English only · Odia translation in progress

Digital Transformation and Exponential Compression in Odisha

Research compiled: 2026-03-28 Purpose: Reference document for The Long Arc series — Ch6 (Exponential Arrival), Ch7 (Scorecard), Ch8 (Ninety-Year Question) Scope: Internet penetration, UPI/financial inclusion, Aadhaar, IT sector, startups, AI in governance, digital agriculture, e-governance, the compression paradox

1. Internet and Smartphone Penetration

Pre-JIO Era (~2015)

Before September 2016, India’s internet landscape was defined by scarcity and cost. The country’s overall internet penetration stood at approximately 26% in 2015 (roughly 350 million users in a population of 1.3 billion), but the headline figure masked a brutal rural-urban divide. Urban internet penetration hovered around 55-60%, while rural India — where 69% of the population lived — had penetration rates of approximately 12-15%.

Odisha occupied the lower rungs of this already low national distribution. The state’s telecom infrastructure reflected its broader developmental profile: underserved, under-invested, and structurally disadvantaged. Several factors compounded this:

Cost barrier: Pre-Jio mobile data cost Rs 200-250 per GB. For a household earning Rs 5,000-7,000 per month (the median rural Odisha income range), spending Rs 200 on a single gigabyte of data — enough for perhaps 2-3 hours of video or a week of moderate browsing — was economically irrational. Internet was a luxury, not a utility.

Device barrier: Smartphones capable of running modern apps cost Rs 5,000-8,000 at the low end. Feature phones dominated rural markets. The combination of expensive devices and expensive data meant that for most of Odisha’s 47 million population (2015), the internet was something encountered in district towns at cybercafes, not in daily life.

Infrastructure barrier: Cell tower density in rural Odisha was among India’s lowest. The state’s geography — forested hills in the west and southwest, flood-prone river plains in the east, and dispersed settlement patterns across 51,311 villages — made tower economics unfavorable for private telecom companies. BSNL, the state-owned provider, had the widest rural reach but its 2G/3G infrastructure delivered speeds that made most modern internet applications unusable.

Access patterns: For the minority who did access the internet before 2016, the primary modes were:

  • Cybercafes in district headquarters (Sambalpur, Berhampur, Balasore, Koraput): Rs 20-40/hour, used primarily for government applications, email, and job searching
  • Office/institutional broadband in Bhubaneswar, Cuttack, and Rourkela
  • College campuses (NIT Rourkela, KIIT, Utkal University) where students accessed wifi
  • A thin slice of urban middle-class homes with Airtel/BSNL broadband (1-2 Mbps, Rs 500-1,000/month)

The result: Odisha in 2015 was informationally isolated. A farmer in Kalahandi could not check market prices on a phone. A student in Koraput could not access educational content beyond what their school provided. A migrant worker in Surat could not video-call home. The information asymmetry that defined colonial-era Odisha — where knowledge of markets, prices, regulations, and opportunities was concentrated among middlemen and urban elites — had not been fundamentally disrupted by the first two decades of the internet age.

The JIO Revolution (September 2016)

On September 5, 2016, Reliance Jio launched commercial operations with a promotional offer that was, in retrospect, the most significant infrastructure intervention in Indian daily life since rural electrification: free voice calls forever, and free data for six months. When paid plans began, data was priced at approximately Rs 10 per GB — a 96% reduction from the pre-Jio price.

The mechanism was straightforward but the consequences were not. Mukesh Ambani had spent approximately Rs 1.5 lakh crore ($22 billion) building an entirely new 4G-only network. The strategy was not to compete with existing operators on their terms but to make data so cheap that it became a universal utility, and then to monetize the ecosystem that cheap data enabled. The existing operators — Airtel, Vodafone, Idea, BSNL — had no choice but to match prices, triggering a price war that permanently restructured India’s telecom economics.

For Odisha specifically, Jio’s impact had distinctive characteristics:

Leapfrogging: Odisha largely skipped the desktop/broadband internet phase entirely. The state went from near-zero internet for the majority to smartphone-based mobile internet. This meant the internet arrived through a 5-inch screen, not a computer monitor — a form factor that shaped usage patterns differently from how internet adoption unfolded in urban India or the developed world.

Speed of adoption: Odisha’s wireless+wireline subscriber base grew from 33.55 million (2022-23) to 35.65 million (2024-25) — the state recorded the highest subscriber growth rate in India at 4.24% (Economic Survey 2025-26, Ch. 7 §7.5.3). Internet subscribers in Odisha reached 25.88 million by March 2025 (17.10 million rural + 8.78 million urban), up from 22.58 million in March 2023 (Economic Survey 2025-26, Ch. 7 §7.5.5). Tele-density stood at 80% in March 2025, with Odisha running ahead of the national average in both rural (64.76 vs India 59.06) and urban (143.79 vs India 131.45) segments (Economic Survey 2025-26, Ch. 7 §7.5.4). Wireless remains 95%+ of total internet connections (wired broadband remains negligible outside Bhubaneswar).

Smartphone penetration: As of 2025, 78.1% of Odisha’s population aged 15+ used a mobile phone, and among mobile phone users, 77.1% used a smartphone — implying that roughly 60% of the adult population directly accessed a smartphone, with the remaining ~20% of mobile users still on feature phones (Economic Survey 2025-26, Ch. 6 §6.9.13, Table 6.12). Low-cost handsets from Xiaomi, Realme, Samsung, and Jio’s JioPhone platform underpin this ownership. However, “owned a smartphone” requires qualification: many households share a single device, and in rural areas the smartphone is often controlled by the male head of household, limiting women’s independent access.

Usage patterns: India-wide data (IAMAI/Kantar) suggests the dominant uses of mobile internet in tier-3 and rural areas are: (1) YouTube/video entertainment, (2) WhatsApp messaging and calling, (3) social media (Facebook, Instagram, ShareChat), (4) Google search, and (5) digital payments. Productive uses — online education, job searching, e-governance — are secondary. This pattern likely holds for Odisha, with the addition of Odia-language content consumption on YouTube and ShareChat.

The scale of change was extraordinary. In approximately seven years (2016-2023), Odisha went from a state where the majority of the population had no practical internet access to one where a substantial majority had a smartphone with affordable data. This was not the result of state planning or development policy; it was a private-sector infrastructure play that happened to transform information access as a side effect.

The Digital Divide That Remains

The headline numbers — approximately 25.88 million internet subscribers as of March 2025, 77.1% smartphone share among mobile users (Economic Survey 2025-26, Ch. 7 §7.5.5; Ch. 6 §6.9.13) — tell a story of transformation. But the gaps within that story are as important as the transformation itself.

Geographic exclusion: Approximately 20% of Odisha’s 51,311 villages still lack reliable mobile connectivity. These are disproportionately located in the forested, hilly terrain of western and southern Odisha — the same geography that has been systematically underserved since the British period. The tribal-dominated districts of Malkangiri, Kandhamal, Nabarangpur, Rayagada, and Koraput face the deepest connectivity gaps. The economics are simple: a cell tower serving a village of 200 people in a forested valley will never generate enough revenue to justify its cost to a private operator. The state-owned BSNL, theoretically the provider of last resort, has struggled with maintenance, equipment procurement, and fiber connectivity to its rural towers.

Urban-rural gap: Overall internet subscription density in Odisha reached 58.09 per 100 population by March 2025 (up from 50.97 in March 2023), still trailing the national 68.63 (Economic Survey 2025-26, Ch. 7 §7.5.6). Household-level data tells the more important spatial story: 75.2% of rural households versus 91.4% of urban households had in-premises internet access in 2025, up from 61.9% and 82.6% respectively in 2022-23 (Economic Survey 2025-26, Ch. 6 §6.9.7). The rural-urban gap is narrower than national comparators but still means that the digital economy in Bhubaneswar or Cuttack operates in a fundamentally different informational environment than a village in Kalahandi or Mayurbhanj.

Device quality gap: Smartphone ownership is not a binary. A Rs 6,000 Redmi phone with 2GB RAM running Android 10 on a congested rural 4G tower delivers a qualitatively different internet experience than a Rs 30,000 Samsung on Bhubaneswar’s 4G/5G network. App crashes, slow loading, and storage limitations mean that the “internet” experienced by rural users is a degraded version of what urban users take for granted.

Gender gap: National data (GSMA Mobile Gender Gap Report) consistently shows Indian women are 28-30% less likely than men to own a mobile phone, and the gap is wider in rural areas and eastern states. In rural Odisha, women’s independent smartphone access is significantly lower than men’s, constrained by household economics (the single phone goes to the earning male), cultural norms (surveillance concerns around women’s independent communication), and literacy barriers.

Tribal districts — the extreme case: During the COVID-19 lockdown (2020-2021), surveys revealed the depth of the divide in tribal districts. In Malkangiri, Kandhamal, Nabarangpur, Rayagada, and Deogarh, 80%+ of parents lacked smartphones entirely. When schools shifted to online teaching, these children received nothing — no Zoom classes, no WhatsApp study groups, no recorded lectures. The digital transformation that reached Bhubaneswar’s middle class simply did not reach Malkangiri’s Bonda or Koraput’s Kondh communities.

Bandwidth quality: Even where 4G coverage exists nominally, the quality of service in rural Odisha is often poor. Tower congestion (too many users per tower), inconsistent power supply (towers running on batteries or diesel generators), and poor backhaul connectivity (fiber not reaching the tower) mean that advertised 4G speeds of 20-40 Mbps often deliver 2-5 Mbps in practice. This makes video-based learning difficult, real-time applications unreliable, and the overall experience frustrating enough to limit sustained use.

Language barrier: Most internet content, apps, and interfaces default to English or Hindi. While Odia-language support has improved (Google Translate, Odia keyboard apps, YouTube auto-captions), the internet remains disproportionately accessible to those who can navigate English. For the 27% of Odisha’s population classified as Scheduled Tribes, many of whom speak Gondi, Santhali, Kui, Bonda, or other tribal languages as their first language, with Odia as a second language and no English, the internet’s utility is further constrained.

The overall picture: the JIO revolution delivered information access to a majority of Odisha’s population in record time. But “a majority” is not “all,” and the 30-40% who remain partially or fully excluded are precisely the populations — tribal communities, rural women, western Odisha districts — that were already most disadvantaged. Digital inclusion has not dissolved the old geography of exclusion; it has layered a new form of access on top of the same structural inequalities.

2. UPI and Financial Inclusion

Jan Dhan Yojana (PMJDY) — The Account Infrastructure

The Pradhan Mantri Jan Dhan Yojana, launched on August 28, 2014, was the largest financial inclusion drive in history. Its core mechanism was simple: open a zero-minimum-balance bank account for every unbanked Indian adult, linked to their Aadhaar number, with a RuPay debit card. The target was to eliminate the first barrier to formal financial participation — not having an account.

Odisha’s numbers reflected both the scale of unbanking and the ambition of the program:

  • Estimated 2.5-3 crore (25-30 million) Jan Dhan accounts opened in Odisha
  • Initial zero-balance accounts were extremely high — in many districts, 70-80% of new accounts had no deposits
  • Over time, activation increased as Direct Benefit Transfer (DBT) programs began routing payments through these accounts
  • Banking correspondent (BC) model deployed to reach villages without bank branches — a human intermediary with a biometric device serving as a mini-bank

The significance of PMJDY was not financial but infrastructural. It created the plumbing — the account-Aadhaar-phone linkage (JAM: Jan Dhan-Aadhaar-Mobile) — through which money could flow digitally. The accounts themselves were often dormant. But dormancy was a feature of the old economic reality, not a failure of the account infrastructure. When DBT payments began flowing, the accounts activated.

The banking desert reality: Despite PMJDY, Odisha’s banking infrastructure remains thin. As of 2024, the state has approximately 4,500-5,000 bank branches for a population of 46+ million — roughly one branch per 9,000-10,000 people. In tribal and western districts, the ratio is worse: one branch per 15,000-20,000 people. The banking correspondent model was supposed to fill this gap, but BC agents are often poorly trained, inconsistently available, and dependent on connectivity that fails in remote areas.

UPI Transaction Growth

The Unified Payments Interface, launched by the National Payments Corporation of India (NPCI) in August 2016, was arguably India’s most consequential fintech innovation. By linking bank accounts to phone numbers and enabling instant, free, person-to-person and person-to-merchant transactions via apps like PhonePe, Google Pay, and Paytm, UPI collapsed the cost and friction of digital payments.

India-wide UPI trajectory:

  • 2016: Negligible (launch year)
  • 2018: ~500 million transactions/month
  • 2020: ~2 billion transactions/month
  • 2022: ~7 billion transactions/month
  • 2024: ~13-15 billion transactions/month
  • Transaction value (2024): approximately Rs 18-20 lakh crore/month

Odisha’s position in UPI adoption reflects its economic profile — a medium-adopter state, neither leading (like Karnataka, Maharashtra, or Delhi) nor trailing (like some northeastern states):

Urban adoption (Bhubaneswar, Cuttack, Rourkela, Berhampur): By 2022, UPI acceptance was near-universal among small shopkeepers, auto-rickshaw drivers, vegetable vendors, and street food sellers in these cities. The QR code became as ubiquitous as the cash register. College students, IT workers, and government employees were early adopters who pulled the retail ecosystem toward digital payments.

Semi-urban and rural penetration: Slower but growing. The primary driver in rural areas was not merchant payments but person-to-person transfers, particularly remittances. A Ganjam construction worker in Surat sending Rs 5,000 home via PhonePe replaced the older informal hawala network and the costly bank wire transfer (Rs 25-50 per transaction plus travel to a branch). The receiver in the village might not use UPI for purchases, but they received money digitally.

Merchant adoption curve: The progression in a typical Odisha district town (say, Jajpur or Phulbani) followed a pattern: first, chain stores and pharmacies (2018-2019); then restaurants and mobile recharge shops (2019-2020); then vegetable markets and auto stands (2020-2022, accelerated by COVID cash-avoidance); and finally, weekly haat markets and smaller villages (2022-present, still partial).

Direct Benefit Transfer (DBT)

If UPI was the payment rail, DBT was the cargo it was designed to carry. The concept: instead of routing welfare money through layers of bureaucracy (state → district → block → panchayat → beneficiary), transfer it directly from the central or state government to the beneficiary’s Aadhaar-linked bank account, bypassing every intermediate handler.

Odisha is one of India’s largest DBT recipient states because of the scale of its welfare programs:

KALIA (Krushak Assistance for Livelihood and Income Augmentation): Launched in 2019 by the Naveen Patnaik government as an alternative to PM-KISAN. Rs 10,000/year (Rs 5,000 per kharif and rabi season) to approximately 52 lakh (5.2 million) farmer families, transferred directly to bank accounts. Total outflow: approximately Rs 5,200 crore/year. The scheme covered small and marginal farmers, landless agricultural laborers, and sharecroppers — categories often excluded from central schemes that require land ownership documentation.

BSKY (Biju Swasthya Kalyana Yojana): Odisha’s health insurance scheme covering approximately 3.5 crore (35 million) people — nearly 70% of the state’s population. Hospital payments are routed through Aadhaar-linked accounts, with cashless treatment at empaneled hospitals. Annual coverage: Rs 5 lakh per family (Rs 10 lakh for women members). Claims processing involves Aadhaar authentication and direct settlement.

PDS (Public Distribution System) modernization: Biometric authentication at ration shops (Fair Price Shops) using Aadhaar-linked point-of-sale devices. The intent: eliminate “ghost rations” — rations drawn against fake or duplicate beneficiary entries. Estimated leakage reduction: 5-8% of total PDS distribution value, though precise numbers are contested. Odisha distributes rice at Rs 1/kg to approximately 3.5 crore beneficiaries under the National Food Security Act.

MGNREGA wage payments: Aadhaar-linked payments reduced fake enrollments and duplicate wage claims in the Mahatma Gandhi National Rural Employment Guarantee Scheme. Odisha generates approximately 15-18 crore person-days of MGNREGA employment annually, making it one of the top five states by employment generation. Digital payments reduced (but did not eliminate) the chronic problem of delayed wage payments.

Estimated corruption bypass: National estimates (by NITI Aayog and others) suggest that DBT saved the central government Rs 2.25 lakh crore cumulatively by 2023 through elimination of fake beneficiaries and reduced leakage. Scaling this to Odisha’s share of national welfare spending suggests savings of Rs 2,000-3,000 crore annually, though these figures are approximate and politically charged — the government has incentives to overstate savings, critics to understate them.

Financial Inclusion vs. Economic Inclusion: The Critical Distinction

The metrics of financial inclusion in Odisha look impressive: near-universal bank accounts, growing UPI adoption, DBT reaching millions. But the metrics obscure a fundamental distinction that matters for understanding what digital transformation actually changes.

Financial inclusion means having the infrastructure for financial transactions: a bank account, a UPI-capable phone, an Aadhaar-linked identity. By this measure, Odisha has made extraordinary progress in a decade.

Economic inclusion means having regular income, productive assets, and access to markets that generate wealth. By this measure, the progress is far less dramatic.

The gap is vivid at the individual level:

  • A Kalahandi farmer has a Jan Dhan account, receives KALIA payments via DBT, and can use UPI. His bank balance never exceeds Rs 5,000 because his income from 2 acres of rain-fed paddy minus input costs leaves nothing to save. He is financially included and economically excluded.
  • A Ganjam migrant worker in Surat sends Rs 7,000/month home via PhonePe. The technology of transfer has been transformed. The economic structure that requires him to migrate 1,500 km for work has not changed at all.
  • A Mission Shakti SHG (Self-Help Group) in Jajpur manages its accounts on a smartphone app. The digital efficiency is real. The loan amounts (Rs 50,000-1 lakh) and interest rates (12-24%) reflect the same credit constraints that existed before digitization.

The measurement problem: The government counts Jan Dhan accounts opened, UPI transactions processed, and DBT amounts transferred. These are input metrics — they measure the infrastructure of financial inclusion. What they do not measure is whether this infrastructure translates into economic participation: higher incomes, asset accumulation, reduced poverty, or reduced inequality. The distinction matters because it is entirely possible — and in Odisha’s case, largely true — to digitize the financial infrastructure of poverty without changing the economic structures that produce poverty.

Remittance flow digitization: One concrete change worth noting: the digitization of remittance flows from migrant workers. The traditional pathway for a Surat textile worker to send money to Ganjam involved either:

  • Bank wire transfer: Rs 25-50 fee, required visiting a bank branch, 1-3 day processing
  • Hawala/informal networks: faster but opaque, with a commission of 1-3%
  • Physical cash carried by returning migrants or trusted travelers

PhonePe and Google Pay reduced this to: instant, free, from the worker’s phone to a family member’s phone. The volume is significant: Ganjam district alone receives an estimated Rs 120 crore/month in remittances from Surat, Kerala, and other destinations. The digitization of this flow is genuine economic progress — it reduces transaction costs, increases speed, and creates a digital record. But it does not change the underlying equation: the money is earned elsewhere because there is no comparable employment at home.

3. Aadhaar as State Infrastructure

Enrollment and Coverage

Aadhaar — India’s biometric identity system, administered by the Unique Identification Authority of India (UIDAI) — achieved near-universal adult enrollment in Odisha by the early 2020s, with an estimated 97-98% of adults enrolled by 2024. This makes it the most comprehensive identity infrastructure in the state’s history, exceeding the coverage of any previous government database including the voter roll, ration card system, or land records.

The enrollment process — capturing fingerprints, iris scans, and demographic data — created a biometric database that enables identity verification at scale. When linked to a bank account and a mobile phone number (the JAM trinity), it enables the state to:

  • Verify identity remotely and instantly
  • Route payments directly to verified individuals
  • Authenticate welfare recipients at the point of service delivery
  • Deduplicate databases to identify ghost beneficiaries

For Odisha, where previous welfare delivery systems were plagued by identity fraud (fake ration cards, duplicate MGNREGA enrollments, non-existent beneficiaries on various scheme lists), Aadhaar represented a genuine capability upgrade. The state could, for the first time, be reasonably confident that a payment was reaching a real person.

Welfare Delivery Transformation

The impact of Aadhaar-based authentication on Odisha’s welfare infrastructure has been substantial in specific areas:

PDS reform: Before Aadhaar-linked authentication, Odisha’s Public Distribution System lost an estimated 20-25% of grain to diversion — fake beneficiaries, corrupt shop owners selling rations on the open market, and ghost entries. The introduction of electronic point-of-sale (ePoS) devices with biometric authentication at Fair Price Shops reduced identifiable leakage to an estimated 12-17% (precise figures vary by study and methodology). The improvement was real but not transformative: corruption adapted to the new system through device tampering, authentication workarounds, and front-end compliance with back-end diversion.

MGNREGA: Aadhaar-linked wage payments addressed the specific problem of fake worker enrollments. In districts where local bureaucrats had historically inflated muster rolls with fictitious names and pocketed the wages, biometric verification made this specific form of fraud significantly harder. The state’s estimated savings from reduced MGNREGA fraud are Rs 300-500 crore annually, though the more persistent problem — delayed wage payments caused by administrative processing delays — is not solved by Aadhaar.

BSKY (health insurance): Aadhaar authentication at empaneled hospitals verifies that the patient is a legitimate BSKY beneficiary before treatment is authorized. This reduces hospital fraud (billing for non-existent patients) but does not address the more systemic problem: the shortage of empaneled hospitals in rural and tribal areas, meaning that the insurance card exists but the hospital to use it in does not.

Pension and scholarship disbursements: Old-age pensions (Madhu Babu Pension Yojana), disability pensions, and student scholarships — all now routed through Aadhaar-linked bank accounts, reducing the historical phenomenon of pensions being drawn by dead beneficiaries or scholarships diverted by institutional intermediaries.

The Aadhaar Paradox: Efficient Redistribution Without Productive Transformation

The deeper significance of Aadhaar in Odisha is not what it does but what it reveals about the nature of digital transformation in a low-income state.

Aadhaar makes the state more efficient at delivering welfare. It makes redistribution leaner, faster, and more targeted. These are genuine improvements. But redistribution efficiency and productive economy transformation are fundamentally different capabilities. Aadhaar makes it easier to give people money; it does not create the conditions for them to earn money.

Consider the analogy: a hospital that installs a state-of-the-art patient management system (digital records, automated scheduling, instant lab results) becomes a more efficient hospital. But if the hospital has only 2 doctors for 500 beds, the management system doesn’t treat patients — it just manages the queue more efficiently. The constraint was never the administrative system; it was the productive capacity.

Odisha’s digital welfare infrastructure is analogous. KALIA payments reach farmers faster through Aadhaar-linked accounts. But KALIA doesn’t make the farm more productive, doesn’t connect the farmer to better markets, and doesn’t change the underlying economics of rain-fed single-crop agriculture. The payment arrives efficiently into an economic system that generates insufficient income.

This is what might be called the efficiency trap: the state invests heavily in making redistribution work better (Aadhaar, DBT, digital PDS) while the productive economy — the thing that would reduce the need for redistribution — receives proportionally less transformative investment. The result is a welfare state that runs increasingly well, serving a population that remains structurally dependent on welfare.

Exclusion and Failure Points

Aadhaar’s near-universal coverage coexists with systematic exclusion of specific populations:

Biometric authentication failures: Estimated 5-10% failure rate in rural areas, caused by:

  • Worn fingerprints: manual laborers, agricultural workers, and brick kiln workers whose fingerprint ridges are worn down by physical labor
  • Age-related degradation: elderly beneficiaries whose biometric data has changed since enrollment
  • Skin conditions: common in humid tropical environments
  • Device calibration issues: cheap biometric readers in rural areas with poor maintenance

Network connectivity failures: Aadhaar authentication requires server communication. In areas with poor mobile connectivity — which, as noted above, includes approximately 20% of Odisha’s villages — authentication simply fails. The fallback is supposed to be offline authentication, but in practice, many service points default to “come back later” when the network is down.

Demographic groups disproportionately affected:

  • Elderly: both biometric degradation and difficulty traveling to enrollment/correction centers
  • Disabled persons: physical inability to provide fingerprints or iris scans in some cases
  • Homeless and migrant populations: address requirements for enrollment
  • Tribal communities: language barriers during enrollment, name spelling inconsistencies between Aadhaar and other documents, address changes due to seasonal migration

Single-point-of-failure risk: Aadhaar authentication creates a system dependency that didn’t exist before. If the UIDAI server is down, PDS authentication fails statewide. If a beneficiary’s Aadhaar is deactivated (due to data quality issues), they lose access to all linked services simultaneously. The previous system was corrupt and inefficient, but it was not fragile in this specific way.

The exclusion is not random; it is structurally correlated with poverty and marginalization. The populations most likely to experience Aadhaar authentication failures are the same populations most dependent on the welfare programs that require Aadhaar authentication. This creates a paradox: the system designed to ensure welfare reaches the right people sometimes prevents the most vulnerable people from accessing welfare.

4. IT Sector in Odisha

Infocity, Bhubaneswar

Bhubaneswar’s IT sector is concentrated in and around Infocity (also known as Infovalley), located in Chandrasekharpur, established in the late 1990s and early 2000s as part of a national push to create IT hubs beyond Bangalore, Hyderabad, and Pune. The Software Technology Parks of India (STPI) Bhubaneswar center facilitated IT exports through infrastructure, customs exemptions, and single-window clearances.

Major employers present:

  • TCS (Tata Consultancy Services): The largest IT employer in Bhubaneswar, with an estimated 6,000-8,000 employees. TCS Bhubaneswar primarily operates as a delivery center for clients based elsewhere — the code is written in Bhubaneswar, but the client relationship, architecture decisions, and business development happen in Mumbai, Chennai, or the US.
  • Infosys: Estimated 2,000-4,000 employees in its Bhubaneswar development center. Similar delivery center model.
  • Wipro: Estimated 4,000-5,000 employees. Multiple offices in and around Infocity.
  • LTIMindtree (formerly Mindtree + L&T Infotech): Significant presence, primarily in BPO/BPM operations.
  • Tech Mahindra: Growing presence, focused on telecom and digital transformation services.
  • DXC Technology: Major employer, primarily in infrastructure management and BPO.
  • Other notable names: HCL Technologies, Cognizant (smaller presence), multiple mid-tier IT services companies, and a growing number of captive centers (GCCs) for multinational corporations.

Total employment: Direct employment in Odisha’s IT and IT-enabled services stood at approximately 60,000 in 2023-24, with most of this concentrated in Bhubaneswar; the state targets growth to over 2 lakh by 2036 under the Odisha IT Policy 2025 (Economic Survey 2025-26, Ch. 6 §6.8.1). This makes IT among the largest private-sector organized employer groups in the city, though the number is dwarfed by government employment (Bhubaneswar being the state capital with a dense concentration of government offices).

STPI Bhubaneswar: The Software Technology Parks of India center in Bhubaneswar has been operational since the early 2000s, facilitating IT exports through infrastructure provision, data communication services, and single-window regulatory clearance. STPI units in Odisha export primarily to the US and Europe, with services ranging from application development and maintenance to business process outsourcing.

Software Exports

Odisha’s IT export value: Software, electronics, and semiconductor exports grew from Rs 4,500 crore in 2019-20 to nearly Rs 12,900 crore in 2023-24 — roughly a 2.8x increase in four years, or a CAGR of around 30% (Economic Survey 2025-26, Ch. 6 §6.8.1). The base remains small compared to tier-1 IT hubs but the trajectory is steep, and the state has articulated higher aspirational targets (Rs 10,000+ crore under the Odisha IT Policy 2025).

Comparison with major IT hubs:

  • Bangalore/Karnataka: Rs 7+ lakh crore ($85+ billion)
  • Hyderabad/Telangana: Rs 2.5+ lakh crore ($30+ billion)
  • Pune/Maharashtra (IT component): Rs 1.5+ lakh crore ($18+ billion)
  • Chennai/Tamil Nadu: Rs 1.8+ lakh crore ($22+ billion)
  • Bhubaneswar/Odisha: Rs 5,000-8,000 crore ($0.6-1 billion)

Odisha’s share: Less than 1% of India’s total IT exports (approximately $245 billion in FY2024). The growth trajectory is positive, but the gap with leading states is not closing in absolute terms. Bangalore adds more IT export revenue in a single quarter than Bhubaneswar generates in an entire year.

Why Bhubaneswar Is Not and Cannot Easily Become Bangalore

The absence of an IT ecosystem — as opposed to IT offices — in Bhubaneswar is not an accident. It reflects the absence of specific ecosystem ingredients that took decades to assemble in successful IT hubs:

No anchor company that started here: Every major IT hub has anchor companies that were founded there and grew organically. Bangalore has Infosys, Wipro, and a cluster of product companies. Hyderabad has its historical association with DRDL, ECIL, and later attracted Microsoft and Google to establish major campuses. Bhubaneswar has no IT company of national significance that was founded in the city. Every major IT employer in Bhubaneswar is a branch office — a delivery center for a company headquartered elsewhere.

This matters because anchor companies create the ecosystem: they train talent that then starts new companies, they attract other companies to locate nearby, they create a managerial class with the experience and network to fund and mentor startups. A delivery center does none of this. It employs people; it doesn’t create an ecosystem.

No VC/PE ecosystem: Venture capital is geographically clustered. India’s VC funds are concentrated in Bangalore (Sequoia, Accel, Matrix, Elevation), Mumbai (debt funds, PE), and Delhi-NCR (early-stage funds). There are zero VC funds of national significance based in Odisha. This means that a Bhubaneswar startup must travel to Bangalore or Mumbai to fundraise — a friction that doesn’t exist for a Bangalore startup. VC investment follows pattern recognition: investors fund what they know, in ecosystems they understand, with founders they can meet regularly.

Limited tier-1 talent retention: NIT Rourkela produces approximately 1,000 engineering graduates annually, with placement rates of 80-90% at top companies. An estimated 85-90% of these graduates leave Odisha — for Bangalore, Hyderabad, Pune, Delhi-NCR, or the US. The same pattern holds for KIIT, IIIT Bhubaneswar, and other engineering colleges: the best students leave. The talent pipeline exists (Odisha produces a significant volume of engineers), but the pipeline flows out of the state.

Quality of life gap: IT professionals choosing between Bhubaneswar and Bangalore consider: international flight connectivity (Bhubaneswar: zero direct international flights; Bangalore: 50+ international destinations), cultural and entertainment infrastructure, restaurant and nightlife scene, school quality for children, spouse employment opportunities, and general urban amenity levels. On every dimension, Bhubaneswar trails significantly.

The fundamental structural difference: The IT companies in Bangalore make products and decisions. The IT offices in Bhubaneswar execute tasks decided elsewhere. This is the service delivery center vs. innovation hub distinction, and it mirrors a pattern seen across Odisha’s economic history: the state provides the inputs (labor, in this case skilled engineers), while the value-added activity (product development, business strategy, client relationships) happens elsewhere. The parallels with iron ore extraction — Odisha mines the ore, other states make the steel — are not coincidental. They reflect the same structural position in the value chain.

5. Startup Ecosystem

Startup Odisha

The Odisha government launched its Startup Odisha initiative in 2016, recognizing the national trend toward entrepreneurship-led economic development. The program offers:

  • Seed funding (up to Rs 10-20 lakh in early stages)
  • Incubation space and infrastructure
  • Mentorship programs connecting startups with industry veterans
  • Tax incentives and regulatory simplification
  • Participation in national and international startup events

By 2024, Startup Odisha reported 2,000-3,000+ registered startups. The number, however, requires qualification: “registered startup” includes very early-stage ventures, many of which have no revenue, no product, and limited viability. The attrition rate for registered startups nationally is estimated at 80-90% within 5 years.

Comparison: T-Hub in Hyderabad — often cited as India’s most successful government-backed startup incubator — houses 800+ startups that have collectively raised $800+ million in venture funding. Startup India’s overall numbers show Karnataka (Bangalore) with 12,000+ DPIIT-recognized startups, Maharashtra with 10,000+, and Odisha in the middle ranks with its 2,000-3,000. The gap is not just numerical but qualitative: Bangalore startups include unicorns (companies valued at $1 billion+), while Odisha’s most successful startups remain at the early or growth stage.

Incubation Centers

KIIT TBI (Technology Business Incubator): The largest incubation center in eastern India, located at KIIT University, Bhubaneswar. Offers physical infrastructure, mentorship, access to KIIT’s student pool, and connections to industry. Has incubated several hundred startups across sectors including healthtech, edtech, agritech, and fintech.

IIIT Bhubaneswar: Runs an incubator focused on IT/software startups, leveraging its academic expertise in computer science and AI.

O-Hub: Located at IDCO Tower, Bhubaneswar, O-Hub is a state government-backed incubation facility providing co-working space, mentorship, and access to the Startup Odisha ecosystem.

Other incubators: CII’s Odisha chapter, NASSCOM’s 10,000 Startups program (with Bhubaneswar touchpoints), and several private co-working spaces that function as informal incubation environments.

Why No Unicorn

Odisha’s startup ecosystem has produced no unicorn and very few companies that have achieved national prominence. The reasons mirror and compound the IT sector’s structural limitations:

No experienced founder recycling: In Bangalore, successful founders from one generation (Infosys, Wipro, Flipkart) become angel investors and mentors for the next generation. This creates an experience flywheel: each generation of startups benefits from the knowledge, networks, and capital of the previous generation. Odisha has no equivalent. There are no Odia-origin tech billionaires based in Bhubaneswar who are recycling their wealth and networks into the local ecosystem. (There are successful Odia-origin tech professionals globally, but they are based in Bangalore, San Francisco, or Singapore — not Bhubaneswar.)

Limited domestic market demand: A startup’s first customers are usually local. Bangalore’s domestic market includes hundreds of IT companies, thousands of startups, and a wealthy, tech-savvy consumer base. Bhubaneswar’s domestic market is the state government, a few IT companies, and a consumer base with significantly lower purchasing power. An edtech startup in Bangalore can sell to corporate training departments of 50 IT companies within the city. An edtech startup in Bhubaneswar has a much smaller addressable market locally and must compete nationally from day one — against companies based in ecosystems with all the advantages described above.

VC distance: The physical and psychological distance between Bhubaneswar and India’s VC centers (Bangalore, Mumbai, Delhi) is a real barrier. VCs invest in people they can meet regularly, in ecosystems they understand. A Bhubaneswar founder pitching in Bangalore is an outsider — unfamiliar ecosystem, unfamiliar networks, unfamiliar references. This is not insurmountable (remote investing increased post-COVID), but it remains a significant friction.

Brain drain of entrepreneurial talent: The most ambitious, risk-tolerant, and well-networked individuals — precisely the people most likely to start successful companies — are also the most likely to leave Odisha for larger ecosystems. A potential founder from NIT Rourkela or KIIT who goes to Bangalore for their first job accumulates networks, experience, and pattern recognition in that ecosystem. When they start a company, they start it in Bangalore, not Bhubaneswar. The startup ecosystem loses its best potential participants to the same talent drain that affects IT employment.

Regulatory and infrastructure friction: Starting and running a business in Odisha involves regulatory compliance (GST, labor laws, municipal permissions) that, while nationally standardized, is administered by a state bureaucracy that ranks lower than leading states in ease of doing business indices. Infrastructure — reliable power, quality office space, logistics connectivity — is adequate in Bhubaneswar’s IT zones but falls short of what founders can access in Bangalore’s Koramangala or Hyderabad’s HITEC City.

6. Digital Literacy and Education

Government Programs

PMGDISHA (Pradhan Mantri Gramin Digital Saksharta Abhiyan): The national digital literacy program aimed at making one person per household in rural India digitally literate. In Odisha, the program trained millions, but “digital literacy” as defined by the program is minimal — operating a smartphone, using basic internet functions, making digital payments. It does not extend to the skills needed for productive digital work: coding, data analysis, digital marketing, or even effective online research.

Smart classroom initiatives: The state government has deployed digital boards (interactive screens or projectors connected to content servers) in a subset of government schools. Programs include BOSS (Biju Odisha Smart Schooling) and integration with central programs like PM eVIDYA and DIKSHA (Digital Infrastructure for Knowledge Sharing). The rollout has been partial — covering a minority of the state’s 60,000+ schools — and effectiveness is constrained by teacher training (many teachers are unfamiliar with the technology), maintenance (broken equipment without repair budgets), and content quality (generic national content not always relevant to Odisha’s curriculum).

Mo School (My School): A state initiative allowing alumni to contribute financially to their government schools, with matching government funds. Includes digital infrastructure components — computers, internet connectivity, digital content — in schools that receive alumni support. A creative model that has generated significant engagement but structurally favors schools in wealthier areas with successful alumni.

COVID Digital Education Crisis (2020-2021)

The COVID-19 lockdown, beginning March 2020, exposed the full depth of Odisha’s digital education divide. When schools closed nationwide and education shifted online, the assumption was that digital infrastructure could substitute for physical infrastructure. For Odisha’s government school students, this assumption was catastrophically wrong.

Key data points (from ASPIRE Survey 2021, Down To Earth, and OdishaTV reporting):

  • 80%+ of Odisha’s government school students received NO instructional resources of any kind during the lockdown
  • Only 8.8% of enrolled children aged 7-14 received learning materials on a weekly basis
  • In tribal districts (Malkangiri, Kandhamal, Nabarangpur, Rayagada, Deogarh), 80%+ of parents lacked smartphones entirely
  • Even where smartphones existed, connectivity was often too poor for video classes
  • Radio and TV-based alternatives (Doordarshan educational programming, All India Radio) had limited reach and lower engagement
  • Private school students in Bhubaneswar, Cuttack, and other cities continued classes on Zoom, Google Meet, and similar platforms with minimal disruption

The learning loss: While precise quantification is difficult, multiple studies (ASER, Azim Premji Foundation) documented significant learning loss nationally, with the poorest states and communities worst affected. For Odisha, the estimated learning loss for government school students — already below grade level before COVID — was equivalent to 1-2 years of instruction. The gap between private school students (who continued learning digitally) and government school students (who received nothing) widened dramatically during the 18-month school closure period.

The structural revelation: COVID didn’t create the digital education divide; it revealed it. The divide was always present — private schools had computer labs and English-medium instruction; government schools had overcrowded classrooms and under-trained teachers. What the pandemic revealed was that the digital transformation celebrated in other contexts (UPI, JIO, Aadhaar) had not meaningfully reached the education system. A student in Malkangiri had the same smartphone access to YouTube as a student in Bhubaneswar — in theory. In practice, one had a Rs 30,000 tablet, a home wifi connection, and parents who could supervise online learning; the other had no device, no connectivity, and parents who were illiterate.

The Quality Paradox

The most analytically interesting aspect of digital education in Odisha is not the gap but the paradox:

Access to information has been democratized; access to knowledge infrastructure has not.

A motivated, English-literate student in Koraput can now access MIT OpenCourseWare, Khan Academy, Stanford Engineering Everywhere, and thousands of hours of world-class educational content on YouTube — all for free, on a Rs 8,000 smartphone. This would have been inconceivable in 2015.

But:

  • That student’s local college has infrastructure, faculty, and pedagogy quality that is orders of magnitude below the content available on their phone
  • They have no laboratory to practice what they watch
  • They have no peer group studying the same material
  • They have no mentor to guide their learning path
  • They have no credential from YouTube watching that the job market recognizes
  • They have no network of alumni from YouTube that can provide career opportunities

The content is available; the infrastructure to act on it is not. YouTube can teach you how a transistor works; it cannot give you a lab, a study group, a degree, a reference letter, or a job. For the small percentage of students who are extraordinary self-learners, the internet is genuinely transformative. For the median student, it is entertainment with educational characteristics — helpful at the margin but not a substitute for institutional quality.

The language dimension: Most high-quality educational content online is in English or Hindi. Odia-language educational content exists but is thin compared to Hindi or Tamil. A tribal student whose first language is Kui, second language is Odia, and who has no English, faces a triple barrier: device access, connectivity, and language.

7. AI in Odisha Governance

OSDMA and Disaster Prediction

The Odisha State Disaster Management Authority (OSDMA), established after the devastating 1999 super cyclone that killed approximately 10,000 people, is arguably Odisha’s most capable government institution. Its track record — reducing cyclone deaths from 10,000 (1999) to 64 (Cyclone Fani, 2019) despite Fani being a stronger storm — represents genuine institutional excellence, built over two decades of systematic investment in early warning, evacuation, and response capabilities.

AI/ML integration with OSDMA builds on this institutional foundation:

Cyclone trajectory prediction: Machine learning models that process satellite imagery, sea surface temperatures, atmospheric data, and historical cyclone paths to improve trajectory and intensity forecasts. These supplement IMD (India Meteorological Department) forecasts, providing OSDMA with additional data points for evacuation decision-making. The improvement is incremental but valuable: hours of additional lead time in evacuation decisions save lives.

Flood monitoring: AI-enhanced analysis of satellite imagery (Sentinel, INSAT) to map flood extent in real-time during the monsoon season. Odisha’s river systems (Mahanadi, Brahmani, Baitarani, Subarnarekha) flood annually, affecting millions. Real-time flood mapping enables more targeted relief distribution and evacuation orders.

Early warning dissemination: Integration with mobile alert systems to push cyclone and flood warnings to smartphones in affected areas. The combination of near-universal mobile phones and AI-optimized warning timing creates an early warning chain that reaches from satellite to smartphone in approximately 30 minutes.

Evacuation route optimization: AI models that factor in road conditions, bridge status, population density, and shelter capacity to generate optimized evacuation routes for specific cyclone scenarios.

The key insight: AI amplifies OSDMA’s already-excellent institutional capacity. OSDMA had the trained personnel, the institutional memory, the political authority, and the operational protocols. AI makes these work faster and with better data. This is the positive case for AI in governance.

Agricultural AI

AI applications in Odisha agriculture are primarily at pilot stage:

Crop advisory systems: AI-driven platforms that combine weather data, soil information, and crop models to recommend planting dates, seed varieties, fertilizer quantities, and irrigation timing. These exist as government pilots and private-sector apps (DeHaat, Ninjacart’s farm advisory) but penetration among Odisha’s smallholder farmers is extremely limited.

Pest surveillance: AI-powered image recognition apps where farmers photograph affected plants and receive pest identification and treatment recommendations. Functional in concept; limited in practice by smartphone quality (camera resolution on Rs 6,000 phones), connectivity (uploading images in low-bandwidth areas), and the fundamental constraint that knowing what pest is attacking your crop doesn’t help if you can’t afford the recommended pesticide.

Market price information: Real-time price data from mandis (agricultural markets) available via apps and SMS. Reduces information asymmetry between farmers and traders. But market price information is only useful if the farmer has: (a) the transport to reach the better-priced mandi, (b) the storage to hold produce until prices improve, and (c) the scale to make the transaction cost worthwhile.

Weather prediction for farming: Hyper-local weather forecasts combining IMD data with local sensor networks and ML models. More accurate than traditional radio forecasts. Useful for timing planting, harvesting, and fertilizer application. But rain-fed farming in western Odisha is fundamentally constrained by the monsoon’s binary character: either enough rain or not. A better forecast of “no rain this week” doesn’t create water.

The core limitation: AI advisory is only useful if the farmer has the resources to act on advice. An AI system that correctly identifies that a farmer should switch from paddy to maize, apply specific micronutrients, and irrigate at precise intervals is useless if the farmer has no irrigation, no access to micronutrient fertilizers, and no market assurance for maize. The bottleneck is not information; it is resources, infrastructure, and market access.

Healthcare AI

Telemedicine platforms: Video consultation platforms connecting patients in primary health centers (PHCs) and community health centers (CHCs) in rural Odisha with specialist doctors in Bhubaneswar, Cuttack, or other urban hospitals. COVID accelerated deployment. Addresses the critical shortage: Odisha has approximately 1 doctor per 2,000 people (national average: 1 per 1,000; WHO recommendation: 1 per 1,000), with distribution heavily skewed toward urban areas. Many rural PHCs operate without any doctor.

AI diagnostic tools: Pilot deployments of AI-powered diagnostic assistance in district hospitals — automated interpretation of X-rays, analysis of pathology slides, screening for diabetic retinopathy from retinal scans. These are genuine advances in capability, particularly for districts where specialist radiologists or pathologists are absent.

COVID applications: Oxygen monitoring dashboards tracking availability across hospitals, bed availability tracking (integrated with the national system), and digital contact tracing. These demonstrated the state’s ability to deploy digital tools rapidly under crisis conditions.

The substitution fallacy: AI in healthcare is sometimes presented as a solution to India’s doctor shortage. It is not. AI can assist diagnosis (flagging an X-ray abnormality for a doctor to confirm), but it cannot treat a patient. A PHC in Nuapada district that has no doctor cannot function because an AI can read X-rays. The patient still needs a human being to examine them, diagnose them, and treat them. AI improves the productivity of existing doctors; it does not create doctors where none exist.

E-Governance AI Applications

The state has deployed or piloted several AI applications in governance:

  • Multi-lingual chatbots: For transport services (vehicle registration, driving license) and e-District services (certificate applications, land records), chatbots handle routine queries in Odia, Hindi, and English. Reduces the need for citizens to visit offices for simple information requests.
  • Face analytics for e-Prisons: Facial recognition technology deployed in the state’s prison management system for identity verification.
  • Text summarization for cybersecurity: AI tools that process and summarize intelligence reports for the state’s cybersecurity operations.
  • Blockchain for land records (Bhulekh): Integration of blockchain technology with Odisha’s online land records system to create tamper-proof ownership records. Land disputes are endemic in Odisha (and India broadly), and blockchain-based records aim to reduce title fraud.

The Amplification Principle

The pattern across all AI applications in Odisha governance can be stated as a principle:

AI amplifies existing institutional capacity. It does not create institutional capacity where none exists.

This has profound implications:

  • OSDMA + AI = better OSDMA. Because OSDMA already had excellent people, processes, and institutional authority, AI made it better.
  • A district health system with no doctors + AI diagnostic tools = still no functional healthcare. The AI has nothing to amplify.
  • Districts with strong governance (typically the coastal districts with better administrative traditions) use AI tools effectively. Districts with weak governance (typically the western tribal districts with chronically understaffed, undertrained, and under-resourced administrations) see little benefit from the same tools.
  • The technology is the same everywhere. The institutional capacity to use it varies enormously. AI therefore tends to widen the governance quality gap between well-governed and poorly-governed districts.

This mirrors the broader digital transformation pattern: technology is a force multiplier, not a force creator. Multiplying zero by any number still gives zero.

8. Gig Economy and Digital Work

Platform Economy in Odisha

The platform-mediated gig economy — Swiggy (food delivery), Zomato (food delivery), Amazon/Flipkart (e-commerce delivery), Ola/Uber (ride-hailing), Urban Company (home services) — is active in Bhubaneswar, Cuttack, Rourkela, Berhampur, and Sambalpur. Smaller platforms like Rapido (bike taxi) and Dunzo (hyperlocal delivery) have more limited presence.

Estimated gig workers in Odisha: No reliable state-level data exists. Nationally, NITI Aayog estimated 7.7 million gig workers in India (2020-2021), projected to reach 23.5 million by 2030. Odisha’s share, estimated proportionally and adjusted for lower urbanization, might be in the range of 50,000-150,000 — but this is highly speculative. The majority are concentrated in Bhubaneswar.

Worker profile: Primarily urban, primarily male, primarily young (18-30). Delivery workers for Swiggy/Zomato/Amazon typically earn Rs 12,000-18,000/month in Bhubaneswar (lower than Bangalore or Hyderabad rates of Rs 18,000-25,000 for comparable hours). Ride-hailing drivers earn Rs 15,000-25,000/month depending on hours.

The gig work proposition: For a young man from a rural Odisha district with 10+2 education and no specialized skills, the choices in Bhubaneswar are: (a) gig delivery work at Rs 12,000-15,000/month, (b) retail employment at Rs 8,000-12,000/month, (c) construction labor at Rs 10,000-15,000/month, or (d) migration to Surat/Kerala/Bangalore for Rs 15,000-25,000/month. Gig work offers the advantage of staying in Odisha (near family, lower cost of living) with the disadvantage of lower earnings and no benefits.

Remote and Freelance Work

Post-COVID remote work: The pandemic normalized remote work for IT and knowledge workers. Some Odia IT professionals working in Bangalore or Hyderabad returned to Bhubaneswar temporarily (or permanently) while working remotely for their employers. This created a small but visible cohort of well-paid remote workers in Bhubaneswar — people earning Bangalore salaries while living in Bhubaneswar’s lower-cost environment.

However, this remains a thin slice of the population — perhaps a few thousand individuals. It does not constitute a structural change in Odisha’s employment patterns. And the trend has partially reversed as many companies mandate return-to-office policies.

Freelancing platforms: Odisha’s participation in global freelancing platforms (Upwork, Fiverr, Freelancer.com) is limited. National data suggests India is the second-largest source of freelancers on these platforms (after the US), but the supply is concentrated in tech hubs (Bangalore, Delhi, Hyderabad) where English fluency, technical skills, and platform familiarity are higher. Odia freelancers exist but are not a statistically significant share of the national freelance workforce.

YouTube and content creation in Odia: A growing niche that represents one of the more organically digital-native economic activities in Odisha. Odia-language YouTube channels covering comedy, cooking, vlogging, news commentary, devotional content, and music have built audiences ranging from thousands to millions. The top Odia YouTube channels generate significant revenue from advertising, brand partnerships, and merchandise. This is a genuinely new economic activity enabled by digital platforms — it did not and could not exist before 2016.

Digital marketing and social media management: A small but growing segment of young professionals in Bhubaneswar offering social media management, content creation, and digital marketing services as micro-enterprises or freelance work. These are typically self-taught through YouTube tutorials and online courses, serving local businesses (restaurants, real estate, education) that are beginning to invest in digital marketing.

Gig Work: Employment or Exploitation?

The gig economy in Odisha raises the same questions it raises globally, with an additional Odisha-specific dimension:

Low wages with no safety net: Gig workers are classified as independent contractors, not employees. They receive no provident fund, no health insurance (though the government’s BSKY provides a partial substitute), no paid leave, and no job security. A Swiggy delivery person in Bhubaneswar who is injured on the job has no worker’s compensation and may lose income during recovery.

Partial substitute for migration: For some young men who would otherwise migrate to Surat or Kerala, gig work in Bhubaneswar or Cuttack offers an alternative — lower earning but closer to home. Whether this is net positive (less family separation, lower living costs) or net negative (lower lifetime earnings, no skill development, no benefits) depends on the individual case.

The wage comparison problem: A Ganjam construction worker in Surat earns Rs 18,000-22,000/month with overtime, in an environment where employer-provided housing (however basic) reduces living costs. A Zomato delivery worker in Bhubaneswar earns Rs 12,000-15,000/month, pays his own rent and expenses, and wears out his motorcycle. The gig economy in Bhubaneswar often pays less than physical labor in Gujarat, making it a questionable economic alternative to migration.

Not a structural solution: The gig economy cannot solve Odisha’s employment problem. The state needs to employ a working-age population of approximately 25-30 million people in productive work. The gig economy in Bhubaneswar provides work for perhaps 50,000-100,000 people, primarily in one city, in low-wage, precarious positions. It is a phenomenon worth studying but not a development strategy.

9. The Compression Paradox — Data Points and Analysis

What “Exponential” Means Materially

The word “exponential” is overused in technology discourse but precisely accurate when applied to Odisha’s informational transformation over three generations:

Literacy trajectory:

  • 1936 (formation of Odisha province): estimated 15.8% literacy rate
  • 1951 (first post-independence census): 15.8%
  • 1971: 26.2%
  • 1991: 49.1%
  • 2011: 72.9%
  • 2026 (estimated): approaching 80-85%, with near-universal basic literacy among those under 40

The acceleration is dramatic: it took 35 years (1936-1971) to increase literacy from 16% to 26% — ten percentage points. It took the next 40 years (1971-2011) to add 47 percentage points. The first generation after independence barely moved the needle; the third generation approached universality.

Communication technology:

  • 1936: No radio, no telephone, no telegraph outside colonial administrative offices. Communication between a Koraput village and Cuttack took days by letter, weeks by person.
  • 1970s: Radio reaches some households (All India Radio). Telephone lines exist in district headquarters. The vast majority of the population has never used a phone.
  • 1990s: Landline telephones reach middle-class urban homes. STD/ISD booths provide public access. Television arrives (Doordarshan). Rural areas remain disconnected.
  • 2005: Mobile phones begin penetrating urban areas. Feature phones cost Rs 2,000-3,000.
  • 2016: JIO launches. Smartphones become affordable.
  • 2024: An estimated 55-65% of adults own a smartphone. Video calling is routine. WhatsApp is the dominant communication platform across urban and rural areas.

The communication leap in three generations — from zero electronic communication to near-universal smartphone access — is historically unprecedented. No previous technology transition (printing press, telegraph, radio, television) compressed this much change into this few decades.

Information access:

  • 1936: One university-level institution in the territory (Ravenshaw College, Cuttack). Access to information was through a tiny number of libraries, newspapers (Odia-language press existed but with limited circulation), and personal networks.
  • 1960s-1970s: University system established (Utkal, Berhampur, Sambalpur). Government schools expanded. But textbooks were scarce, libraries inadequate, and the information universe available to a student was bounded by what existed physically within reach.
  • 1990s: Private engineering and medical colleges begin. Television brings visual information. But information access remains constrained by physical proximity and purchasing power.
  • 2016+: A student anywhere in Odisha with a smartphone can access YouTube, Wikipedia, Khan Academy, MIT OCW, Google Scholar, news from any source in any language. The entire digital repository of human knowledge is theoretically accessible.

The informational leap in three generations exceeds the informational change of the previous thirty generations. A person born in 1940 in a Kalahandi village and their grandchild born in 2010 in the same village inhabit radically different information environments, even if they inhabit the same economic environment.

Specific Compression Images

The paradox of exponential digital change layered on top of structural economic stasis produces images that capture the compression:

The oxen-UPI farmer: A farmer in Bargarh district uses PhonePe to receive KALIA payments and pay his electricity bill. He plows his field with a pair of oxen and a wooden plow — equipment and techniques unchanged in two hundred years. Both facts are simultaneously true about his daily life. He is digitally modern and agriculturally pre-modern. The phone connects him to 2024; the plow connects him to 1824. He lives in both centuries.

The tribal YouTuber: A Kondh student in a village outside Koraput watches MIT lectures on semiconductor physics on a Rs 8,000 Redmi phone, sitting under a tree because his house has intermittent electricity and the tree gets better mobile signal. The village has no paved road to the nearest district hospital (45 minutes by motorbike on a dirt track). He can access the same lecture as a student at IIT Delhi. He cannot access the same lab, library, teacher, peer group, or job market. The information gap has collapsed. The opportunity gap has not.

The digital mason: A construction worker from Ganjam district, working on a high-rise in Surat, sends Rs 7,000 to his wife via PhonePe at 11 PM after his shift. The money arrives instantly in their village account. His wife will use part of it to buy rice and part to pay their daughter’s school fees. There is no factory, no processing plant, no organized employer within 50 kilometers of their village. Digital connectivity has been achieved. Economic connectivity — the kind that would eliminate the need for his 1,500-kilometer migration — has not.

The OSDMA-WhatsApp chain: When Cyclone Fani approached Odisha’s coast in May 2019, OSDMA’s warning reached every smartphone in the target evacuation zone within 30 minutes through a cascade of official alerts, WhatsApp forwards, and news notifications. This contributed to evacuating 1.2 million people in 48 hours, resulting in 64 deaths — compared to 10,000 in a comparable storm 20 years earlier. The digital communication chain worked brilliantly for disaster response. The same digital infrastructure has not been able to deliver agricultural extension advice — what to plant, when to spray, where to sell — to the majority of the same farmers. One-way emergency information flows; multi-step, context-specific advisory information does not.

The SHG on WhatsApp: Mission Shakti, Odisha’s Self-Help Group program, comprises over 6 lakh SHGs with approximately 70 lakh women members as of 2024-25 (Economic Survey 2025-26, Ch. 9 §9.5.5). These groups increasingly coordinate via WhatsApp — scheduling meetings, sharing information, even maintaining basic accounts. The digital coordination is real and useful. Under the SHG bank linkage program, 3.52 lakh SHGs received Rs 17,500 crore in finance in 2024-25, and the Mission Shakti Loan Scheme offers interest-free loans up to Rs 10 lakh (Economic Survey 2025-26, Ch. 9 §9.5.11). Yet the same women often have no land titles (land is registered in their husband’s name) and face market-access constraints that no app dissolves. WhatsApp coordinates their marginalization more efficiently; it does not end it.

The graduate with zero employment options: A young man from Nuapada district has a bachelor’s degree (BA from a local college), a smartphone, access to Naukri.com and LinkedIn, and he can see every job posting in India. He applies to dozens. He is qualified for almost none (his education quality does not match what employers seek). He can see the job market perfectly; he cannot participate in it. Information about opportunity is not the same as access to opportunity.

What Digital Changes

Acknowledging what has genuinely changed is important. The digital transformation is not illusory:

Information asymmetry: collapsed. This is the most consequential change. Previously, a farmer selling rice in a mandi had no way to know prices at other mandis. A middleman could quote below-market prices because the farmer had no alternative information. Now, the farmer can check prices on the AGMARKNET app or simply Google “rice price Sambalpur mandi today.” The information asymmetry that enabled exploitation by middlemen, traders, landlords, and bureaucrats has been significantly reduced.

The same applies to government schemes: a villager can check if they are listed as a KALIA beneficiary online, verify their ration card status, or check their MGNREGA wage payment status. This doesn’t eliminate corruption, but it makes corruption more visible and therefore more contestable.

Awareness and aspiration: massively expanded. Through YouTube, Instagram, and WhatsApp, rural Odias now see how other places work. A young person in Kendrapara can see what Bangalore looks like, what jobs pay there, what students in other states are studying. This expanded awareness creates aspiration (wanting more) and comparison (seeing what’s missing). Whether expanded aspiration without expanded opportunity is net positive or negative is an open question — it may contribute to out-migration, frustration, or political mobilization.

Some genuinely new economic activity: The gig economy, YouTube content creation, digital marketing, online tutoring — these are economic activities that did not exist before 2016 and cannot exist without digital infrastructure. They are small relative to the total economy but they are real, and for specific individuals, they represent genuine new pathways.

Government efficiency: DBT, Aadhaar-authenticated PDS, e-governance portals, digital land records — these have made the state government more efficient at service delivery. Not transformatively better, but measurably better. Wait times have reduced, some corruption channels have been blocked, and citizen access to information has improved.

Social coordination: WhatsApp enables organizing — for SHGs, for protests, for community events, for disaster response. The coordination cost of collective action has dropped dramatically. Whether this leads to productive collective action or merely more efficient social media outrage depends on institutional context.

What Digital Does Not Change

The harder truth: the things that digital transformation does NOT change are precisely the things that determine whether Odisha’s economy and society fundamentally transform:

Productive assets: A smartphone does not create a factory, modernize a farm, build a processing plant, or establish a research laboratory. The physical infrastructure of production — the capital stock that generates employment and income — is not affected by digital connectivity. Odisha’s industrial capacity, its agricultural productivity, its manufacturing base — these require physical investment, not digital access. The phone connects you to information about what could exist; it does not create what needs to exist.

Institutional capacity: WhatsApp does not build a functioning industrial development agency. An AI chatbot does not create a competent district administration. Digital tools can make existing institutions work better, but they cannot substitute for institutions that don’t exist or don’t function. Odisha’s governance gaps — understaffed block offices, absent PHC doctors, unfilled engineering positions in the Public Works Department — are human resource and institutional design problems, not technology problems.

Structural employment: You cannot employ 60% of the population through apps. Odisha needs approximately 15-20 million new productive jobs for its working-age population. The entire gig economy, IT sector, and digital services sector combined employs perhaps 100,000-200,000 people in the state. The employment problem requires manufacturing, agro-processing, mining value-addition, and organized services at a scale that digital platforms do not create.

Power structures: Digital tools layer on top of existing caste, class, and gender hierarchies — they do not dissolve them. A Dalit farmer with a smartphone is still a Dalit farmer. The social dynamics of the village — who speaks in panchayat meetings, whose land gets irrigated first, whose children get recommended for government jobs — are determined by power structures that predate and outlast any app. Anecdotal evidence suggests that social media, far from dissolving caste consciousness, has in some ways reinforced it — caste-based WhatsApp groups, caste-based online mobilization, caste-based content communities.

The fundamental equation: You can digitize poverty without eliminating it. A state where every citizen has a bank account, a smartphone, and a UPI ID, but where 35% live below or near the poverty line, has not been transformed by digital technology. It has been connected by digital technology. Connection and transformation are different phenomena.

10. E-Governance Innovations

Key Platforms

Odisha has invested significantly in e-governance, producing several platforms that are recognized nationally:

Bhulekh (Online Land Records): Odisha’s digitized land records system, providing online access to Record of Rights (RoR), plot information, and land maps. Land records in Odisha were historically maintained in handwritten registers, prone to manipulation, loss, and corruption. Digitization makes records accessible, searchable, and harder to tamper with. The integration of blockchain technology adds an additional layer of tamper-resistance. For a state where land disputes are endemic — estimated to constitute 40-60% of civil litigation — digital land records represent a genuine improvement. However, digitization of incorrect records produces digitally accessible incorrect records; the underlying land survey and settlement data, much of it dating to the colonial period, contains errors that digital systems inherit.

Mo Sarkar (My Government): Perhaps the most innovative of Odisha’s e-governance initiatives. After a citizen receives a government service (hospital visit, police interaction, district office transaction), they receive a phone call from a government call center asking them to rate the service and the specific official who served them. This inverts the traditional power dynamic: instead of citizens being grateful for any service received, officials are rated on their performance by the citizens they serve. The data feeds into performance assessments for officials.

The concept is powerful. Whether it transforms behavior sustainably or becomes a routine that officials learn to game (coaching citizens before they receive the call, selective service improvement for educated/vocal citizens who might give negative ratings) is an open question. But the design insight — using technology to invert the accountability direction — is genuinely innovative.

5T Framework: Chief Minister Naveen Patnaik’s governance framework, articulated in the mid-2010s: Teamwork, Transparency, Technology, Time, Transformation. Technology is positioned as one of five pillars, not a standalone solution. The framework has been implemented through: government dashboards tracking project progress, digital monitoring of scheme implementation, grievance redressal portals, and performance tracking of district and block officials.

JAGA Mission (Odisha Livable Habitat Mission): Uses drone mapping and digital records to provide land rights to urban slum dwellers. Approximately 1,725 slums in Odisha’s cities have been mapped using drones, with land rights certificates issued to residents. This is internationally recognized — JAGA Mission won the World Habitat Award in 2019. The use of drone technology to map informal settlements that were previously “invisible” to the land records system is a genuine innovation in governance.

e-Panchayat Sabha: Digital platform for conducting panchayat meetings, including livestreaming for transparency, digital record-keeping of decisions, and online display of budgets and expenditures. Addresses the historical opacity of local governance.

SKOCH Rankings and National Recognition

Odisha has performed exceptionally well in governance rankings in recent years:

  • SKOCH State of Governance 2023: Odisha ranked #1 nationally
  • SKOCH sectors where Odisha was top-ranked: Disaster Management, Education, Infrastructure, Water, Mining, Sports
  • District Governance and e-Governance: Improved from #4 to #2 nationally
  • United Nations: JAGA Mission recognized internationally; OSDMA’s cyclone response studied as a global model
  • NITI Aayog: Acknowledged Odisha’s improvement in multiple development indices

These rankings reflect genuine institutional effort and investment. Odisha under Naveen Patnaik’s 24-year tenure (2000-2024) built a reputation for clean, competent welfare administration — a stark contrast to the state’s earlier reputation as a governance backwater.

COVID Digital Governance

The COVID-19 pandemic tested digital governance infrastructure under crisis conditions:

  • Oxygen tracking dashboard: Real-time monitoring of oxygen availability across all hospitals in the state, enabling rapid reallocation during the second wave (April-May 2021) when oxygen shortages were critical.
  • Bed availability monitoring: Digital dashboard showing real-time bed availability in COVID hospitals, reducing the chaotic hospital-shopping that characterized the worst days of the second wave.
  • Vaccination scheduling: Integration with the national CoWIN platform, with state-level innovations in session planning and outreach to remote areas.
  • Digital contact tracing: App-based and call-center-based contact tracing during the first wave, transitioning to more targeted approaches in subsequent waves.

Assessment: E-Governance as Efficiency, Not Transformation

Odisha’s e-governance achievements are genuine and nationally significant. The state has invested more consistently in digital governance infrastructure than most Indian states, and the results are measurable in improved service delivery, reduced corruption in specific channels, and increased citizen access to government information.

But e-governance, by definition, optimizes existing governance. It does not transform what governance does; it transforms how governance does what it already does.

What e-governance improves: Speed of service delivery, transparency of records, accountability of officials (through Mo Sarkar-type feedback), accuracy of targeting (through Aadhaar-linked databases), and citizen access to information.

What e-governance does not change: The scope of what the state does. Whether the state’s primary orientation is welfare distribution or productive economy building is a political and institutional choice, not a technology choice. Odisha’s e-governance makes the welfare-and-extraction equilibrium run more efficiently — KALIA payments arrive faster, PDS leakage is reduced, disaster response is more precise. But the equilibrium itself — extractive mining plus efficient welfare — is not challenged by better technology.

The best e-governance in India still runs the same governance paradigm. A perfectly optimized digital system for distributing Rs 10,000/year to farmers via KALIA does not address the question of whether Rs 10,000/year is adequate, whether direct cash transfers are the optimal use of state resources, or whether the state’s agricultural policy produces sufficient income for farmers to not need transfers.

11. Comparison with Major IT/Digital Ecosystems

DimensionBhubaneswarHyderabadBangalorePune
IT/ITeS employment~60,000 (state-wide, Ch. 6 §6.8.1)~800,000~2,000,000~500,000
IT exports (annual)~Rs 12,900 cr (2023-24, Ch. 6 §6.8.1)~Rs 2.5 lakh cr~Rs 7+ lakh cr~Rs 1.5 lakh cr
VC funding (annual)Minimal~$2B~$15B~$2B
Unicorns (cumulative)010+50+10+
Anchor companies HQ’d0Several (e.g., T-Hub ecosystem)Many (Infosys, Wipro, etc.)Several
International flights0 direct30+ destinations50+ destinations15+ destinations
Higher ed pipelineNIT Rourkela, KIIT, IIITIIT-H, IIIT-H, ISB, OsmaniaIISc, IIM-B, multiple engineeringMultiple engineering, management
Ecosystem age~20 years~25 years~40 years~30 years
Product companiesNear zeroGrowingDominantGrowing
GCC/captive centersFewManyManyMany

The gap is not closing. Bangalore adds more IT jobs in a single quarter than Bhubaneswar has total. Hyderabad’s annual IT export growth exceeds Bhubaneswar’s total IT export value. The nature of technology ecosystems is that they exhibit increasing returns to scale: talent attracts companies, companies attract talent, success breeds investment, investment funds more companies. The ecosystem dynamics work against late entrants.

This does not mean Bhubaneswar’s IT sector is worthless — ~60,000 IT jobs state-wide (Economic Survey 2025-26, Ch. 6 §6.8.1), the bulk of them concentrated in Bhubaneswar, is significant and these are among the highest-paying private-sector jobs in the state. But it means that IT/ITeS will not be the engine of Odisha’s economic transformation in the way it was for Karnataka. The numbers simply do not scale.

What Bhubaneswar could realistically become: Not the next Bangalore (impossible — the ecosystem conditions cannot be replicated). More plausibly, a mid-tier IT delivery center that provides good employment for a few tens of thousands of people, supplements the state’s GDP modestly, and serves as a training ground that feeds talent into the national IT ecosystem. This is worth having. It is not transformative.

12. Synthesis: The Digital Layer on the Old Structure

The Pattern

Across every domain examined — internet access, UPI, Aadhaar, IT, startups, education, AI, e-governance — the same pattern emerges:

  1. Digital infrastructure has been deployed at unprecedented speed and scale. The JIO revolution, Aadhaar, UPI, and e-governance platforms have reached most of Odisha’s population in less than a decade.

  2. Informational barriers have collapsed. The farmer knows prices. The student can access content. The citizen can check records. The migrant can send money instantly. These are real changes.

  3. Institutional and economic structures have not correspondingly transformed. The factory that doesn’t exist cannot be created by an app. The doctor who isn’t there cannot be replaced by a chatbot. The job that requires migration still requires migration.

  4. The digital layer makes the old structure more visible, more efficient, and more connected — but does not change it. DBT delivers welfare more efficiently through the same extraction-redistribution model. E-governance optimizes the same governance paradigm. UPI digitizes the same remittance economy. AI amplifies the same institutional inequalities.

The Question for The Long Arc

The research documented here points to a central question for Chapters 6-8 of The Long Arc series:

Does exponential informational change eventually force structural economic change? Or can a society absorb radical informational transformation while maintaining structural economic continuity?

The optimistic case: information asymmetry collapse → awareness expansion → political demand for change → institutional response → structural transformation. In this view, digital technology is a slow-acting solvent that dissolves the old equilibrium over time. The farmer who sees on YouTube how agriculture works in Punjab demands better from his own state. The student who sees Bangalore’s opportunities demands them at home. The voter who can track government performance on a dashboard holds politicians accountable.

The pessimistic case: digital technology makes the existing equilibrium more bearable without disrupting it. Welfare arrives more efficiently, reducing the urgency of structural change. Entertainment (YouTube, social media) absorbs the energy that might otherwise fuel political mobilization. Information access creates the feeling of participation without actual economic participation. The system digitizes itself and persists.

The honest assessment: as of 2026, the evidence supports the pessimistic case more than the optimistic one. Digital transformation has been absorbed by Odisha’s existing structures — efficiently. The extraction economy has not changed. The migration patterns have not changed. The employment structure has not changed. What has changed is the information environment in which these persistent structures operate. Whether information eventually becomes a sufficient condition for structural change, or whether it is merely a necessary but insufficient one, is the open question for the next generation.

The Ninety-Year Frame

In the ninety-year frame (1936-2026) that The Long Arc examines:

  • First thirty years (1936-1966): Agricultural, pre-industrial, informationally isolated. Change was slow: literacy from 16% to 26%.
  • Next thirty years (1966-1996): Gradually urbanizing, dams and steel plants built, information limited to radio/TV/newspapers. Change was moderate: literacy from 26% to 49%.
  • Last thirty years (1996-2026): The exponential phase. Internet, smartphones, UPI, Aadhaar, AI. Change was explosive in the informational dimension: literacy from 49% to 80%+, phone access from near-zero to near-universal, information access from constrained to infinite.

But in the economic dimension — GSDP per capita relative to national average, industrial employment share, migration rates, agricultural productivity — the last thirty years look more like incremental improvement from a low base than exponential transformation.

This is the compression paradox: the informational curve is exponential. The economic curve is linear. The two coexist in the same people, the same villages, the same lives. A person can be informationally in 2026 and economically in 1996. Digital transformation has created a new kind of inequality — not between the connected and disconnected (though that persists), but between the informational reality and the material reality of the same person.

Whether this gap between informational exponential and economic linear is sustainable — whether it will eventually close through structural change or persist indefinitely as a feature of the new equilibrium — is perhaps the most important question for Odisha’s next thirty years.

Sources

  1. TRAI Quarterly Performance Indicators Reports (2016-2024) — telecom subscriber data, internet penetration, tele-density
  2. Reserve Bank of India: Digital Payments Statistics — UPI transaction volumes and values
  3. Ministry of Electronics and IT: India’s IT Industry Data — national IT export figures
  4. NASSCOM: Technology Sector Reviews (2020-2024) — IT employment, exports, startup data
  5. STPI Annual Reports — IT exports from STPI centers including Bhubaneswar
  6. Odisha Economic Survey (various years) — state-level economic indicators, IT sector data
  7. NITI Aayog: Digital India Dashboard — DBT savings estimates, digital inclusion metrics
  8. PMJDY Progress Reports (pmjdy.gov.in) — Jan Dhan account data
  9. Down To Earth: “No smartphones, internet access: Odisha’s rural kids caught in digital divide” (2020) — COVID education gap data
  10. OdishaTV: “Digital Divide Deprives Odisha’s Disadvantaged Children” (2020) — tribal district smartphone access
  11. ASPIRE Survey 2021: Learning During Lockdown — weekly learning material receipt rates
  12. Business Standard: “TCS, Wipro to expand headcount in Bhubaneswar” (2015) — IT employment expansion
  13. SKOCH State of Governance Report 2023 — governance rankings
  14. NIC Informatics: Odisha Digital Transformation (October 2024) — AI applications in Odisha governance
  15. Startup Odisha Portal (startupodisha.gov.in) — registered startup counts, incubation data
  16. KIIT TBI Annual Reports — incubation center data
  17. UIDAI: Aadhaar Dashboard (uidai.gov.in) — enrollment and authentication statistics
  18. World Bank: India Digital Economy Report (2024) — digital inclusion and economic impact analysis
  19. GSMA Mobile Gender Gap Report (2022, 2023) — women’s mobile phone access data
  20. ASER (Annual Status of Education Report) — learning loss during COVID
  21. Azim Premji Foundation: Loss of Learning during the Pandemic (2021) — state-level learning loss estimates
  22. NITI Aayog: India’s Booming Gig and Platform Economy (2022) — gig worker estimates
  23. NPCI: UPI Statistics Dashboard — monthly transaction data
  24. Odisha Computer Application Centre (OCAC) — e-governance implementation data
  25. JAGA Mission documentation and World Habitat Award citation (2019)
  26. Mo Sarkar evaluation reports — citizen feedback mechanism data
  27. OSDMA: Cyclone Fani Response Report (2019) — evacuation and response statistics
  28. Census of India 2011 — population, literacy, urbanization data for Odisha
  29. India Internet Report (IAMAI/Kantar) — internet usage patterns in rural and semi-urban India
  30. T-Hub Hyderabad annual reports — comparison data for startup ecosystem benchmarking

Cited in

The narrative series that build on this research.