contract, invoice through a third-party aggregator, and report income that fluctuates with weather, festivals, and fuel prices. When a household survey asks whether the respondent is employed, the answer is often yes; when it asks for the employer, the answer is a mobile application. That ambiguity has produced systematic undercounting, and undercounting has consequences that travel far beyond statistical convention.
The decision to adjust labour survey methods is not a technical footnote. It is an admission that official employment data has been describing an industrial-era labour market while the country's fastest-growing employment category operates closer to a just-in-time supply chain. If platform work is invisible or miscounted, then wage benchmarks, social security contributions, accident insurance coverage, and the employment-to-population ratios that investors and multilateral institutions read are all distorted. India's economic formalisation story, often measured through payroll additions and provident fund registrations, misses a large, growing, and relatively unprotected segment that is neither fully formal nor entirely informal.
Registration Is the First Gate, Not the Finish Line
The e-Shram portal was designed to give unorganised workers a portable identity and a pathway to accident insurance and welfare access. Platform workers are gradually being onboarded, but registration is only the first gate. A worker who registers in Maharashtra may work in Karnataka the next month; a worker who registers once may churn out of app work entirely after six months. The challenge is not merely getting names into a database but keeping the database connected to portable benefits that follow the worker across states, platforms, and periods of unemployment. That requires more than survey redesign; it requires the operational rules under India's social security legislation to move from paper to practice.
The legal scaffolding for platform-worker welfare exists in pieces, a social security code that acknowledges gig workers, an accident insurance scheme, a national database, but the connective tissue of portable benefits is still being assembled. Several states have attempted aggregator-specific welfare levies; the centre's survey recalibration suggests a preference for measuring first, then mandating. If Maharashtra's registration base is to become a welfare base, the next step is not more registration drives but benefit portability across state lines and across apps.
What Better Counting Buys
Accurate platform-work data is not an academic luxury. It feeds the base numbers for wage policy, tax collection design, and insurance product structuring. If the labour survey undercounts gig workers by even fifteen percentage points, the resulting policy arithmetic is wrong: contribution rates for an accident cover cannot be set without knowing how many riders and drivers need it, a state welfare levy cannot be calibrated without knowing the size of the base, and the central government's employment claims cannot be benchmarked against credible denominators. The projected 2.3 crore gig workers by 2030 is not a prediction of generosity; it is a planning constraint. The state that cannot count its workers cannot cover them.
The overseas temptation is to import the European model of employment classification, which converts platform workers into employees through judicial tests. India has little to gain from that path. The Indian labour market is not the European labour market; it is larger, more informal, more fragmented, and more dependent on flexible work as a first step out of subsistence. A homegrown framework that protects workers without eliminating the aggregator model is the more plausible route. The survey recalibration is a necessary precondition for that framework: before India legislates how platform workers are treated, it must first prove it knows how many they are, where they work, and how they move.
The rise of platform work also sharpens an older question in Indian policy: whether job creation should be measured by headcount alone or by the durability and protection attached to each job. A gig worker counted by a revised survey is not automatically a formal worker; they may still lack paid leave, employer contributions, or a predictable schedule. The survey revision can settle the counting question, but it will not settle the quality question. That remains a contest between the flexibility platforms need to expand and the security workers need to survive.
Transaction Data Is the Next Frontier
There is a further step the survey revision should contemplate. Platform operators hold transaction-level data on every ride, delivery, and task completed; that data is structured, time-stamped, and auditable in ways household surveys are not. Linking anonymised platform transaction data to the labour survey would correct the undercount in near real time, though it raises legitimate questions about data access and commercial confidentiality. The alternative is to keep asking workers to recall their app earnings over a reference period, which is the statistical equivalent of measuring rainfall with a sieve.
The Maharashtra registration number is less a destination than a baseline. It tells us where the formal trail begins; it tells us nothing yet about how many workers never register, how many drop out, and how many slip back into the unmeasured informal economy when the app deactivates their ID. The labour survey revision that follows is the state's attempt to see its own economy more clearly. That clarity must come before protection, and protection must come before the 2030 projection becomes a census of missed opportunities.
