- The Unified Agent Protocol aims to enable AI agents to transact on UPI for users in India
- Agentic commerce could replace traditional search and clicks with instant AI-driven purchases
- AI agents may provide richer attribution by recording all options considered in transactions
Somewhere in the corridors of the National Payments Corporation of India, a quiet piece of plumbing is being laid that could rearrange the economics of persuasion.
The proposed Unified Agent Protocol (UAP) would create a common trust layer through which AI agents, registered, verified and authorised, transact on UPI on behalf of their human principals. If it materialises as envisioned, India could become the first country with national infrastructure for agentic payments. Grocery replenishment, bill payments and other low-consideration purchases could increasingly be discovered, compared and bought by software while the human attends to life.
ALSO READ: Agentic AI Is Here, 'Physical AI' To Power Next Chain Of Growth, Says Nvidia CEO
Immediately it suggests what will begin to disappear.
The search. The scroll. The shelf. The click.
For three decades, the rise and rise of digital advertising has revolved around a single human gesture and a single assumption that the last touchpoint before purchase deserves the credit. Agentic commerce compresses discovery, comparison and purchase into seconds.
The click loses its privileged place.
The decline of the click
Attribution was always a polite fiction. Among the cognoscenti, last-click logic was less a science more a settlement. A nervous truce among parties who each wanted credit and agreed to award it to whoever stood nearest the till. The model assumed a human journey with visible touchpoints be it an ad seen, a review read, a link clicked or a cart filled.
An agent can collapse those steps into a single transaction executed in milliseconds. Entire categories may soon have no meaningful customer journey left to map.
Everyone will claim the sale regardless. The agent platform will claim it because its algorithm made the decision. The marketplace will claim it because it held the inventory and sold the retail media placement. The publisher will claim it because its reviews and rankings shaped the machine's preferences. The brand will claim it because someone, somewhere, once chose to trust it. All four will be partly right.
ALSO READ: The Hidden Cost Of Deploying Agentic AI For The Wrong Processes
Credit migrates to whoever influences the agent's consideration set. The discipline shifts from measuring clicks to measuring presence. Was the brand included in the shortlist the machine assembled and why?
Ironically, agentic commerce could produce richer attribution than today's ecosystem. An AI agent can record every option considered, every product rejected and every reason for doing so. Ownership, access and interpretation become the real battleground.
The UAP presents an intriguing possibility. If its architecture eventually includes standardised records of verified agents and their transactions, it could become one of the most neutral sources of attribution available, sitting outside the walled gardens that have marked their own homework for two decades. The protocol was conceived for payment trust. It may also reshape measurement.
Two audiences, one budget
The deeper shift is philosophical. Agentic commerce splits advertising's audience in two. The human still writes the brief. The machine increasingly executes it.
Machines read evidence before they read emotion. Structured product data, verified reviews, pricing integrity, stock availability, service records and delivery reliability become persuasive assets. These have long belonged to catalogue and operations teams. In the agentic era they become strategic advantage. The catalogue becomes creative infrastructure.
A new discipline will emerge alongside search optimisation. Brands will optimise for AI agents, ensuring products are consistently understood, retrieved and recommended across machine ecosystems.
Humans require something older and far harder. Brand strength powerful enough to be named in the prompt.
Consider three briefs.
"Order Surf Excel."
Memory wins.
"Order the best-value detergent."
Price wins.
"Order whichever detergent removes children's stains best."
Proof wins.
Those three prompts describe three different economies of marketing.
Bill Bernbach argued that a communicator must concern himself with unchanging man, whose instincts took millions of years to form and will take millions more to vary. He remains right about half the audience. The other half changes with every model update.
The paradox at the heart of it
Agentic commerce delivers a remarkable paradox. Brand building grows more valuable precisely as traditional advertising becomes less visible.
Byron Sharp argued that brands grow through mental and physical availability. The agent era gives both a literal address. Mental availability lives inside the consumer's prompt. Physical availability lives inside the machine's data feed.
Enduring brands become the instruction itself. Everyone else enters the comparison table.
What Indian marketers should do now
Treat product data as media. Audit the feed with the rigour once reserved for the television commercial.
Invest in verifiable trust. Reviews, service records, fulfilment performance and product accuracy increasingly shape machine recommendations. Machines reward evidence.
Defend the name. Every rupee of brand building should answer one question. Will the consumer type us into the brief?
Interrogate every attribution claim with fresh scepticism. When the platform that decides also reports, demand transparency.
India has a habit of leapfrogging. UPI skipped the card era. The UAP could allow India to leapfrog fragmented agentic commerce standards emerging elsewhere and become the world's first large-scale laboratory for machine-mediated demand. The advantage will belong to marketers who rehearse before the curtain rises.
ALSO READ: Navigating The Next Frontier: Agentic AI In Financial Services
Advertising has been declared obsolete before, by the remote control, the ad blocker and the disappearing cookie. Each wave pushed the discipline upstream, closer to memory and meaning. Preference always came first. The click simply revealed it.
In the agentic era the advertisement may go unseen. The purchase still begins with memory. Memorable enough to be typed. Trusted enough to be transacted.
When the machine buys, memory becomes the first point of sale.
Shubhranshu Singh is a Fellow of The Marketing Academy, London. He serves as a global board member of the Effies Lion foundation, New York. He is a member of SEBI's Apex and Executive Committee for project Jagruk.
Disclaimer: The views expressed in this article are solely those of the author and do not necessarily reflect the opinion of NDTV Profit or its affiliates. Readers are advised to conduct their own research or consult a qualified professional before making any investment or business decisions. NDTV Profit does not guarantee the accuracy, completeness, or reliability of the information presented in this article.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.