Social, Commerce and the UPI Rails, 2016 to 2024

Where this era begins: a rail with no traffic, a network with no data

The previous study in this series, on the smartphone inflection and the Jio shock of 2010 to 2018, ends with cheap data arriving in Indian pockets faster than almost anywhere else on earth. This study picks up a narrower question: once that data was flowing, what did Indians actually do with it, and how did businesses learn to reach them and be found by them. Two surfaces run through everything that follows. The marketing surface, how a business reaches a person, moved decisively onto a small set of apps: WhatsApp for conversation, Instagram and YouTube for attention, a vernacular layer of short-video and regional social apps for the hundreds of millions the English-language platforms undercounted. The discovery surface, how a person finds a business, moved just as decisively onto marketplaces and aggregators that behaved less like stores and more like search engines for their category: Amazon and Flipkart for products, Zomato and Swiggy for food, MakeMyTrip for travel, and by the back half of the era, Blinkit, Zepto and Instamart for anything a household needed inside ten minutes.

Underneath both surfaces sat a rail that had almost nothing to do with either when it launched. The Unified Payments Interface, built by the National Payments Corporation of India and regulated by the Reserve Bank of India, went live on 11 April 2016, with a pilot inaugurated by then RBI Governor Raghuram Rajan in Mumbai and twenty-one member banks on board. Reliance Jio, the network that had made the data itself cheap, would not become publicly available until 5 September 2016, almost five months later. The two launches were not connected at the time. Jio was a telecom price war; UPI was a central-bank plumbing project. What this era did was fuse them: by its end, a resident of a small town could discover a product on a marketplace, decide to buy it after seeing it in a WhatsApp group or an Instagram reel, and pay for it in three seconds over a rail that barely existed when the decade began.

The scale of that fusion is visible in one deal from deep inside this period. In 2020, Reliance's Jio Platforms raised roughly 1,52,056 crore rupees (about 20.52 billion dollars) in exchange for a 32.97 percent stake, from an investor list that included Facebook, the owner of WhatsApp and Instagram, and Google, the owner of YouTube. The two companies that would go on to define the marketing surface of this era bought a direct stake in the pipe that carried it. That is the frame for what follows: a rail, a set of surfaces, and a state that spent the back half of the decade writing rules for both.

UPI becomes the rail everything else runs on

UPI's early growth was unremarkable by the scale it would later reach, and the previous era in this series left it there deliberately: monthly transaction value on the rail was still only around 48 crore rupees in October 2016, a rounding error against what was coming. Major Indian banks began rolling out UPI-enabled apps to customers from August 2016, and adoption built gradually through demonetization that November, which pushed millions of first-time users toward digital payment out of necessity rather than preference. The real acceleration came from third-party apps built on top of the interface rather than from bank apps themselves. PhonePe, founded in December 2015 by Sameer Nigam, Rahul Chari and Burzin Engineer, was acquired by Flipkart in April 2016 and brought its UPI-based app to market in August that year, giving India's largest online marketplace a direct stake in the rail from the very start (Flipkart would fully spin PhonePe off as an independent company in December 2022). Google Pay followed, and the two apps, alongside a slower-moving WhatsApp Pay, became the interface through which most Indians actually touched UPI, even though NPCI itself owned the underlying rail.

The volume that followed is worth stating plainly rather than in the abstract. One lakh crore rupees is one trillion rupees, roughly 12 billion dollars at exchange rates typical of this period. By calendar year 2023, UPI carried roughly 117.7 billion transactions worth about 182.8 lakh crore rupees, on the order of 2.2 trillion dollars. By 2024 that had risen to roughly 140.0 billion transactions worth about 246.8 lakh crore rupees, close to 3 trillion dollars moved over a single domestic instant-payment rail in one year. A single month inside the period gives a sharper sense of the trajectory: in July 2024 alone, UPI processed 14.44 billion transactions worth 20.64 lakh crore rupees, an average of 466 million transactions a day, up 35 percent from the same month a year earlier. By the end of this era, the government and NPCI were routinely describing UPI as the world's largest real-time payments platform by transaction volume, a claim that had become measurably true well before it became a marketing line.

WhatsApp Pay's path through the rail illustrates how tightly regulation, not just adoption, shaped this decade. NPCI approved a phased rollout for WhatsApp Pay in February 2020, starting with a cap of 20 million users, a fraction of WhatsApp's Indian base even then. That cap rose to 40 million in November 2021 and 100 million in 2022, before NPCI finally removed the onboarding limit entirely, letting WhatsApp Pay reach the app's full Indian user base. The caps were a regulator managing concentration risk in a payments system it did not want any single app, however dominant in messaging, to dominate outright. That instinct, keep the rail neutral even as the apps on top of it concentrate, recurs later in this era's politics, in the IT Rules and in ONDC.

The social surface: WhatsApp, Instagram, YouTube and the languages of India

The TikTok shock and the short-video scramble

No single event in this era moved the marketing surface faster than a border clash that had nothing to do with marketing. TikTok had become one of the fastest-growing apps in Indian history by early 2020; the app passed 2 billion downloads worldwide that April, and roughly 30 percent of those downloads had come from India alone, by some estimates its single largest national market. Then, on 15 and 16 June 2020, Chinese and Indian troops clashed in hand-to-hand fighting in the Galwan river valley in Ladakh, the deadliest confrontation on the disputed border in decades, killing twenty Indian soldiers.

Thirteen days later, on 29 June 2020, India's Ministry of Electronics and Information Technology invoked Section 69A of the Information Technology Act, 2000, to block 59 mobile apps it said were "engaged in activities which is prejudicial to sovereignty and integrity of India, defence of India, security of state and public order," citing unauthorized transmission of user data to servers outside the country. TikTok topped the list; ShareIt, Helo, CamScanner, Likee, Club Factory and Shein were also named. It was, at the time, one of the largest single-day app removals any government had ordered against consumer platforms with hundreds of millions of active users. The government widened the ban that September, blocking 118 more apps including the mobile version of PUBG, and formalized a permanent ban on the original 59 in the following months.

What happened next is the part of the story that matters most for a study of marketing surfaces: the vacuum did not stay empty for long, and the companies that filled it moved with a speed that only made sense as a direct response to the ban. Instagram rolled out its short-video feature, Reels, to India in July 2020, weeks after TikTok disappeared, ahead of Reels' global rollout that August. YouTube moved even faster and more pointedly: it launched the beta of YouTube Shorts, a product built explicitly as what TechCrunch at the time called "its TikTok rival," in India first, on 15 September 2020, before any other market in the world, only bringing the beta to the United States the following March. Domestically, Moj, the short-video app from ShareChat's parent company Mohalla Tech, grew into what industry reporting described as India's largest short-video app within about a year, reaching a monthly user base cited at roughly 160 million with a creator community above 50 million. Four different companies, three of them foreign-owned and one homegrown, treated a geopolitical decision as a starting gun, and the shape of Indian social video for the rest of the decade was set in the scramble that followed.

Selling inside the chat window: WhatsApp Business and social commerce

The WhatsApp Business app of January 2018 was the opening move in a longer shift toward what the industry came to call conversational commerce: selling that happens inside a chat thread rather than on a separate storefront page. Over the rest of the era, Meta layered catalog, cart and payment features on top of the basic Business app and its associated API, so that a customer could discover a product, ask a question, and complete a purchase without ever leaving the conversation. For millions of small Indian merchants who had never built a website and had no reason to, WhatsApp effectively became their storefront by default, because it was already the app on every customer's phone.

No company built a bigger business on that default than Meesho. Founded in December 2015 by Vidit Aatrey and Sanjeev Barnwal, Meesho let resellers, overwhelmingly women running a business from home for the first time, share a catalog of products directly into WhatsApp, Facebook and Instagram groups and earn a commission on whatever their network bought, with zero inventory and effectively zero starting capital. By July 2021 the company had reported roughly 250 million registered consumers and a million resellers reachable through that model, at a moment when other social-commerce startups were still explaining the category to investors. Meesho's bet was specific and, in hindsight, correct: in a market where a website was still a foreign concept to many small sellers but a WhatsApp group was not, the chat app itself could be the entire commerce stack.

Payments closed the loop. As the previous section describes, UPI on WhatsApp moved from a 20 million user cap in 2020 to the app's full Indian base by the middle of the decade, meaning that by the back half of this era, discovery, conversation and payment for a huge share of small commercial transactions in India could happen inside a single app, without a browser, a card terminal or a separate payment gateway ever entering the picture.

Marketplaces become search engines, and the festival becomes an economic event

The pandemic, the D2C wave and the ten-minute delivery war

The state writes the rules: intermediary law, data protection and the open network

The same platforms absorbing this much of India's commercial attention drew the state's attention in return, and the second half of this era is when that attention turned into binding law. The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, notified on 25 February 2021, created a new category, the significant social media intermediary, for any platform with more than 5 million registered users in India, and imposed on that category a set of obligations that had not previously existed in Indian law: a resident chief compliance officer, a nodal law-enforcement contact, a resident grievance officer, and, for messaging apps specifically, a duty under Rule 4(2) to identify the first originator of a message inside India when ordered to by a court or competent authority.

That traceability requirement produced the era's sharpest confrontation between an Indian government and a foreign platform. On 25 May 2021, the last day intermediaries had to comply with the new rules, WhatsApp sued the Government of India, arguing the traceability rule was incompatible with the end-to-end encryption at the center of its product and would force it to break that encryption for every message on the platform; the government called the move "clear act of defiance." Weeks later, on 5 July 2021, the government stated that Twitter had lost its intermediary liability protection, the legal safe harbor that shields a platform from being treated as the publisher of what its users post, for failing to comply with the same rules, after Twitter had not appointed the required resident compliance officers. Both disputes were, at their core, arguments over how much control an elected government could exert over platforms that most of the country now used as its default channel for speech, commerce and information.

The Digital Personal Data Protection Act, 2023, extended that regulatory reach from platform conduct to the data those platforms held. The bill passed the Lok Sabha on 7 August 2023 and the Rajya Sabha two days later, receiving presidential assent on 11 August 2023, and made India the nineteenth G20 member to pass a comprehensive personal-data-protection law. It created the Data Protection Board of India to adjudicate breaches, set penalties as high as 250 crore rupees (roughly 30 million dollars) for the most serious violations, and required plain-language privacy notices, adopting a deliberately simple drafting style the government called SARAL. Its detailed rules were not notified until November 2025, triggering a phased rollout running to May 2027, which places full enforcement past the end of this study's period, but the Act itself, passed inside this era, is the legal foundation everything since has built on.

The most structurally ambitious response came from a different direction entirely: rather than regulate the dominant marketplaces, the government tried to build an alternative to them. The Department for Promotion of Industry and Internal Trade constituted a nine-member advisory council for what became the Open Network for Digital Commerce on 5 July 2021, and ONDC was formally incorporated as a non-profit Section 8 company on 31 December 2021, seeded by the Quality Council of India and Protean eGov Technologies alongside a group of public and private banks that each bought a small equity stake. ONDC's premise was structural rather than promotional: instead of another app competing for a share of India's commerce, it built an open, shared protocol that let any seller's app and any buyer's app transact with each other, unbundling discovery, ordering, logistics and payment in a way no single marketplace controlled end to end. The network piloted on 29 April 2022 in five cities, New Delhi, Bengaluru, Bhopal, Shillong and Coimbatore, with its first transaction a Bengaluru seller receiving an order placed through the buyer app Paytm; within roughly a year it had scaled to more than 236 cities and over 45 network participants, averaging more than 13,000 retail orders and 36,000 mobility rides a week, according to figures reported by the Ministry of Commerce. Whether an open protocol seeded by the state can meaningfully dent the concentration built by Amazon, Flipkart, Zomato and the quick-commerce trio is a question this era leaves open rather than settled, and it is one of the threads the next study in this series, on the answer era, AI and machine discovery from 2023 onward, has to pick up directly, because by the last year of this period the same question, who mediates discovery when the mediator is no longer a search box, had already started to be asked about a different technology entirely.