The Networks of Everyday Life

The complete report has seven parts and one 204-page master PDF. The research cutoff for this part is 2 October 2026. Every figure retains its own observation year. Page references in the original reading guide below refer to this volume’s printed pages within the master PDF; use the contents here to move through the web edition.

A country is also the things that work.

An author's note

Part 1 began with money, work and prices. This part starts closer to the front door. Is there a road you can use in the rain? Does the tap run? Can you reach a hospital, make a payment or get home without losing half your day? These are ordinary questions. Together, they tell us a lot about a country.

I wanted this volume to explain infrastructure without making it sound like an engineering textbook. A gigawatt should mean something to the person paying an electricity bill. A kilometre of metro should mean something to the person trying to get to work.

The chart shared by The Matrix India made a striking claim about how much was added after 2014. Some of its comparisons hold up well. Others need a different starting date, a clearer definition or a newer figure. I have kept the original claims visible so you can see where the report agrees and where it changes the comparison.

I also want to keep two things in view at the same time: building something and making it useful. A house on a sanction list is still a promise. A house completed is a physical result. A house that a family can safely live in is the reason for doing the work.

At MisterLove, I write about society, history, public policy and technology in plain language. I'm Lovepreet Singh, founder of Lovelace and an MBA student in Business Strategy and Analysis, working toward doctoral research. This report follows the same habit: take the claim apart, explain it, and leave the evidence where the reader can find it.

I am not asking you to memorise these numbers. I want you to understand them well enough that the next dramatic chart takes a little less effort to read.

Lovepreet Singh
MisterLove · 2 October 2026

Sources: 1, 2.

From the highway to the kitchen tap

Part 2 of the seven-part India report card.

The figures are organised by what the service does. Transport moves people and goods. Power and gas supply energy. Water, sanitation and housing support daily life. Digital networks help people communicate, pay and use services.

Most earlier observations are from March 2014 or 2013-14. Some programmes and surveys began later. Those dates remain visible. The recent column means the latest suitable figure checked for that comparison, not a measurement of everything on the same day.

This is Part 2 of the complete seven-part report. The other parts cover the economy; health and education; farming and industry; crime, institutions and environment; state differences and rankings; and the final UPA–NDA comparison.

The research cut-off is 2 October 2026. A 2026 publication may describe 2025 activity. A target for 2028 is still a target. Each table tells you which kind of date you are reading.

Sources: 2.

A bigger number needs a clear meaning.

The unit, the date and the stage of delivery come first.

A stock is what exists at a date, such as railway length or tap connections. A flow is activity during a period, such as annual cargo or monthly payments. A change in stock is not automatically entirely new construction. A financial year such as 2013-14 runs from April 2013 to March 2014.

One lakh is 1,00,000. One crore is 1,00,00,000, or ten million. MMT means million metric tonnes. MTPA means million tonnes per year of handling capacity. A port able to handle 100 MTPA has capacity; it may actually handle much less cargo that year.

A gigawatt, written GW, is 1,000 megawatts of power capacity. It describes how much power equipment can produce under the relevant conditions. A kilowatt-hour, or one electricity unit on a bill, measures energy used over time. Capacity and annual electricity generation are different.

The Matrix chart's percentage uses the later total as its denominator. If a network goes from 100 km to 200 km, the extra 100 km is 50% of the new total. The usual growth rate is 100%, because the increase is divided by the earlier 100 km. Both calculations are valid; they answer different questions.

That chart formula works best for a consistently measured stock. Annual coal production or defence exports are flows. Their increases are not the share of a physical network built. Recognised startups are an administrative category, and tap-water coverage is a household percentage.

Sanctioned means approved. Under construction means work is progressing. Commissioned means put into service under the relevant reporting rules. Usage tells us what people or goods actually did. We keep those stages separate because each is a different part of the story.

Finally, a change after 2014 does not tell us exactly who caused it. Union and state decisions, local bodies, private investment, earlier plans and technology can all contribute. These comparisons establish recorded changes; they are not a calculation of political credit.

Sources: 2, 3, 7, 15, 17.

The main road network became much longer.

National highway length is one part of India's roads.

A national highway is a road in a designated national network. It is not every road in India. State highways, district roads, village roads and city streets are counted under other headings. The larger later number therefore describes one important network, rather than every route a person can travel.

Imagine a map with 91 km of a particular road category. A rise of about 61% would take that category to roughly 147 km. That gives a feel for the scale. The actual national figure is a thousand times larger.

The network can grow through new alignments, extensions and the declaration of existing roads as national highways. Widening an existing stretch can improve its capacity without adding the same number of route kilometres. So the difference of 55,285 km should not be described as 55,285 km of brand-new road built on previously roadless land.

Road kilometres and lane kilometres also differ. A 10 km four-lane road is 10 road km but 40 lane km. Adding lanes can make construction substantial while the route length remains 10 km. Mixing the two units makes an increase look larger or smaller than it is.

What does this mean in daily life? A useful new or improved connection can shorten a journey and help goods reach a market. The actual gain depends on where it goes, congestion, tolls, maintenance and access from the surrounding area. A bigger national total cannot tell you whether your particular journey became easier.

The later endpoint is the dated ministry report. Network growth includes changes in designation.

Sources: 3, 4.

More lanes and expressways. Different counts.

The road category must stay the same in both columns.

Four-lane and above tells you the road's width in traffic lanes. Access-controlled tells you how traffic enters and leaves it: vehicles normally use planned interchanges rather than every roadside turning. A wide highway and an expressway are therefore not automatically the same thing.

The chart's expressway pair, 1,000 to 6,700 km, was not independently reproduced for a consistent all-India definition in this volume. National corridors are a narrower series than a total that might also include state expressways. The official national figure above cannot simply replace an all-India total without explaining the change in scope.

Construction speed also varies by year. The ministry's annual table reports 12.1 km a day in 2014-15, 36.5 in 2020-21, 33.83 in 2023-24 and 29.21 in 2024-25. The last of those is 10,660 km across the year. The peak rate was not maintained every year.

These annual construction figures can include improvement works, not just entirely new roads. The 2014-15 observation is also after the change of government. It should not be labelled the entire pace of the earlier government.

A faster road can save time when its complete usable stretch connects the places people need. An awarded project, an opened section and a complete corridor are separate milestones. Safe crossings, emergency response, local access and upkeep remain part of the result.

Road series and later dates differ by row. No single expressway total is certified here.

Sources: 2, 4, 5.

A village road can change an ordinary day.

PMGSY counts programme work, including construction and upgrades.

PMGSY is the Pradhan Mantri Gram Sadak Yojana. It began in 2000 to provide all-weather rural road connectivity. Later phases also improve important routes and links to facilities. It existed before 2014 and continued afterward.

The June backgrounder separately reports about 4.11 lakh km completed during 2014-2026. That is work during the later period, not a later cumulative total of only 4.11 lakh km. The July cumulative figure of nearly 7.97 lakh km makes that distinction clearer.

Programme work is not the same as the length of a unique national rural-road network. An upgraded road can be counted as work even though a route already existed. Different programme phases also have different purposes. Adding all their work lengths does not prove that every kilometre was a first-time connection.

Think about a farm connected to a market by a track that becomes unusable during heavy rain. A durable road can make the trip more predictable. It can also help a school bus or ambulance reach the settlement. The benefit comes from the usable connection, not from the sign announcing the project.

Eligibility matters. A habitation is a settlement, and programme rules use population thresholds and special provisions for some difficult areas. “All eligible habitations connected” would not mean every small settlement in the country has every kind of road. Maintenance and bridge condition determine whether the connection stays useful.

Sources: 6, 7.

Most broad-gauge routes now have electric wires.

Electrifying a route is different from building a new one.

An electric locomotive receives power through equipment above the track. Electrification provides that system on a route. The railway can then run suitable electric trains there. A route may be decades old even when its electric equipment was installed recently.

Route km measures the distance between places along the network. Track km also counts parallel tracks. A 100 km route with two running tracks is still 100 route km, although it has 200 running-track km. Electrification statistics must not be silently compared with total track length.

The displayed electrified length is about 3.2 times the ministry's earlier baseline. The increase is about 48,072 route km. That is a major change in the type of traction the network can support. It is not a claim that the whole railway became 3.2 times longer.

There is a historical reporting difference worth keeping visible. The older revised railway statistical summary records 21,614 electrified route km for 2013-14. The 2026 backgrounder uses 21,801 for its retrospective 2014 comparison. The table above retains the later ministry vintage; it does not pretend both historical totals are identical.

Electric trains can reduce direct diesel use and support efficient operation. The wider environmental effect depends on how their electricity is generated. Wires alone do not settle punctuality, ticket availability, crowding, track quality or safety.

Broad gauge describes a particular distance between rails. The 99.6% figure applies to that network. It should not be expanded into a claim that every railway, every route or every train in India has become electric.

Sources: 8, 9, 17.

More freight. Fewer passenger journeys.

Physical use belongs beside the infrastructure total.

The earlier figures come from the revised railway statistical summary. The recent figures come from the ministry's annual release. On these reported measures, freight is about 59% higher, while passenger journeys are about 12% lower. The recent headline data are not a substitute for the final detailed statistical accounts.

A passenger journey is a trip, not a unique person. Someone making repeated trips can be counted repeatedly. Railway passenger counts also combine different kinds of travel, including frequent suburban journeys and longer-distance trips.

A lower total does not prove that every type of railway travel declined. Nor does it, by itself, establish why the total changed. To understand the cause, we would need comparable suburban and non-suburban data, passenger kilometres, service availability, fares and changes in reporting.

The historical series shows the pandemic's disruption plainly: 808.6 crore journeys in 2019-20, 125 crore in 2020-21 and 351.9 crore in 2021-22. A later recovery from that low point can be real while the level remains below an older year. Both facts belong in the same explanation.

Freight tonnes count how much was carried. Tonne kilometres also count how far it travelled. Carrying one tonne for 1,000 km is a different amount of transport work from carrying it for 100 km. Higher freight loading therefore needs distance and commodity mix for a fuller interpretation.

For a traveller, the practical questions remain simple: can I get a seat, can I afford it, and will the journey be safe and reasonably dependable? A larger capital budget or more electric routes cannot answer those questions on its own.

Earlier freight: revenue-earning tonnes. Later figure: ministry headline loading; final detailed accounts may refine comparability.

Sources: 9, 10.

Routes for goods trains.

A real addition, with a history that began before 2014.

Freight means goods. A dedicated freight corridor is a railway designed mainly for goods trains. Separating much of that traffic can reduce competition for space on conventional lines that also carry passenger trains.

The August 2026 parliamentary answer says both corridors were commissioned. The September announcement describes the last links being joined. These are dated accounts of completion. Their 2,843 km is the stated combined length, not all Indian freight routes.

The zero means no commissioned kilometres of these corridors in the cited before-2014 account. India already had freight trains and goods routes, and preparation for this project had begun.

The project company, DFCCIL, was registered on 30 October 2006. Planning, finance, land acquisition and construction have different timelines. Earlier preparation and later commissioning both belong in the history.

Imagine trucks and commuter buses competing on one road. A separate route for many trucks can help both services if its junctions and connections work. Dedicated freight capacity follows a similar idea.

The practical tests are dependable delivery, useful factory and port connections, costs and actual use. The chart's 100% is calculated from zero commissioned kilometres. It does not measure the share of planning, spending or preparation that occurred after 2014.

Sources: 8, 11, 12, 13.

Metro reached more cities.

A network can expand without serving every neighbourhood.

Metro is urban rail designed for frequent travel within a city and its connected areas. The reported route length is about 4.7 times the 2014 level, an increase of 907 km. The Matrix chart's rounded 1,000 km recent endpoint is below the later official snapshot used here.

Cities covered tells us how widely the system has spread. It does not tell us how much of each city has a useful connection. A small operating section and a large connected network can each put a city into that count.

Ridership normally counts trips or entries under the operator's method, not unique people. A commuter travelling to work and back can generate two trips. The national daily total also needs its reporting date and operator coverage; it is not a household survey of how many Indians can afford metro travel.

Consider a 15-minute train journey that requires a 30-minute trip to the first station and another 20 minutes from the last station to work. The whole commute is 65 minutes, before waiting. This is an example, not a measured national average. It explains why feeder buses, walking routes and interchange design matter.

A metro can offer dependable travel where demand and connections support it. Its usefulness also depends on fares, operating frequency, accessible stations, safety and maintenance. A completed line that few people can conveniently reach has a different value from a well-used one.

Keep regional rapid transit, ordinary metro and water-based urban transport labelled separately where the source does. They can all improve mobility, but a kilometre of one is not automatically interchangeable with a kilometre of another.

The recent daily ridership is a reported 2026 snapshot, not a matched full-year average.

Sources: 5, 6.

More places can connect by air.

An operating airport is not a promise of frequent cheap flights.

The operational-airport total is higher by 91, or about 123%. That does not mean 91 entirely new airports were all built from scratch. A total can change through new facilities, revived facilities and changes in operating status or coverage.

UDAN is a regional air-connectivity programme launched in 2016. It supports selected routes to improve access to underserved or unserved places. Its route count is a different measure from the total airport count.

The UDAN snapshot also reports over 1.68 crore passengers cumulatively. Cumulative means across the programme's operation up to that date. It is not 1.68 crore passengers in a single year. A person taking several flights can be represented several times.

Heliports and water aerodromes serve different aircraft and operations from conventional airports. The 95 connected sites should therefore keep all three categories in the label. Adding them to the national airport total would risk counting overlapping facilities twice.

A route that was operationalised at some point is not automatically a route running every day now. Airlines, schedules and services can change. Before judging regional access, look at current flights, continuity, the fare people actually pay and the time needed to reach the airport.

A useful local flight can save a long road trip to a distant airport. For someone unable to afford a ticket, the airport alone does not create affordable access. Infrastructure and the travel experience need to be assessed together.

The chart shows 164 airports. The selected later official snapshot reports 165; the date is kept visible.

Sources: 14.

Ports can handle more. They also handled more.

Capacity and actual cargo are separate rows.

A port is where ships load and unload. Capacity is the amount it can handle annually under the source's method; MTPA means million tonnes per year. Throughput is the cargo actually handled. One is a capability, the other an activity.

On the matched figures, capacity roughly doubled: about 101% growth. Actual cargo increased by about 72%. Both are substantial increases. They are different because the ability to handle cargo is not the same as the quantity arriving to be handled.

The Matrix chart uses 800 to 2,760 MTPA. Its starting point is not the all-ports 2013-14 capacity in the matched ministry table. Comparing an older major-ports figure with a later all-ports figure would change the basket. This volume uses all ports in both columns instead.

A port can be physically capable of handling more cargo while ships still wait for a berth, paperwork or a road and rail connection. Faster unloading can also be undermined by slow movement outside the port. The full delivery chain matters.

A ship-turnaround figure is a measure of time, but averages can cover different vessel types and operations. One must keep the scope and period consistent before comparing them. Capacity alone is easier to measure than the complete experience of moving goods from a factory to a customer.

More cargo is not automatically more domestic value added. Ports also handle imports, raw materials and different kinds of bulk freight. The cargo mix is needed before making a claim about the value of what India produces.

Sources: 15.

More freight moves on rivers and canals.

A declared waterway and an operating one are different.

Inland waterways are navigable routes within the country, including rivers and canals. National Waterway status identifies a route under the legal and development framework. It does not make that route commercially usable all year.

An operating route needs a usable channel, vessels, terminals and land connections. Water depth, river conditions and maintenance affect what it can carry. A legal declaration is an earlier stage than dependable transport.

The Matrix chart compares five earlier waterways with 32 later ones. Five is the earlier declared count; the matched table gives three operational waterways in 2013-14. Compare declared with declared, or operational with operational.

The chart's earlier cargo figure of 29 million tonnes differs from the roughly 18 million tonnes in the matched table. The dated pair used here shows about twelve times the earlier cargo, with the rounded baseline visible.

Water transport can suit heavy goods when the route and terminal connections fit the shipment. Distance from the river, loading time, deadlines and seasonal reliability matter. It is not automatically the best option for every product.

Cargo tonnes do not measure the full environmental effect or the share of physical waterways built. Ferries, tourist cruises and freight are different uses. Adding their counts would not give one meaningful transport total.

The matched 2013-14 cargo baseline is rounded to 18 million tonnes in the parliamentary table.

Sources: 6, 15.

India can generate much more power.

Installed capacity is the equipment total, not the electricity bill.

The later total is about 2.23 times the earlier level, or roughly 123% higher. That is a comparison of installed generation capacity. It is not a claim that electricity generation or every household's supply grew by exactly the same percentage.

Think of capacity as the kitchens available, and generation as the meals actually produced. Some kitchens operate often; others operate only at particular times. Equipment and output must be counted separately.

The CEA's August 2026 table gives 250.21 GW of fossil-fuel capacity and 304.334 GW of non-fossil capacity. Non-fossil includes renewable sources and nuclear power. It is broader than renewable energy alone.

By capacity, the table reports 54.88% non-fossil and 45.12% fossil. Coal alone is 40.47% of the total. Those are equipment shares. They do not say that 54.88% of all electricity generated during the year came from non-fossil sources.

A solar plant cannot generate its rated power continuously through the night. Other plants also have outages, fuel limits or maintenance needs. Supply has to match demand at the moment it is needed, using generation, storage, transmission and operating arrangements together.

Installed equipment can grow while a locality still has interruptions because of distribution faults. Better operations can also improve supply without as much new generation equipment. The chart shows a rounded 540 GW; this volume uses the dated August CEA total.

Sources: 16, 17.

Solar grew especially fast.

Keep large hydro in or out consistently.

The solar total is about 59.5 times the earlier value. A tiny starting point makes a large multiple possible, but the absolute addition also matters: about 165.2 GW. Wind is about 2.8 times its earlier level. Not every renewable technology grew at the same pace.

Renewable sources include solar, wind, hydro and forms of bioenergy under the official categories. Large hydro means larger hydroelectric projects; whether it is included changes the total substantially. In August 2026, the CEA figure excluding large hydro is about 243.49 GW, compared with 295.55 GW including it.

The older 76.38 GW is the retrospective inclusive baseline in the June 2026 government backgrounder. Solar and wind use MNRE's historical series. These source vintages remain labelled, rather than being silently replaced with older differently classified totals.

The chart's renewable pair, 54 to 248 GW, was not reproduced on a clearly consistent basis here. It needs dates and an inclusion rule for large hydro. It should not be read as a universal total of all clean-energy equipment. Nuclear power is non-fossil, but it is outside the renewable total used above.

Large additions to solar capacity create more ability to generate electricity when conditions allow. Turning that into dependable supply throughout the day also requires grid connections, flexibility and storage or complementary sources. A target or a tender award is not installed capacity.

Rooftop solar can reduce purchased electricity. The outcome depends on sunlight, system size, costs, maintenance and billing rules. A national capacity total does not mean every roof has panels or every home has a zero bill.

GW figures converted from MW. Current renewable totals include the categories shown in CEA/MNRE reporting.

Sources: 5, 17, 18.

Longer hours of electricity.

Connections, supply and reliability are different measures.

The March 2026 reply gives these later averages. Rural supply rose by 10.1 hours. It calls the earlier year FY2015, or 2014-15; the backgrounder calls it 2014 and supplies the urban baseline. These figures do not promise uninterrupted supply to each household.

Averages hide differences between states, seasons and localities. A business affected by brief interruptions needs more than a total-hours measure. Voltage quality and repair time also matter.

Transmission moves large amounts of electricity over longer distances. Distribution carries it through local substations, transformers and lines to consumers. A strong national grid can still meet a weak local link.

Circuit km is a counting unit for transmission circuits, not simply the distance along a map. For example, two circuits running over the same 100 km route make 200 circuit km. The table covers lines at 220 kilovolts and above; it does not include every local wire.

Transmission crossed five lakh circuit km in January 2026. The release gives 71.6% growth since April 2014. The chart's roughly 42% divides the increase by the later total instead. The different denominators answer different questions.

A household connection gives access to the network. Regular supply, an affordable bill, safe wiring and prompt repairs make it useful. The February 2026 backgrounder repeats an older urban average of 23.4 hours; the March reply gives 23.6 for FY2025. The dated versions are not identical.

Supply hours: FY2025 in the March 2026 reply, not a new household survey. Transmission milestone: 14 January 2026.

Sources: 19, 20.

Gas reached more households.

PNG is piped natural gas; LPG usually arrives in cylinders.

Long-distance pipelines carry gas to consuming areas. City-gas networks distribute it to homes and businesses. A pipeline passing an area does not mean every home there has a kitchen connection.

The recent pipeline total is about 79% above the historical ministry figure. It comes from PNGRB's March 2026 publication, but that particular table explicitly dates the pipeline data to December 2025. The publication month must not replace the observation date.

A domestic PNG connection provides a fixed pipe to a household. It is different from a commercial or industrial connection. The earlier approximate reach of 25-26 lakh homes is a figure stated in the March 2026 Prime Minister's speech, rather than an independently reconstructed household survey.

PNGRB records 1,69,49,377 domestic connections for March 2026, of which 1,07,25,930 were billed connections: about 63%. The regulator distinguishes having a connection from being billed for use. The exact administrative meaning matters; the gap cannot automatically be called permanent abandonment.

The August 2026 ministry release also describes converting unbilled connections into working connections. Its later 1.74 crore total is a connection-stock figure. This report does not invent a later billed-use percentage from that stock.

Piped gas can remove cylinder deliveries where service is dependable. LPG provides another route to cleaner cooking where PNG is unavailable. For either fuel, costs, safety and supply matter. A connection alone does not prove that a household can afford enough fuel or has stopped using other fuels.

Pipeline observation: December 2025.

Sources: 21, 22, 23, 24.

A tap is a start. Water is the service.

The correct programme baseline is August 2019, not 2014.

Reported coverage rose by 65.52 percentage points. About 82 of every 100 rural households were recorded with taps in the update, leaving about 18 outside coverage. States and Union Territories report these figures; they are not a fresh survey of water from every tap.

The earlier 16.72% is from 2019. Putting it under 'before 2014' moves it back five years and changes the story. This report keeps the correct programme date, rather than implying the entire recorded addition began in 2014.

The separate 2024 functionality assessment examined selected villages and households. Among households with taps in that sample, 87% had received water during the previous week, 80% received at least 55 litres per person per day, and 76% met the combined functionality conditions. Those conditions concern quantity, quality and regularity.

Fifty-five litres covers daily household needs, not just drinking. For four people, that corresponds to 220 litres a day. A pipe supplying too little or unsafe water provides a different service from a dependable tap.

The assessment's 98.1% connection figure applies to its surveyed villages. It is not India's national coverage. Nor can we multiply the 2026 national connection percentage by the 2024 sample's functionality percentage and call the result a measured national rate for 2026.

Wider access is a substantial reported change. Its everyday value depends on whether water arrives when needed, meets quality standards and can be supplied reliably through changing seasons.

Connection coverage and the independent functionality sample answer different questions and refer to different years.

Sources: 25, 26.

Building toilets changed access.

Using them and safely handling waste complete the job.

The programme count shows a large construction effort. One household toilet can serve several people, so twelve crore toilets is not twelve crore people. The cumulative count is also not a survey of every facility's present use.

The 100% refers to reported rural coverage and 2019 open-defecation-free declarations under programme rules. Open defecation means going outside without a toilet. A declaration does not establish that every Indian has used a safe, maintained toilet ever since.

NFHS measures people in households using improved sanitation, across rural and urban India. Improved facilities separate waste from human contact under its definition; shared facilities are treated separately. This survey percentage cannot replace rural programme coverage. It rose by 21.7 percentage points from 2015-16 to 2019-21. The periods show changed living conditions, not a measurement for 2014 or October 2026.

A usable toilet needs water where the design requires it, privacy, maintenance and safe arrangements for waste. Emptying a pit or tank unsafely can move the health risk to workers or nearby land and water. Access, use and safe disposal are connected, but each needs its own evidence.

Sanitation programmes existed before Swachh Bharat. Zero under a new scheme only means that scheme had not begun. The comparison concerns expanded delivery and use, not the invention of toilets in 2014.

Programme coverage is rural. NFHS figures shown here are all India; they are not a like-for-like rural coverage series.

Sources: 5, 27, 28.

More homes completed.

Approvals and completed homes are different counts.

IAY means Indira Awaas Yojana, the earlier rural housing programme. PMAY-Gramin restructured support from 2016. The row compares annual programme completions, not all homes built; rules and reporting also differ. Annual completions are about 2.45 times the earlier count. A finished home may have been approved earlier. The 26.57 lakh count does not mean every house was approved, started and finished within twelve months.

The August reply lists 4.18 crore houses in allocated targets, 3.91 crore sanctioned and 3.12 crore completed. Targets assign work; sanctions approve it; completions record a later stage. The roughly 79 lakh gap between sanction and completion is not additional completed housing.

The urban milestone combines PMAY-U, launched in 2015, and PMAY-U 2.0, introduced in 2024. The release describes over one crore homes completed or delivered. This wording does not establish an independently surveyed count of homes occupied today.

These totals cover programmes, not the whole housing stock. Rural and urban schemes count homes under different arrangements. Private construction and older housing programmes also existed. Zero under a newly launched scheme is not zero homes in India.

A completed home becomes useful when it is safe and connected to essentials. Location, water, electricity, sanitation, transport and maintenance shape that experience. Construction counts explain one stage; they cannot settle every question about living conditions.

Cumulative totals cover multiple years.

Sources: 29, 30, 31.

Internet use grew fast.

Subscriptions count connections; they do not count unique people.

Broadband meets an official minimum speed definition. That threshold changed from 512 kilobits per second to 2 megabits per second over the period. Subscriptions expanded greatly, but the measurement did not stay completely unchanged.

A subscription is a connection or account under telecom reporting rules. One person may have two mobile connections and a home line. Several people may share one device. Therefore 101 crore broadband subscriptions does not mean 101 crore different Indians each have their own dependable internet.

M2M means machine-to-machine: connections for devices such as meters or trackers. TRAI includes about 9.09 crore in its broader August total. The main comparison excludes them from the recent total; the broader total is labelled separately.

Data use measures the amount transferred, not speed or quality. On the decimal convention, 1 GB is 1,000 MB. Moving from tens of MB to tens of GB marks very different use. The average does not describe every subscriber.

TRAI reports average operator revenue of about Rs 7.51 per GB in the March 2026 quarter. This is not the price of every customer's plan. Retail plans bundle amounts, validity and other services.

Living in a covered district does not remove weak signals, phone costs or difficulty using a service. The next page looks at household access and payments alongside the network totals.

August broadband data were released on 28 September 2026. The data-use observation is for the March 2026 quarter.

Sources: 32, 33, 34.

Using the digital network.

Transactions, households and people are separate measures.

The survey asks whether access exists in the home. It does not say every member owns or controls a phone, or can complete digital tasks alone. Its percentages describe 2025, not October 2026.

Among people aged 15-29, the survey reports internet use during the previous three months by 92.7% in rural areas and 95.7% in urban areas. Those figures concern young people in the sample. They do not describe all ages or establish that everyone can use every online public service.

UPI, the Unified Payments Interface, allows payments between accounts through participating services. It began in 2016. A 2014 UPI total is therefore 'not yet launched', not evidence that electronic payments or banking transfers did not exist before 2014.

NPCI's August 2026 table records 24,508.96 million transactions. Ten million make one crore, so that is 2,450.896 crore transactions. Its value figure, Rs 29,82,355.95 crore, is about Rs 29.82 lakh crore. Count tells us how many payment transactions occurred; value tells us how much money those transactions moved.

The same money can move more than once. A payment from one person to a shop and a later payment from that shop to a supplier are both transfers. Adding the payment values does not measure newly created wealth, government spending or GDP.

A useful digital service needs a workable device, connectivity and a way to understand and complete the task. Someone who needs help should also be able to get it. More transactions show widespread use of the system; they do not prove that access is equally easy for every person.

UPI is a monthly flow, not a count of unique users. August 2026 was the latest monthly table verified for this volume.

Sources: 35, 36.

Keep the claim beside the source.

Original chart pairs first; the correction or update follows.

The chart's percentages divide the increase by the later value. They are not the usual growth rate. For example, 250 becoming 1,000 means 75% of the later stock was added, while the stock grew by 300% from its starting level. Neither calculation alone proves how much entirely new infrastructure was physically built.

MMT means million metric tonnes. A missing matched pair is kept visible rather than filled with a guess.

Sources: 2, 5, 8, 11, 12, 15, 17, 18, 22, 23, 24, 25.

Check what each total counts.

The remaining infrastructure rows, plus the scope of later parts.

The chart also lists startups (500 -> 2.30 lakh), defence exports (Rs 700 -> 38,424 crore), AIIMS (8 -> 23), MBBS seats (51,348 -> 1,39,000), medical colleges (387 -> 844), steel (109 -> 222 MT) and coal (565 -> 1,000 MT). These seven claims belong to Parts 3 and 4. They are recorded here, not certified by the infrastructure checks.

The rows above complete this volume's check of all 16 infrastructure claims. Dates, definitions and stages of delivery explain the differences; the full series will examine the remaining topics in their own context.

MTPA is annual handling capacity. MT here means million tonnes; the steel and coal claims still need their own definition checks.

Sources: 2, 5, 8, 9, 14, 15, 16, 17, 20, 21, 22.

Sources and reading notes