Indian Telecom Sector Report: 5G, ARPU & Broadband

Q: How big is the Indian telecom market?

India has 1,282.4 million mobile connections (June 2026), the second largest base in the world. The mobile connectivity market is estimated at ₹2.9 trillion in FY2026 and forecast at ₹5.2 trillion by FY2031. Home broadband adds ₹340–350 billion and business connectivity ₹490–500 billion in FY2026. These sizes come from an Analysys Mason report quoted in the Jio Platforms prospectus.

Q: What is ARPU in telecom?

ARPU is average revenue per user: the revenue for a period divided by the average number of customers, usually shown per month in rupees. It moves with tariffs, plan mix and how many low-value users are in the base. India's mobile broadband ARPU was ₹199.5 a month in FY2026 and is forecast at ₹326.4 by FY2031.

Q: Who are the main telecom companies in India?

Four operators run mobile networks: Reliance Jio (49.78% of wireless broadband in June 2026), Bharti Airtel (35.47%), Vodafone Idea (12.51%) and state-owned BSNL (2.20%). The three private players hold over 97% of mobile broadband customers.

Q: Why is mobile data so cheap in India?

Jio launched in 2016 with free voice and very cheap data, and the rest of the industry had to follow. Revenue per GB fell from about ₹228 in 2016 to ₹7.5 in Q4 FY2026, against ₹33.4 in China and ₹228.4 in the USA. Heavy use (26.7 GB a person a month) spreads the network's fixed cost over a lot of data.

Q: What is fixed wireless access (FWA)?

Fixed wireless access brings home broadband over the air, using 5G or unlicensed-band radio links, instead of laying fibre to each home. It is faster and cheaper to roll out. In FY2026 it gave 9.8 million of the 14.2 million new home broadband connections in India.

Q: Is this report a recommendation to buy telecom stocks?

No. This is an educational sector study. It explains how the industry earns money and what to track. It does not recommend buying, selling or holding any security, and its forecasts come from a third-party report commissioned by Jio Platforms.

Snapshot · read this in a minute

The sector on one page

Telecom is the pipe under India's digital economy. Payments, shopping, streaming, government services and now AI all ride on it. This report turns the industry chapter of the Jio Platforms prospectus (pages 152 to 195) into one clear read: how the market got here, where the money is, who holds the airwaves, and which numbers to watch each quarter. Each chapter opens with a few lines in plain words, then the charts; the sources chapter at the end lists every page used, and the glossary explains new words.

Mobile broadband users: 1,021.4 mn; June 2026 · 69.62% of people
Still on 2G: 261.0 mn; June 2026 · the upgrade pool
Home broadband: 66.1 mn; June 2026 · only ~21% of homes
Digital economy: ₹49.6 tn; ~14% of GVA, 5th largest in the world

Four things to know

Data got 30 times cheaper, and India used it
One gigabyte cost about ₹228 in 2016 and ₹7.5 in Q4 FY2026. Monthly use per user went from under 0.2 GB to 26.7 GB. Cheap data built the habit that every app now depends on.
The market is now three private players
About ten operators competed before 2016. Today Jio holds 49.78% of wireless broadband, Bharti Airtel 35.47%, Vodafone Idea 12.51% and state-owned BSNL 2.20%.
Growth has moved from users to ARPU
Mobile broadband ARPU rose from ₹106.1 in FY2019 to ₹199.5 in FY2026 and is forecast at ₹326.4 by FY2031. Tariff hikes, 5G plans and a better customer mix drive it.
Home broadband is the open field
Only about 21% of Indian homes have fixed broadband, against 63% to over 100% in large economies. Fixed wireless (5G and radio links to the home) is filling the gap much faster than fibre could.

Explainer

How a telecom company earns, in one picture

Five things stand between an auction and your monthly plan. Almost all of the cost comes first and stays fixed.

Step 1 of 5 · Airwaves

Spectrum

Paid at auction

Radio airwaves. The government sells them at auction, and the right to use them usually runs for 20 years.

How it is paid for. Won at the auction and paid in one go or in instalments over the years. A very large bill that comes before any revenue.

20 yearsUsual term of spectrum won at auction
26,801 MHzJio's holding, the largest; about 15.8 years of life left on average, as of 30 June 2026
59,285 MHzHeld by the four operators shown in Figure 3.7 together (Jio, Airtel, Vodafone Idea, BSNL/MTNL), June 2026 (our sum)derived
₹1,04,514 crSpectrum still owed by Jio in instalments, March 2026

What it costs

Mostly fixed

Spectrum, tower rent, power, fibre and staff cost about the same whether the network is half empty or full.

CostRevenueLossProfitbreak-even

More users on the same network

Too few users: the fixed network costs more than they pay. A loss.

Illustration, not to scale. Drag the handle, tap the chart, or focus it and use the arrow keys.

What it earns

Users × ARPU × 12

Once the network is built, every extra user or extra rupee of ARPU is close to pure profit.

Hover, tap or tab to a number to see where it comes from.

The RHP's own size of the mobile broadband market: ₹2.3 trillion (FY2026). Our multiplication gives a little more, likely because the user count is a year-end figure while ARPU is worked out on the average base across the year. Users is not fixed: 1.67% to 3.9% of customers leave each month, so operators must keep winning new ones just to stand still.

The proof

Of every ₹100 a customer pays, what is left

Day-to-day costs come first. Then the network itself has to be paid for: spectrum, towers and equipment are written down a little each year. Jio and Airtel keep ₹30 to ₹33. Vodafone Idea, with the fewest users and the smallest bills, spends more than ₹100.

Before the write-down, Airtel India keeps the most (₹58.55), helped by the biggest bills. After it, Vodafone Idea's costs run past the ₹100 line: a loss of ₹6.92 on every ₹100. Hover, tap or tab to a bar.

EBITDA margin is profit before interest, tax, depreciation and amortisation; EBIT margin is after depreciation and amortisation (both FY2026). The gap between them is the yearly write-down of spectrum, network equipment and, under Ind AS 116, most leased tower and site rent. Airtel India leaves out its tower business and its ARPU is for mobile only, so compare direction more than level.

The economy

The economy that telecom runs on

Phone and broadband spending grows with incomes. The RHP opens its industry chapter with the economy, and so do we.

A phone plan is one of the first things a family pays for and one of the last it gives up. Below are the numbers the prospectus uses to argue that incomes, and the share of life lived online, will keep rising. Every one of those online habits, from UPI payments to streaming, runs over a telecom network.

GDP, FY2026: ₹346.4 tn; US$3.8 tn; 11.77% a year since FY2021
GDP, FY2031: ₹579.0 tn; US$6.4 tn forecast; 10.82% a year
Real growth, FY2027: 6.7%; Reserve Bank of India estimate
Household income: ₹0.71.1 mn; Median a year, FY2026 → FY2031; 9.19% a year
Median age: ~28; A young, phone-first population
Consumer spending: 56%+; Private consumption share of GDP, FY2026

Steady foundations

Fiscal deficit: 6.4%4.4%; Of GDP, FY2023 → FY2026
Retail inflation: ~2.1%; FY2026, a six-year low
Current account deficit: <1.5%; Of GDP in each of FY2024–26
Foreign investment: US$70 bn+; Gross FDI in each of the last five years

Exhibit

India's digital economy

₹ trillion · the % under each year is the share of gross value added (GVA). FY2031 is a forecast.

FY2031Forecast21.7% of GVA

Hatched = forecast in the RHPThe years between columns are not equal; the strip under the axis shows each step.Changes (% or ×) are our calculation from the RHP figures.

India's digital economy: ₹15.3 trillion in FY2019 (8.49% of GVA), ₹49.6 trillion in FY2026 (14.4%), ₹125.8 trillion forecast for FY2031 (21.7%)
Column	Digital economy
FY2019 · 8.49% of GVA	₹15.3 tn
FY2026 · 14.4% of GVA	₹49.6 tn
FY2031P · 21.7% of GVA	₹125.8 tn

Growth of 18.25% a year to FY2026, then 20.47% a year forecast. India is the world's fifth-largest digital economy (ICRIER).

The habits cheap data built

Times bigger derived

UPI usersFY2019 → June 2026
80555 mn6.9x
Digital paymentsTransactions a year, FY2019 → FY2026; the world's largest real-time volume
23282 bn12.3x
OTT video usersFY2016 → 2025
30601 mn20x
Rural internet usersFY2019 → March 2026
227441 mn1.9x
Start-ups2016 → March 2026; 100+ unicorns, third-largest hub
500223,000446x
Mobile data trafficQ4 FY2026; second only to China
76.1 EB–

What it means: Median household income is forecast to rise from about ₹7 lakh to ₹11 lakh a year by FY2031. The report links the ARPU forecast to that rise: as incomes grow, people pay more for data, and more homes can afford fixed broadband (137 to 201 million).

The 2016 reset

How 2016 reset the whole industry

Before 2016 about ten operators competed, 87% of users were on 2G, and data was a luxury. One launch changed the price of data, and the rest of the industry had to follow.

In September 2016 Jio launched with free calls and very cheap data. Rivals had to cut prices and race to build 4G. Within a few years the field shrank to three private operators plus state-owned BSNL; Vodafone and Idea merged in 2018 to survive.

The scissors · India's mobile data, FY2016 to FY2031

Data got 30x cheaper, while use per customer rose from under 0.2 GB to 26.7 GB a month.

After Jio arrived in 2016, price and use moved apart like the two blades of a pair of scissors. Both lines start from their FY2016 level, so you can see how far each has moved.

FY26 · Jan–Mar 2026 · point at, tap or tab to a yeartap a year

Price of 1 GB

₹7.5 a GB

About US$0.1. 6% below FY19; 30x cheaper than FY16derived

Data per customer

26.7 GB a month

Up 16.70% a year since FY19

Height shows each figure as a multiple of its FY2016 level (1x). Each gridline is 10 times the one below. derived

Price of 1 GB

₹228 to ₹7.5

30x cheaper since FY16 derived

Price paid per GB, Jan–Mar 2026

India₹7.5~30x below the USA, ~4.5x below Chinaderived
China₹33.4
USA₹228.4

The RHP measures this as revenue per GB. India paid ₹228 in FY16, about what the USA pays today. Revenue per GB kept falling even as revenue per user (ARPU) rose.

Data per customer, a month

<0.2 to 26.7 GB

Over 130x more since FY16 derived bases differ, see note

Data used per customer a month, latest

India26.7 GBAbout 1.4x the USA and Chinaderived
USA and China~18–19 GB

India: per data customer, Jan–Mar 2026 (TRAI). USA and China: one combined figure, total traffic divided by customers, GSMA Intelligence 2025. The two bases differ slightly.

Exhibit

Thirty years of the mobile phone in India

What a phone does for people, on a relative scale. Redrawn from the RHP's three-phase illustration.

Hover, tap or tab to a numbered dot to read each moment. Esc closes it.

Jio launched as an all-internet network built for 4G voice (VoLTE) and video. It made calls free and charged only for data. It reached 100 million 4G users in under 180 days and 42.67% of broadband users within a year. Rivals copied the model, and within three years data use rose more than fifty-fold.

Exhibit

India stopped waiting for new networks

Years between the first launches abroad and India's commercial launch

How long India waited for each new network

8–10years for 3G2–3years for 5G

The delay fell from nearly a decade for 3G to two to three years for 5G.

Point at, tap or tab to a row to see its details below.

5GFifth generation networksLatest

First launched abroad: 2019–2020; South Korea, USA, Singapore, Japan
India's commercial launch: 2022; By private operators
India's wait: 2–3 years; After the first launches abroad

India is one of the fastest 5G adopters: about 43% of mobile broadband customers moved to 5G in less than four years (TRAI, as of 31 March 2026).

Timeline

Ten years that remade the industry

The dated steps from Jio's launch to 1.28 billion connections.

7.7×more people on mobile broadband: 132.2 mn in March 2016, 1,021.4 mn in June 2026.

−71%fewer on 2G-only phones: 900.9 mn down to 261.0 mn.

We worked out 7.7× and −71% from the RHP figures.

On mobile broadband (4G, 5G, 3G)
Not on broadband (mostly 2G)
Years not reported
A step by Jio
A step by the whole industry

The RHP reports these lines at four dates: March 2016, March 2019, March 2026 and June 2026. It gives no yearly values in between, so that stretch is dashed. Pins sit on the date of each step; the two steps the RHP dates only loosely say so on their card.

Exhibit

What cheap data built

Users of each digital service in India, million, FY2019 and FY2026

E-commerce

Changes (% or ×) are our calculation from the RHP figures.

Digital service users in India, FY2019 and FY2026, million
Column	FY2019	FY2026
E-commerce	120	270
OTT video	275	601
Govt services	21	539
Digital wallets	80	555
Online gaming	300	517

How many Indians are online, FY2019 → FY2026

Population: 1,3781,467 mn; Everyone, for scale
Mobile broadband users: 5441,002 mn; 68% of people today (derived)
Unique internet users: 479772 mn; 53% of people today (derived)

What it means: The cheapest data in the world was not a gift. It was a strategy to make the network the default way India does everything. That habit is what lets operators raise prices today without losing most users.

India vs the world

India against China and the USA

India already uses more mobile data per person than either country, yet most digital services reach far fewer of its people. That gap is the growth case.

The gap is widest in everyday services: 18% of Indians shop online, against 69% in China and 82% in the USA. Read each gap as room to grow, not as a promise. The report uses these gaps to argue that Indians will spend more on connectivity and digital services as incomes rise.

Exhibit

How many people use each service

% of population, FY2026. Above 100% means more connections than people.

Mobile broadband

Gaps against India (percentage points) are our calculation from the RHP figures.

Share of population using each digital service in India, China and the USA
Column	India	China	USA
Mobile broadband	68%	132%	124%
Internet users	53%	84%	83%
E-commerce	18%	69%	82%
OTT video	41%	71%	88%
Govt services	37%	63%	29%
Digital wallets	38%	71%	61%
Online gaming	35%	49%	63%

India against other countries

Four measures of headroom

India trails on all four. The big number on each tile is how far apart India and the top country are; a “+” means the true gap is larger still.

Homes with fixed broadband

% of households · India as of Q1 FY2027 · bars run 0 to 100%

Room to grow

The USA, China, the UK, Japan and South Korea have more fixed broadband connections than homes (business lines included). India has about 1 in 5.

The benchmarks include business connections (RHP footnote 48), which is why some pass 100%. Canada is ~94%; the world outside India was above 70% in FY2026. In FY2016 India was ~6%, the USA ~81% and China ~60%. Inside India: ~43% of urban homes, ~7% of rural homes, Q1 FY2027.

Homes with a computer

% of households · bars run 0 to 100%

Room to grow

Fewer than 1 in 10 Indian homes own a computer. In the USA it is more than 95%.

Smartphones are common in India; computers are not, which limits who can use productivity and AI tools. Jio's answer in the RHP is JioPC, a “virtual PC” that runs on a TV or set-top box.

Dedicated business lines

Million dedicated connections, FY2026 · bars scaled to the largest value

Room to grow

The USA has about 12 times as many dedicated business lines as India.

Fewer than 4 mn businesses in India have fixed broadband, under 5% of its 79 mn+ MSMEs (FY2026).

Enterprise digital spending

% of GDP, FY2026 · bars run 0 to 5% of GDP

Room to grow

In the UK, Canada, Japan and South Korea, digital services are 3–5% of GDP. Jio's addressable market in India is 1.48%.

The two are defined differently: the peers' figure is all digital services spending; India's is the part Jio can sell to, ~₹5.12 trillion (US$56.8 bn). The RHP says such spending can reach up to 5% of a country's GDP.

The multiples (×) are our arithmetic from the RHP figures, rounded; they are not stated in the RHP. “~” means about, “+” means more than, “<” means less than. Periods are FY2026 or Q4 FY2026 unless a tile says otherwise.

Is India's ARPU too low for its income?

The report plots mobile ARPU against income per person for many countries; the two move together closely (R² = 0.81). India's FY2026 mobile broadband ARPU (US$2.2) and even its FY2031 forecast (US$3.6) sit below that line, which is the report's case for more tariff headroom.

What it means: India leads the world on data used and trails it on homes with broadband, computers at home and businesses online. Each of those gaps is a line in an operator's growth plan, and each is a forecast, not a promise.

Mobile

1.28 billion connections, and the room still left

India is the second largest mobile market in the world. Almost four in five connections are now broadband, and the rest are the easiest upgrades left.

India has 1,282 million mobile connections. Some people carry two SIMs, so these are connections, not people. 1,021 million of them are broadband, meaning 4G or 5G. The other 261 million are still on 2G, mostly in villages, mostly because a smartphone and a data plan still cost too much.

The size of the mobile market is simple arithmetic: users × what each pays a month × 12. The calculator below lets you try it.

Exhibit

Broadband users, and the 2G base still to move

Million connections · the % under each year is mobile broadband users as a share of population

Q1 FY20271,282.4 mn connections

Mobile broadband connections equal 69.62% of the population

Cities are above 110% and villages about 45% (March 2026). Above 100% means more connections than people.

Mobile broadband1,021.4 mn79.6% of all connections+88% since FY2019

Other (2G)261.0 mn20.4% of all connections−58% since FY2019

Point at a year, or use the arrow keys. Totals, shares and changes are our arithmetic on the RHP figures; the striped column is a forecast.Penetration counts connections (SIMs) per head, not people, so one person with two SIMs counts twice. In FY2016 there were just 132.2 mn broadband connections against 900.9 mn others. The 261.0 mn users still on 2G in June 2026 are the easiest upgrades left. By FY2031 the report expects 1,339.4 mn on broadband, about 98% of all connections (our arithmetic).

Exhibits

Where broadband stands: by network, and by place

Mobile broadband users, March 2026. 5G took 43% in under four years; cities are full and villages are not.

5G took 43% in under four years

one square ≈ 10.0 mn customers

42.6%of mobile broadband customers are on 5G426.7 mn customers

Jio alone had 284.9 mn customers on its standalone 5G network by June 2026, the third-largest 5G base in the world.

Shares are of the 1,002.3 mn mobile broadband customers in March 2026. 4G and 5G add up to 999.7 mn; the RHP does not split the other 2.6 mn (our arithmetic).Speed is the pull: 5G downloads already average more than 180 Mbps, against just over 68 Mbps for all networks together in December 2025.

Cities are full; villages are not

One figure = 10 connections per 100 people Past 100% = second SIMs

The gap is about 65 points. Rural India is where the next users will come from (derived: 110 − 45). Tap a row for the detail.

Still on 2G, all India: 261 mn; Q1 FY2027 · held back by phone and plan prices
Mobile broadband, all India: 69.6% → 87.8%; Q1 FY2027 to FY2031 forecast

Build the market yourself

Revenue = users × ARPU × 12 months

Each rectangle's area is a year of mobile broadband revenue: users run across, ARPU (₹ a month per user) runs up. Drag the corner, tap the chart, or use the sliders.

₹2.40 tn

a year, our sum

Same as FY2026

Mobile broadband users1,002.3 mnARPU, ₹ a month per user₹199.5

FY2026 pool1,002.3 mn users × ₹199.5 × 12₹2.40 tn

Move a slider or drag the corner: new revenue shows in green, lost revenue in red.

Check against the report

₹ trillion a year · report figures from the RHP

Our sum ₹2.40

Our arithmeticFrom FY2026 to FY2031, higher ARPU adds about ₹1.5 tn a year and new users about ₹0.8 tn, with ₹0.5 tn from the two together. Users grow about 6% a year and ARPU about 10% a year, so price, not just reach, does most of the work.

The report puts FY2026 mobile broadband at ₹2.3 trillion; our sum is about 4% higher, probably because it uses the March 2026 user count, not the year's average, and the report rounds to one decimal. FY2031 figures are forecasts from the Analysys Mason industry report in Jio's RHP, which the company commissioned and paid for. Growth rates per year are our arithmetic.

What it means: Both levers still have room. Users can grow by another 300 million or so as 2G users upgrade, and ARPU is forecast to rise by about two-thirds. The report expects the mobile market to grow from ₹2.9 trillion to ₹5.2 trillion by FY2031.

ARPU

From more users to more from each user

Average revenue per user (ARPU) is revenue divided by the average number of customers, per month. It is the one number that tells you if the industry is earning more or just getting bigger.

ARPU is the average monthly bill: if an operator earns ₹2,000 crore in a month from 100 million customers, its ARPU is ₹200. For years operators chased users with cheap plans, so ARPU stayed low. Since then the industry has raised prices together, most recently in July 2024. ARPU can rise even while the price of a GB falls, because people keep using more data every year.

Exhibit

Mobile broadband ARPU

₹ a month · FY2031 is a forecast

FY2031Forecast

Hatched = forecast in the RHPThe years between columns are not equal; the strip under the axis shows each step.Changes (% or ×) are our calculation from the RHP figures.

Mobile broadband ARPU ₹106.1 in FY2019, ₹199.5 in FY2026, ₹326.4 forecast for FY2031
Column	Mobile broadband ARPU
FY2019	₹106.1
FY2026	₹199.5
FY2031P	₹326.4

Growth of +9.45% a year to FY2026, then +10.34% a year forecast to FY2031.

What pushes ARPU up

Faster networks
Average download speed went from ~2.5 Mbps (January 2016) to over 68 Mbps (December 2025). 5G averages over 180 Mbps.
Smartphones everywhere
80–85% of mobile users have a smartphone now, up from 20–25% in FY2016. Phones like JioBharat sell for under ₹1,000.
More data each month
Use per user is forecast to more than double, from 26.7 GB to 59.2 GB a month by FY2031.
Steady rules
Spectrum auctions, the waiver of spectrum usage charges and the Telecommunications Act 2023 made the rules more predictable.

Operating leverage, in one dial

What a ₹ rise in ARPU is worth

A price rise adds revenue without adding much network cost, so most of it reaches operating profit (EBITDA). Pick a rise per customer per month.

+₹10 a month

per customer

≈ ₹14,620 cr a year, all three

Jio
533.3 mn customers · ARPU ₹215.6
Extra revenue a year+₹6,400 cr
a 4.6% price rise
Share of FY26 EBITDA8.4%
of ₹76,255 cr
Bharti Airtel · India business
491.9 mn customers · ARPU (mobile) ₹263.9
Extra revenue a year+₹5,903 cr
a 3.8% price rise
Share of FY26 EBITDA7.2%
of ₹82,193 cr
Vodafone Idea
193.1 mn customers · ARPU ₹177
Extra revenue a year+₹2,317 cr
a 5.6% price rise
Share of FY26 EBITDA12.2%
of ₹19,004 cr

Jio: 533.3 mn customers × ₹10 × 12 months = ₹6,400 crore a year, which is 8.4% of its FY2026 EBITDA of ₹76,255 crore.

Our arithmeticThe same +₹10 matters most to Vodafone Idea: 12.2% of a year's EBITDA, because its profit base is the smallest. Jio gains the most rupees because it has the most customers.

Customers (June 2026) × rise × 12 months, before tax, churn, any share kept by distributors, and before the 8% licence fee on adjusted gross revenue and spectrum usage charges, so less than the full amount reaches EBITDA. Customers are each company's total base (Airtel includes Homes and DTH); Jio's EBITDA is Jio Platforms consolidated. EBITDA is FY2026; ARPU is the June 2026 quarter (Airtel is India mobile). Each company counts customers its own way, so treat this as a rough guide.

What it means: Because the network cost is mostly fixed, a rise in ARPU is close to pure profit. That is why tariff rounds move telecom stocks so much, and why a return of price cuts would hurt just as fast.

Competition

Four players, three that matter

A market of about ten players in 2015 is now three private operators and state-owned BSNL. Spectrum, the airwaves bought at government auctions, decides who can offer what.

A mobile network needs three things: spectrum, towers, and fibre to link the towers. Spectrum is the scarcest, and the more an operator holds in the right bands, the more traffic it can carry. Jio and Airtel hold most of it and most of the customers; Vodafone Idea holds far less and has been losing customers; BSNL is small but holds a large slice of the low band that reaches deep indoors. Each company keeps its own colour throughout.

Meet the operators · June 2026 quarter

Each bar is this operator's figure; the coloured ticks are its rivals.Hover or focus a row for rivals' numbers; click or tap it for the full comparison.

Market leader

Jio

Reliance groupIPO in October 2026

Customers

Customers, June 2026
- 1Jio533.3 mn
- 2Airtel491.9 mn
- 3Vi193.1 mn
Connections, not people. Jio includes home connections; Airtel counts India only.
Net new customers, June 2026 quarter
- QuarterFY26
- 1Airtel+9.5 mn+58.0 mn
- 2Jio+8.9 mn+36.2 mn
- 3Vi+0.3 mn−5.4 mn
Airtel: India business. Source: RHP KPI table.
Monthly churn, June 2026 quarter
- 1Jio1.61%
- 2Airtel2.59%
- 3Vi3.8%
Lower is better. Airtel: India mobile only. BSNL is not reported.
5G customersOn standalone 5G, June 2026284.9 mn

Money

Revenue per user a month (ARPU)
- 1Airtel₹263.9
- 2Jio₹215.6
- 3Vi₹177
Airtel's figure is mobile only; Jio's covers all its customers, so treat the gap as rough.
EBITDA margin, June 2026 quarter
- QuarterFY26
- 1Airtel59.14%58.55%
- 2Jio53.26%51.91%
- 3Vi43.07%42.35%
Profit before interest, tax, depreciation and amortisation, per ₹100 of revenue. Airtel: India business excluding towers. BSNL is not reported.
EBIT margin, June 2026 quarter
- QuarterFY26
- 1Jio34.23%33.36%
- 2Airtel31.75%30.37%
- 3Vi−3.70%−6.92%
Profit before interest and tax, after depreciation. Below zero means an operating loss. Airtel: India business excluding towers.

Network

Spectrum held, all bands
- 1Jio26,801 MHz
- 2Airtel22,538 MHz
- 3Vi8,030 MHz
- 4BSNL1,916 MHz
BSNL's figure includes MTNL. RHP Figure 3.7.
Low-band spectrum (below 1 GHz)
- HeldShare*
- 1Jio890 MHz39%
- 2BSNL706 MHz31%
- 3Airtel373 MHz16%
- 4Vi299 MHz13%
*Shares are derived by us: each holding ÷ 2,268 MHz held by the four. BSNL includes MTNL. RHP Figure 3.7.
Data used per customer a month
- 1Jio43.7 GB
- 2Airtel34.4 GB
- 3Vi21.7 GB
Airtel: India mobile only. Source: RHP KPI table.
Home broadband customers, June 2026
- TotalWired + 5G
- 1Jio28.6 mn14.6 + 14.0 mn
- 2Airtel14.9 mn11.1 + 3.8 mn
- 3BSNL4.5 mn4.5 + 0 mn
TRAI data in RHP Figure 4.5. The RHP gives no figure for Vodafone Idea.

Edge

Most users, most spectrum, lowest churn, and the lead in fixed wireless (78.6% share).

Watch out

ARPU below Airtel's (on different definitions); forecasts about it come from a report it paid for.

Number two

Bharti Airtel

Premium mixAfrica too

Customers

Customers, June 2026
- 1Jio533.3 mn
- 2Airtel491.9 mn
- 3Vi193.1 mn
Connections, not people. Jio includes home connections; Airtel counts India only.
Net new customers, June 2026 quarter
- QuarterFY26
- 1Airtel+9.5 mn+58.0 mn
- 2Jio+8.9 mn+36.2 mn
- 3Vi+0.3 mn−5.4 mn
Airtel: India business. Source: RHP KPI table.
Monthly churn, June 2026 quarter
- 1Jio1.61%
- 2Airtel2.59%
- 3Vi3.8%
Lower is better. Airtel: India mobile only. BSNL is not reported.

Money

Revenue per user a month (ARPU)
- 1Airtel₹263.9
- 2Jio₹215.6
- 3Vi₹177
Airtel's figure is mobile only; Jio's covers all its customers, so treat the gap as rough.
EBITDA margin, June 2026 quarter
- QuarterFY26
- 1Airtel59.14%58.55%
- 2Jio53.26%51.91%
- 3Vi43.07%42.35%
Profit before interest, tax, depreciation and amortisation, per ₹100 of revenue. Airtel: India business excluding towers. BSNL is not reported.
EBIT margin, June 2026 quarter
- QuarterFY26
- 1Jio34.23%33.36%
- 2Airtel31.75%30.37%
- 3Vi−3.70%−6.92%
Profit before interest and tax, after depreciation. Below zero means an operating loss. Airtel: India business excluding towers.

Network

Spectrum held, all bands
- 1Jio26,801 MHz
- 2Airtel22,538 MHz
- 3Vi8,030 MHz
- 4BSNL1,916 MHz
BSNL's figure includes MTNL. RHP Figure 3.7.
Low-band spectrum (below 1 GHz)
- HeldShare*
- 1Jio890 MHz39%
- 2BSNL706 MHz31%
- 3Airtel373 MHz16%
- 4Vi299 MHz13%
*Shares are derived by us: each holding ÷ 2,268 MHz held by the four. BSNL includes MTNL. RHP Figure 3.7.
Data used per customer a month
- 1Jio43.7 GB
- 2Airtel34.4 GB
- 3Vi21.7 GB
Airtel: India mobile only. Source: RHP KPI table.
Home broadband customers, June 2026
- TotalWired + 5G
- 1Jio28.6 mn14.6 + 14.0 mn
- 2Airtel14.9 mn11.1 + 3.8 mn
- 3BSNL4.5 mn4.5 + 0 mn
TRAI data in RHP Figure 4.5. The RHP gives no figure for Vodafone Idea.

Edge

Highest ARPU and margin: more postpaid and higher-value users.

Watch out

Smaller fixed-wireless base (21.4% share) and higher churn than Jio.

Number three

Vodafone Idea

Rebuilding

Customers

Customers, June 2026
- 1Jio533.3 mn
- 2Airtel491.9 mn
- 3Vi193.1 mn
Connections, not people. Jio includes home connections; Airtel counts India only.
Net new customers, June 2026 quarter
- QuarterFY26
- 1Airtel+9.5 mn+58.0 mn
- 2Jio+8.9 mn+36.2 mn
- 3Vi+0.3 mn−5.4 mn
Airtel: India business. Source: RHP KPI table.
Monthly churn, June 2026 quarter
- 1Jio1.61%
- 2Airtel2.59%
- 3Vi3.8%
Lower is better. Airtel: India mobile only. BSNL is not reported.

Money

Revenue per user a month (ARPU)
- 1Airtel₹263.9
- 2Jio₹215.6
- 3Vi₹177
Airtel's figure is mobile only; Jio's covers all its customers, so treat the gap as rough.
EBITDA margin, June 2026 quarter
- QuarterFY26
- 1Airtel59.14%58.55%
- 2Jio53.26%51.91%
- 3Vi43.07%42.35%
Profit before interest, tax, depreciation and amortisation, per ₹100 of revenue. Airtel: India business excluding towers. BSNL is not reported.
EBIT margin, June 2026 quarter
- QuarterFY26
- 1Jio34.23%33.36%
- 2Airtel31.75%30.37%
- 3Vi−3.70%−6.92%
Profit before interest and tax, after depreciation. Below zero means an operating loss. Airtel: India business excluding towers.
Net debt ÷ EBITDA, FY26Years of operating profit to repay debt7.75x

Network

Spectrum held, all bands
- 1Jio26,801 MHz
- 2Airtel22,538 MHz
- 3Vi8,030 MHz
- 4BSNL1,916 MHz
BSNL's figure includes MTNL. RHP Figure 3.7.
Low-band spectrum (below 1 GHz)
- HeldShare*
- 1Jio890 MHz39%
- 2BSNL706 MHz31%
- 3Airtel373 MHz16%
- 4Vi299 MHz13%
*Shares are derived by us: each holding ÷ 2,268 MHz held by the four. BSNL includes MTNL. RHP Figure 3.7.
Data used per customer a month
- 1Jio43.7 GB
- 2Airtel34.4 GB
- 3Vi21.7 GB
Airtel: India mobile only. Source: RHP KPI table.

Not in the RHP for Vi

Home broadband

Edge

Customer losses have slowed to almost nil (+0.3 mn in Q1 FY2027).

Watch out

Operating loss, net debt 7.75x EBITDA, and the highest churn of the three.

State-owned

BSNL

Network

Spectrum held, all bands
- 1Jio26,801 MHz
- 2Airtel22,538 MHz
- 3Vi8,030 MHz
- 4BSNL1,916 MHz
BSNL's figure includes MTNL. RHP Figure 3.7.
Low-band spectrum (below 1 GHz)
- HeldShare*
- 1Jio890 MHz39%
- 2BSNL706 MHz31%
- 3Airtel373 MHz16%
- 4Vi299 MHz13%
*Shares are derived by us: each holding ÷ 2,268 MHz held by the four. BSNL includes MTNL. RHP Figure 3.7.
Home broadband customers, June 2026
- TotalWired + 5G
- 1Jio28.6 mn14.6 + 14.0 mn
- 2Airtel14.9 mn11.1 + 3.8 mn
- 3BSNL4.5 mn4.5 + 0 mn
TRAI data in RHP Figure 4.5. The RHP gives no figure for Vodafone Idea.

Not in the RHP for BSNL

CustomersNet new customersMonthly churnARPUEBITDA marginEBIT marginData per customer

The RHP compares Jio only with the listed private operators, Airtel and Vodafone Idea. BSNL is state-owned and not listed, so its customers, revenue and profit figures are not in the peer table.

Edge

Large low-band holding (31% of sub-1 GHz) for rural coverage.

Watch out

Small and shrinking share; no private-sector KPIs in the RHP.

June 2026 quarter unless marked FY26. Sources: RHP KPI peer table, market share, spectrum Figure 3.7, home broadband Figure 4.5. Bars, ranks and low-band shares are derived by us and compare only the operators the RHP reports.

Explainer

What each band of spectrum does

Lower frequencies travel farther and through walls; higher ones carry more data over a shorter distance. Operators need a mix.

The rule: the higher the frequency, the shorter the reach, but the more data it carries.

Hover a band anywhere to follow it. Tap a card to pin it.

Drawing not to scale

Frequency on a log scale. Coloured zones are the frequencies in Jio’s licences. The RHP calls Jio’s 3,300 MHz holding its 5G mid band; this page groups 1–3 GHz as “mid band”, like Exhibit 10.

Low band
Below 1 GHz
Reachmost of the four bandsCapacityleast of the four bands
Travels farthest and gets through walls. Coverage for villages, highways and deep inside buildings.
Strong indoorsRHP: 5G low band
The RHP says 700 MHz “can provide deep indoor 5G coverage”, working together with 3,500 MHz.
Most held:Jio890 MHz
Mid band
1–3 GHz
Reachmore of the four bandsCapacityless of the four bands
Good reach and good capacity. A balance of the two, used across towns and cities.
Fair indoors
Jio’s licences in this group are 1,800 and 2,300 MHz.
Most held:Airtel1,965 MHz
C-band
Around 3.5 GHz
Reachless of the four bandsCapacitymore of the four bands
Shorter reach, wide channels. Fast 5G speeds in cities.
Weak indoorsRHP: 5G mid band
The RHP defines C-band as spectrum around 3.5 GHz and calls Jio’s 3,300 MHz its 5G mid band.
Most held:Jio2,440 MHz
mmWave
26 GHz
Reachleast of the four bandsCapacitymost of the four bands
Short reach, blocked by walls. Huge capacity in a small area, such as wireless broadband to homes.
Blocked by wallsRHP: 5G mmWave
The RHP says 26 GHz lets Jio run fixed wireless home broadband at high capacity.
Most held:Jio22,000 MHz

Jio holds all three 5G tiers

The RHP says Jio is the only operator with 5G spectrum in the low band (700 MHz), mid band (3,300 MHz) and millimetre wave (26 GHz) in every circle. It combines 700 MHz with 3,500 MHz to give deep indoor 5G: the far-reaching band and the high-capacity band working together.

Jio’s 26,801 MHz, by band

Low band890
Mid band1,471
C-band2,440
mmWave22,000

mmWave is 82.1% of Jio’s MHz (derived: 22,000 ÷ 26,801). Wide high bands carry data; the thin low band carries reach.

MHz held and the leader in each band are from RHP Figure 3.7 (via the data file; Exhibit 10 below shows every operator). Reach and capacity dots rank the four bands against each other and are not measurements; the plain-English jobs and indoor ratings are general background, not RHP figures.

Exhibit

Who owns the airwaves

Share of spectrum held in each band, MHz, Q1 FY2027. Low bands reach indoors; high bands carry the most data.

How big each band isShare of all 59,285 MHz the four operators hold derived

Low band3.8%
Mid band9.8%
C-band9.9%
mmWave76.5%

Low bandSub 1 GHz · reaches deep indoorsIncludes 700 MHz, the RHP’s 5G low-band.

2,268 MHz3.8% of all MHz

Mid band1–3 GHz · the 4G workhorseMostly 4G. Not the RHP’s 5G “mid-band”, which is the C-band below.

5,827 MHz9.8% of all MHz

C-band3.5 GHz · the main 5G band3.3–3.6 GHz. The RHP calls this the 5G mid-band (3,300 MHz).

5,840 MHz9.9% of all MHz

mmWave26 GHz · huge capacity, short reachThe RHP’s 5G “millimetre wave” band.

45,350 MHz76.5% of all MHz

All spectrumEvery band together

59,285 MHzfour operators

About 76.5% of these MHz are mmWave (derived), so this row mostly follows the mmWave bar. It does not show who has the most of the bands that reach indoors.

Shares are worked out from the MHz figures (derived). “% of all MHz” is the band’s size against the 59,285 MHz the four operators hold together, not all spectrum in India. Hover or tap any bar or value; pick a company above to follow it through every band. Rounding in the source: Mid band: the companies add to 5,828 MHz, the band total shows 5,827; Vodafone Idea: its bands add to 8,031 MHz, the total shows 8,030.

Exhibit

Speed, and what Jio says about itself

Average download speed, Mbps, Q3 FY2026 (Opensignal). The RHP names only Jio. On the right, Jio's own claims: the company's view, not checked facts.

Average download speed

Jio#1 of 4

107.3 Mbps1.8× Player 2 · 3.5× the other three's average (30.8 Mbps) (our arithmetic)Jio's standalone 5G averaged 198.9 Mbps in the same quarter.

Player 2#2 of 4

58.2 MbpsJio is 1.8× this (our arithmetic)The RHP does not name Players 2 to 4.

Player 3#3 of 4

25.1 MbpsJio is 4.3× this (our arithmetic)The RHP does not name Players 2 to 4.

Player 4#4 of 4

9.0 MbpsJio is 11.9× this (our arithmetic)The RHP does not name Players 2 to 4.

Point at or tab to a bar to compare it with Jio. Ratios and gaps are our arithmetic.Dashed line: all networks together averaged just over 68 Mbps in December 2025.The RHP says Jio's speed has been more than double the average of the other operators for the last three years. 5G alone averages more than 180 Mbps across India; Jio's standalone 5G averaged 198.9 Mbps in Q3 FY2026.

What Jio says about itself

Largest spectrum bank: 26,800.8 MHz with 16.0 years average life left.
About 27 million net active mobile additions in FY2026, about 3 times the number two.
4G/5G coverage of over 99% of people; lower cost to carry 1 GB than the industry average.

What it means: Three players with heavy, sunk spending in spectrum have little reason to start a price war, and every reason to lift ARPU together. Vodafone Idea is the weak link: it lost customers three years running and still carries heavy debt (see the scorecard).

Home broadband

The open field: broadband for the home

Mobile came first in India. Wired internet to the home stayed rare because laying fibre street by street is slow and costly. Fixed wireless is changing that.

Home broadband means a fixed connection, a Wi-Fi router at home, instead of mobile data. Only about one Indian home in five has one, against nearly every home in large economies.

The problem was never demand; it was supply. Laying fibre down every street is slow and costly. Fixed wireless beams 5G or radio to a small box at home, so operators can connect a home without digging. That is why it supplied most new connections in FY2026.

Exhibit

Home broadband finally took off

Fixed broadband customers, million · the % under each year is the share of households

FY2031Forecast46%

Hatched = forecast in the RHPThe years between columns are not equal; the strip under the axis shows each step.Changes (% or ×) are our calculation from the RHP figures.

Fixed broadband customers 17.5 mn FY2016, 19.7 mn FY2019, 66.1 mn June 2026, 150.4 mn forecast FY2031
Column	Fixed broadband customers
FY2016 · 6%	17.5 mn
FY2019 · 7%	19.7 mn
Q1 FY2027 · 21%	66.1 mn
FY2031P · 46%	150.4 mn

Explainer

Two ways to bring the internet home

Why fixed wireless did most of the growth in FY2026.

The same three homes, connected two ways. India has 66.1 mn fixed broadband customers, about 21% of households (June 2026).

The answer in one lineIn FY2026, 9.8 mn of the 14.2 mn net new broadband customers came through fixed wireless.

Way 1 · by cable

Fibre to the home

A glass cable dug into the street, all the way to your door. It is the benchmark: other ways are judged as “fibre-like”.

Share of fixed broadband: 69.11% in FY2026Up from 6.85% in FY2019. All fibre lines (FTTx)
The catch: Slow, costly diggingPermits street by street (right of way); the government is streamlining these rules
Where it fits: Strongest in big citiesFibre roll-out has historically been thin beyond metros and top cities; over 60% of Jio's ~1.0 mn km of fibre is inside cities
Jio on fibre: 10 mn customers in 51 monthsJio was the first in India to cross 10 mn fibre broadband customers; all its wired customers are on fibre

Who runs wired broadband (fibre and old copper), June 202647.7 mn customers

Hover or tap a company for its customers and share.

Way 2 · through the air

Fixed wireless (FWA)

Radio from a nearby tower to a box on your home. Fibre-like speed at a relatively low cost, the report says.

Share of fixed broadband: 26.11% in FY2026Up from 0% in FY2019; 27.05% by June 2026
The catch: Shares the mobile networkOperators short of spectrum often have to put mobile users ahead of fixed wireless
Where it fits: Towns and villagesBuilt for semi-urban and rural areas where fibre is thin. Fixed broadband reaches ~7% of rural vs ~43% of urban homes, June 2026
Radio used: 5G or unlicensed radio (UBR)Jio launched fixed wireless (JioAirFiber) in Sept 2023 and is the first in India to offer it over UBR. UBR is even cheaper than 5G fixed wireless
Jio on fixed wireless: Largest base within two yearsJio had India's largest fixed wireless customer base within two years of launch

Who runs 5G and UBR fixed wireless, June 202617.9 mn customers

Hover or tap a company for its customers and share.

Excludes ~0.5 mn customers on older fixed wireless technology.

Why it matters: fixed wireless brought most of the net growth

Net new fixed broadband customers, how they were connected and who added them. Skipping the trench avoids the delays and costs of laying fibre to each door, so operators can reach towns and villages quickly.

FY2026April 2025 to March 202614.2 mn net new customers

How they were connected

Hover or tap a part for the exact number. * = worked out from RHP figures.

Who added them

Hover or tap a company for its net additions. * = worked out from RHP figures.

Q1 FY2027April to June 20262.5 mn net new customers

How they were connected

Hover or tap a part for the exact number. * = worked out from RHP figures.

Who added them

Hover or tap a company for its net additions. * = worked out from RHP figures.

Net additions of fixed broadband customers, TRAI; they include a small base of business customers (~5%). Fixed wireless and company figures from RHP Figure 4.5. Jio took ~65% of all net additions since its fixed wireless launch in Sept 2023. * Worked out from RHP figures: “Wired and other” and “Everyone else” are the totals minus the printed parts; the fixed wireless and wired % are divisions.

Exhibit

Fibre and fixed wireless took over

Share of fixed broadband customers by technology

FY201919.7 mn customers

FY202663.6 mn customers

pts = percentage points: the later share minus the earlier one (derived). Hover or tap a part to follow it from one year to the next.

Fixed broadband by operator and technology, June 2026Million customers, with each operator's share of the row. The bar under each line shows the split.

Segment	Jio	Bharti Airtel	BSNL	Other ISPs	Total
Wired (fibre, DSL)	14.630.5%	11.123.3%	4.59.5%	17.536.7%	47.7 mn
5G / radio FWA	14.078.6%	3.821.4%	–not listed	–not listed	17.9 mn
All fixed broadband	28.643.3%	14.922.6%	4.56.8%	17.526.5%	66.1 mn
Share of net adds, Q1 FY27	1.561.7%	0.520.1%	–not listed	–not listed	2.5 mnnet adds

Wired (fibre, DSL)

Total 47.7 mn

Jio14.630.5%
Bharti Airtel11.123.3%
BSNL4.59.5%
Other ISPs17.536.7%

5G / radio FWA

Total 17.9 mn

Jio14.078.6%
Bharti Airtel3.821.4%
BSNL–not listed
Other ISPs–not listed

All fixed broadband

Total 66.1 mn

Jio28.643.3%
Bharti Airtel14.922.6%
BSNL4.56.8%
Other ISPs17.526.5%
Not split (FWA)0.50.74%

Share of net adds, Q1 FY27

Net adds 2.5 mn

Jio1.561.7%
Bharti Airtel0.520.1%
BSNL–not listed
Other ISPs–not listed
Not split18.2%derived

– The RHP lists no BSNL or other-ISP customers for 5G / radio fixed wireless, and gives net adds only for Jio and Airtel.
Not split: 0.5 mn fixed-wireless homes on older technology that the RHP does not split by operator. On net adds, the part not split is 100% minus Jio and Airtel (derived). Jio added 1.5 mn and Airtel 0.5 mn of the 2.5 mn.

Home broadband in numbers

Homes that can afford it: 137201 mn; FY2026 → FY2031, income above ₹8.29 lakh a year
FWA share of FY26 adds: 9.8 of 14.2 mn; Fixed wireless did most of the growth
Urban vs rural: ~43% vs ~7%; Fixed broadband reach, Q1 FY2027
Market size: ₹345925 bn; Midpoints, FY2026 → FY2031

Jio's home network, as the RHP describes it (company view)

Fibre network: ~1.0 mn route km; 30+ mn fibre-pair km, over 60% inside cities
C-band spectrum: 100 MHz+; Contiguous in every circle, 130 MHz in eight
Fibre milestone: 10 mn in 51 months; First operator to reach it

Why satellite is not the threat it sounds like

Starlink's entry plan costs about ₹4,800 a month in the USA and ₹2,500–4,500 in most of Africa. (Prices as of August 2026.) The report expects satellite to stay a niche for remote places, because fixed wireless gives fibre-like speed for far less.

What it means: Home broadband is where the next round of growth sits: 66 million homes today, 150 million forecast by FY2031, and a home customer pays far more than a mobile one (Airtel's home ARPU was ₹527 in FY2026 against ₹257 for mobile).

Enterprise

Business connectivity and enterprise services

Operators want to sell businesses more than a SIM: fibre and leased lines, cloud, messaging, IoT and managed IT. These are the pools the report sizes, with its five-year forecasts.

Businesses buy more than SIMs: office internet, private lines between branches, cloud servers, bulk SMS and WhatsApp messages, connected devices and IT management. The report groups these as enterprise digital services.

Big companies already use most of them. Small firms mostly do not: under 3% of SMEs, against 75–90% of large firms. India has over 79 million MSMEs, so even a small rise in adoption makes a large market.

Business connectivity

Business fixed broadband: <4~32 mn; FY2026 → FY2031; under 5% of MSMEs today
Revenue pool: ₹495905 bn; Midpoints, FY2026 → FY2031

Exhibit

Enterprise digital services: where the money is

₹ billion, FY2026 to FY2031 forecast, with yearly growth

All 7 together₹5,116 → ₹11,759 bn2.3× in five yearsour sum, matches the report's ~₹5.12 trillion

Point at, tab to or click a bar for all its numbers.

Exhibit

Inside the biggest pools

₹ billion, FY2026 to FY2031 forecast

Part of Business apps & security₹1,501 → ₹4,300 bn in all
Part of Managed services & IT support₹1,424 → ₹2,740 bn in all
Part of Enterprise communication₹542 → ₹1,220 bn in all; the rest is internal tools

Point at, tab to or click a bar for all its numbers. "est." marks a FY2031 value we worked out from the stated growth rate.

Exhibits

Machines are coming online

IoT connections by country, million; and smart electricity meters installed in India, million, one big driver.

IoT connections, million

China#1 of 10

3,058 mnbar not to scale24.7× India (our arithmetic)

USA#2 of 10

365 mn2.9× India (our arithmetic)

India#3 of 10

124 mnUSA is 2.9× India · India is ahead of Japan (1.5×) (our arithmetic)Figure 5.4 value. M2M connections rose 86% in a year: 66.5 mn in March 2025 to 123.9 mn in March 2026.

Japan#4 of 10

83 mnIndia is 1.5× this (our arithmetic)

Brazil#5 of 10

69 mnIndia is 1.8× this (our arithmetic)

Germany#6 of 10

63 mnIndia is 2.0× this (our arithmetic)

UK#7 of 10

61 mnIndia is 2.0× this (our arithmetic)

Italy#8 of 10

45 mnIndia is 2.8× this (our arithmetic)

South Korea#9 of 10

38 mnIndia is 3.3× this (our arithmetic)

Canada#10 of 10

21 mnIndia is 5.9× this (our arithmetic)

Point at or tab to a bar to compare it with India. Ratios and gaps are our arithmetic.China's bar is cut at the zig-zag so the others stay readable: it is not drawn to scale. Its full value is shown.

Smart meters installed, million

FY26Latest

Changes (% or ×) are our calculation from the RHP figures.

Smart meters installed, FY17 to FY26, million
Column	Smart meters (mn)
FY17	0.0
FY18	0.1
FY19	0.4
FY20	1.7
FY21	2.4
FY22	4.1
FY23	5.6
FY24	10.5
FY25	25.2
FY26	61.3

What it means: These numbers are large, but they are addressable markets drawn up for an IPO, crowded with global players such as AWS, Microsoft, Google and Cisco. Treat them as the size of the pitch, not as revenue any one operator will win.

Consumer services

Digital services for consumers

Streaming, music, games, cloud storage, AI assistants and smart homes, often sold bundled with a phone or broadband plan.

Consumer digital services are what people pay for on top of a data plan: streaming video and music, games, cloud storage, AI assistants and smart-home devices.

Operators rarely own these apps. Their advantage is the bundle: Jio packs 15+ streaming apps into its plans, which puts them in millions of homes cheaply.

Exhibit

Consumer digital services

₹ billion, FY2026 to FY2031 forecast

All 3 together₹818 → ₹1,724 bn2.1× in five yearsour sum

Point at, tab to or click a bar for all its numbers.

Exhibit

Entertainment is moving to streaming

₹ billion, FY2026 and FY2031 forecast

OTT video

Hatched = forecast in the RHPChanges (% or ×) are our calculation from the RHP figures.

OTT, pay TV, audio and cloud gaming revenue FY2026 and FY2031 forecast, ₹ billion
Column	FY2026	FY2031 forecast
OTT video	₹233	₹540
Pay TV	₹210	₹115
Audio streaming	₹80	₹224
Cloud gaming	₹19	₹80

Pay TV is the one line forecast to shrink, from ₹210 to ₹115 billion.

Two more numbers

Productivity suites: ₹63 bn; FY2026; consumer cloud and compute ₹97 bn
Game consoles: <1 mn; Under 1% of homes; cloud gaming skips the console

What it means: Bundling is the operator's edge here: a streaming app or a cloud PC sold inside a data plan reaches millions without its own marketing. The flip side is that Google, Netflix, Amazon and others own the content, so operators mostly earn a share, not the whole pie.

Global

Selling India's network playbook abroad

Jio built much of its own network technology. The report sizes three ways that know-how could be sold to operators abroad. All three are forecasts of opportunity, not business that exists today.

The report also sizes a market outside India: selling the technology Jio built at home to operators in other countries. It looks at three products: fixed wireless kits, a full 5G network, and cheap 4G phones running JioBharat's software. Each column below shows how many people still lack the service, then what fixing that would cost operators and what it would earn them each year.

The market Jio could address
New users on top
Served today
Not in the estimate

Money bars in all three cards share one scale: a full bar is ₹12.84 tn. Point at, tap or tab to any part of a bar to read it.

Fixed wireless kits

Can India's home-broadband playbook work abroad?

357 mn

new home broadband connections needed

That would lift these 97 countries from 139 mn connected homes to about 495 mn, the global average.

659 mn households in 97 low-broadband countries

Shares are our arithmetic (derived). 163 mn = 659 − 139 − 357.

The money

Fully new revenue

One-time network spend₹12.84 tn

US$143 bnUS$300–1,500 a customer

New revenue a year₹6.22 tn

US$69 bnAbout US$17 a customer a month

These homes pay nothing for broadband today.

Worldwide (excluding India), 1,595 mn of 2,121 mn homes have fixed broadband, 70–75%. These 97 countries are below 50%.
Fixed wireless reuses existing towers and backhaul, so it costs less than laying fibre to each door.
Jio's edge: fixed wireless on unlicensed radio (UBR) as well as 5G, which leaves licensed spectrum for mobiles.
Upside not counted: countries above 50% broadband and developed markets, where FWA could replace old lines or be a second line.

5G network stack

Who still has to build 5G?

2,027 mn

customers whose operators still have to build 5G

1,387 mn with no 5G coverage today, plus 640 mn new users these operators expect to win.

7,852 mn mobile customers outside India, plus 640 mn new users expected

Shares are our arithmetic (derived).

The money

Mostly existing revenue

One-time 5G build₹5.58 tn

US$62 bnUS$25–35 a customer

Revenue a year it keeps₹8.24 tn

US$92 bnAbout US$3.5 a customer a month

Operators already earn most of this; 5G protects it.

The gap sits with 254 operators in 131 countries that have not launched 5G.
Jio is the only operator with its own end-to-end 5G stand-alone stack at nationwide scale. India is only the fifth country with this ability.

Smart feature phones

How many mobile customers are still on 2G phones?

729 mn

mobile customers still on 2G handsets

9.3% of mobile customers outside India (derived). Most already have 4G coverage; the phone's price is the barrier.

7,852 mn mobile customers outside India

Shares are our arithmetic (derived). 7,123 mn = 7,852 − 729.

The money

Jio: platform share only

Spent on new phones₹1.31 tn

US$15 bnPaid by users; this goes to phone makers, not Jio

Revenue a year for operators₹1.86 tn

US$21 bnAbout US$2.2 a customer a month

Jio would earn a portion, for its phone software and platform.

Most 2G users are in Africa and developing Asia.
JioBharat moved India's 2G users to 4G with phones under ₹1,000.
Jio's routes to earn: licensing the JioBharat OS and platform, licensing phone reference designs, and embedded media and digital services.

What it means: Selling network gear abroad means competing with Nokia, Ericsson, Samsung and Huawei. For an investor this is an option, not a base case. Watch for signed international contracts before giving it any weight.

Scorecard

The operators, side by side

The prospectus puts the three listed private operators in one table, sourced from each company's filings. We pulled out the lines that matter most.

Three listed private operators compete in India. The table compares them line by line, using each company's own filings as reproduced in the prospectus. Some definitions differ (Airtel reports mobile ARPU, Jio a blended figure), so compare the direction more than the exact level. The chart after it shows three years of the same lines: who is adding users, who is raising ARPU, and who is turning operating profit into cash.

Operator scorecard, FY2026The three listed private operators, line by line. Green marks the best of the three and red the weakest. Lines that are not like for like are shown but not ranked.

Jio5 best0 weakest
Airtel India3 best0 weakest
Vodafone Idea0 best8 weakest

Jio is best on 5 of 8 lines, Airtel India on 3. Vodafone Idea is weakest on all 8. Derived: counted from the 8 lines below that can be compared.

Metric (FY2026)	Jio	Airtel India	Vodafone Idea
Size and growth
Customers, March 2026 (mn)Higher is better	524.4Best , best of the three	482.4	192.8Weakest , weakest of the three
Net customers added (mn)Higher is better	36.2	58.0Best , best of the three	−5.4Weakest , weakest of the three
Revenue (₹ crore)Higher is better	₹1,46,885Best , best of the three	₹1,40,373	₹44,873Weakest , weakest of the three
Each customer
ARPU, exit quarter (₹/month)Higher is betterAirtel is mobile only	₹214.0	₹257.2Bestmobile only , best of the three	₹174Weakest , weakest of the three
Data per customer (GB/month)Higher is betterAirtel is mobile only	42.3Best , best of the three	31.4mobile only	20.2Weakest , weakest of the three
Monthly churnLower is better, so the shorter bar winsAirtel is mobile only	1.67%Best , best of the three	2.43%mobile only	3.9%Weakest , weakest of the three
Profit
EBITDA marginHigher is better	51.91%	58.55%Best , best of the three	42.35%Weakest , weakest of the three
EBIT marginHigher is better	33.36%Best , best of the three	30.37%	−6.92%Weakest , weakest of the three
Profit after tax (₹ crore)Not compared: Vi's figure includes one-off gains, and Airtel India's is not reported	₹30,049	Not reported	₹34,552incl. one-off gainsA loss of about ₹24,055 cr without the one-off gainsderived
Debt and cash
Net debt ÷ EBITDANot compared: Jio leaves out spectrum dues, and Airtel's figure is for the whole group	0.36xexcl. spectrum dues	1.36xwhole group	7.75xHeavy debt: net debt is about 7.75 years of EBITDA
EBITDA less cash capex (₹ crore)Not compared: Vi's figure is not reported	₹42,071	₹51,186	Not reported

Size and growth

Customers, March 2026 (mn)

Higher is better

Jio524.4Best
Airtel India482.4
Vodafone Idea192.8Weakest

Net customers added (mn)

Higher is better

Jio36.2
Airtel India58.0Best
Vodafone Idea−5.4Weakest

Revenue (₹ crore)

Higher is better

Jio₹1,46,885Best
Airtel India₹1,40,373
Vodafone Idea₹44,873Weakest

Each customer

ARPU, exit quarter (₹/month)

Higher is betterAirtel is mobile only

Jio₹214.0
Airtel India₹257.2Bestmobile only
Vodafone Idea₹174Weakest

Data per customer (GB/month)

Higher is betterAirtel is mobile only

Jio42.3Best
Airtel India31.4mobile only
Vodafone Idea20.2Weakest

Monthly churn

Lower is better, so the shorter bar winsAirtel is mobile only

Jio1.67%Best
Airtel India2.43%mobile only
Vodafone Idea3.9%Weakest

Profit

EBITDA margin

Higher is better

Jio51.91%
Airtel India58.55%Best
Vodafone Idea42.35%Weakest

EBIT margin

Higher is better

Jio33.36%Best
Airtel India30.37%
Vodafone Idea−6.92%Weakest

Profit after tax (₹ crore)

Not compared: Vi's figure includes one-off gains, and Airtel India's is not reported

Jio₹30,049
Airtel IndiaNot reported
Vodafone Idea₹34,552incl. one-off gainsA loss of about ₹24,055 cr without the one-off gainsderived

Debt and cash

Net debt ÷ EBITDA

Not compared: Jio leaves out spectrum dues, and Airtel's figure is for the whole group

Jio0.36xexcl. spectrum dues
Airtel India1.36xwhole group
Vodafone Idea7.75xHeavy debt: net debt is about 7.75 years of EBITDA

EBITDA less cash capex (₹ crore)

Not compared: Vi's figure is not reported

Jio₹42,071
Airtel India₹51,186
Vodafone IdeaNot reported

Airtel India excludes the passive tower business. Its ARPU, data and churn are for mobile only; the profit and leverage lines for Airtel India are not reported, so leverage is for the whole Bharti Airtel group.
* Vodafone Idea's FY2026 profit includes exceptional items of ₹58,607 crore. Without them it made a loss, as it did in FY2024 and FY2025.
† Jio's net debt leaves out deferred spectrum dues (₹1,04,514 crore) and leases; the peers include them. The leverage numbers are not like for like.

Exhibit

Three years, line by line

Pick a line to see how each operator moved from FY2024 to FY2026, each company in its own colour.

Customers · Million, at each March year-end. Airtel = India excluding passive infrastructure.

FY2026Latest

Changes (% or ×) are our calculation from the RHP figures.

Customers FY2024 to FY2026
Column	Jio	Airtel India	Vodafone Idea
FY2024	482	406	213
FY2025	488	425	198
FY2026	524	482	193

What it means: Airtel earns the most per user and the best margin; Jio has the most users, the lowest churn and fast-rising free cash; Vodafone Idea loses the most customers and runs at an operating loss. For the full Jio picture, read our Jio Platforms IPO decode.

Risks

What could go wrong

The report lists the threats to the industry. We kept them and wrote each in plain words.

Telecom is a regulated business that needs a lot of capital. Most of its risks come from three places: the government (spectrum prices, licence and tariff rules), competitors restarting a price war, and the next technology needing another round of heavy spending.

Six risks the report lists

Rules can change
Spectrum pricing, licence terms and consumer rules (such as the recent order to offer voice-and-SMS-only vouchers) can limit how tariffs are set.
Price cutting can return
A rival with deep-pocketed backers could offer free or heavily discounted plans again. 2016–2019 showed how fast that resets the whole industry.
Technology keeps costing money
Each generation needs new spending and the retiring of old networks. Satellite is the talked-about disruptor, though the report expects it to stay niche in India.
Cyber security and privacy
More digital services mean more attack surface. A serious breach costs trust and invites regulation.
Customers leave
Tariff hikes, users dropping a second SIM, cleanup of inactive SIMs and easy number portability all push churn up.
A slower economy
In a slowdown people trade down from premium plans and companies cut digital spending, so ARPU growth stalls.

One more the report does not list: its own source. Forecasts drawn up for a company's IPO tend to be generous. Check the ARPU and broadband forecasts against each quarter's actual numbers before relying on them.

Tracker

How to read a telecom result

Eight numbers tell you most of what you need from a telecom quarterly result. Here is where to find each one and how to read it.

Telecom companies report every quarter, and the regulator publishes customer numbers every month. You do not need to read every page of either.

01 ARPU
Where: Results release, KPI table
Good sign: Rising each quarter without big customer losses
Warning sign: Rising only because customers left
02 Net customer additions
Where: TRAI monthly data and results
Good sign: Positive, and active users (VLR) rising too
Warning sign: Additions with falling ARPU: that is a price war
03 Monthly churn
Where: Results release
Good sign: Under about 2% a month
Warning sign: Rising after a tariff hike and not settling
04 Data per customer
Where: Results release
Good sign: Rising: people depend on the network more
Warning sign: Flat use with flat ARPU
05 5G share of users
Where: Results and TRAI
Good sign: Rising: 5G users tend to take bigger plans
Warning sign: 5G given away free with no ARPU lift
06 Home broadband adds
Where: TRAI and results
Good sign: Steady monthly adds, fixed wireless growing
Warning sign: Adds bought with deep discounts
07 EBITDA less cash capex
Where: Cash flow statement
Good sign: Positive and growing after a build cycle
Warning sign: Turning negative: a new build or a spectrum auction
08 Net debt ÷ EBITDA
Where: Balance sheet and notes
Good sign: Falling; check whether spectrum dues are counted
Warning sign: Rising, or dues and leases left out

Monthly: TRAI data
The regulator publishes subscriber numbers every month, by operator and by state. Net additions show up here weeks before results.
Quarterly: results
ARPU, churn, data use, margins and capex come in each company's results release and investor presentation.
Event: auctions and tariffs
Spectrum auctions add debt and capex; tariff rounds lift ARPU across the industry within a quarter or two.

FAQ, sources and disclosures

Questions, sources and disclosures

Words used in this report

ARPU: Average revenue per user: revenue ÷ average customers, per month.
Churn: The share of customers who leave in a month.
Spectrum: Radio frequencies an operator buys at auction to run its network.
MHz: Megahertz: how much spectrum is held. More MHz, more capacity.
4G / 5G: Network generations. 5G is faster and carries far more data.
SA 5G: Standalone 5G: runs without a 4G core; enables network slicing.
FWA: Fixed wireless access: home broadband over 5G or radio, no fibre to the door.
FTTH / FTTx: Fibre to the home (or to the building, kerb and so on).
Mobile broadband: A mobile connection on 4G or 5G; the rest are 2G.
VLR: Visitor location register: the count of SIMs actually active on a network.
EBITDA: Operating profit before interest, tax, depreciation and amortisation.
Capex: Capital spending: towers, fibre, radios and spectrum.

Questions readers ask

How big is the Indian telecom market?

India has 1,282.4 million mobile connections (June 2026), the second largest base in the world. The mobile connectivity market is estimated at ₹2.9 trillion in FY2026 and forecast at ₹5.2 trillion by FY2031. Home broadband adds ₹340–350 billion and business connectivity ₹490–500 billion in FY2026. These sizes come from an Analysys Mason report quoted in the Jio Platforms prospectus.

What is ARPU in telecom?

ARPU is average revenue per user: the revenue for a period divided by the average number of customers, usually shown per month in rupees. It moves with tariffs, plan mix and how many low-value users are in the base. India's mobile broadband ARPU was ₹199.5 a month in FY2026 and is forecast at ₹326.4 by FY2031.

Who are the main telecom companies in India?

Four operators run mobile networks: Reliance Jio (49.78% of wireless broadband in June 2026), Bharti Airtel (35.47%), Vodafone Idea (12.51%) and state-owned BSNL (2.20%). The three private players hold over 97% of mobile broadband customers.

Why is mobile data so cheap in India?

Jio launched in 2016 with free voice and very cheap data, and the rest of the industry had to follow. Revenue per GB fell from about ₹228 in 2016 to ₹7.5 in Q4 FY2026, against ₹33.4 in China and ₹228.4 in the USA. Heavy use (26.7 GB a person a month) spreads the network's fixed cost over a lot of data.

What is fixed wireless access (FWA)?

Fixed wireless access brings home broadband over the air, using 5G or unlicensed-band radio links, instead of laying fibre to each home. It is faster and cheaper to roll out. In FY2026 it gave 9.8 million of the 14.2 million new home broadband connections in India.

Is this report a recommendation to buy telecom stocks?

No. This is an educational sector study. It explains how the industry earns money and what to track. It does not recommend buying, selling or holding any security, and its forecasts come from a third-party report commissioned by Jio Platforms.

Where every number comes from

• Jio Platforms Limited, Red Herring Prospectus dated 14 October 2026: Industry Overview, p.152–195; peer KPI table, p.189–193.
• Underlying sources named in the RHP: TRAI (Telecom Subscription Data and Performance Indicators Reports), GSMA Intelligence, IMF World Economic Outlook (April 2026), MoSPI, MeitY, Opensignal, TeleGeography, Euromonitor, Ormax Media, FICCI & EY, Ministry of Power.
• Analysys Mason prepared the industry report for the issue. Jio Platforms commissioned and paid for it (RHP p.152; risk factor 36, p.60).
• Derived figures (our arithmetic): the 30x fall in price per GB (228 ÷ 7.5), market-builder and ARPU-hike tools (customers × ARPU × 12), shares of 4G/5G in the base, and ₹ crore conversions (₹ million ÷ 10).

Published 18 October 2026 · Data as of 30 June 2026 · Figures in ₹; FY = year to 31 March.

Disclosures

Research entity: INVESTOLOGY · Rohit Singh · SEBI Registered Research Analyst, INH000015297.

Nature of this page: An educational sector study. It explains how the telecom industry works and what to track. It is not a recommendation to buy, sell or hold any security, and it is not an IPO rating.

Third-party data: Market sizes and forecasts come from a report that Jio Platforms commissioned from Analysys Mason, as reproduced in the RHP. They are not our forecasts.

Investments in securities market are subject to market risks. Read all the related documents carefully before investing. Registration granted by SEBI, membership of BASL and certification from NISM in no way guarantee performance of the intermediary or provide any assurance of returns to investors.

Published on the record.

The reports

The week

The ChartBook

A clearer view of the markets.

Build your foundation

A utility learning to charge like one.

Growth now comes from ARPU

Home broadband is the next leg

The heavy 5G spend is behind

The sector on one page

Data got 30 times cheaper, and India used it

The market is now three private players

Growth has moved from users to ARPU

Home broadband is the open field

How a telecom company earns, in one picture

Spectrum

Mostly fixed

Users × ARPU × 12

Of every ₹100 a customer pays, what is left

The economy that telecom runs on

India's digital economy

How 2016 reset the whole industry

Thirty years of the mobile phone in India

India stopped waiting for new networks

Ten years that remade the industry

What cheap data built

India against China and the USA

How many people use each service

Four measures of headroom

Is India's ARPU too low for its income?

1.28 billion connections, and the room still left

Broadband users, and the 2G base still to move

Where broadband stands: by network, and by place

From more users to more from each user

Mobile broadband ARPU

Faster networks

Smartphones everywhere

More data each month

Steady rules

Four players, three that matter

What each band of spectrum does

Low band

Mid band

C-band

mmWave

Who owns the airwaves

Speed, and what Jio says about itself

The open field: broadband for the home

Home broadband finally took off

Two ways to bring the internet home

Why it matters: fixed wireless brought most of the net growth

Fibre and fixed wireless took over

Why satellite is not the threat it sounds like

Business connectivity and enterprise services

Enterprise digital services: where the money is

Inside the biggest pools

Machines are coming online

Digital services for consumers

Consumer digital services

Entertainment is moving to streaming

Selling India's network playbook abroad

The operators, side by side

Three years, line by line

What could go wrong

Rules can change

Price cutting can return

Technology keeps costing money

Cyber security and privacy

Customers leave

A slower economy

How to read a telecom result

Questions, sources and disclosures

Where every number comes from

Disclosures