Future of Commerce
The future of payments is inextricably linked to the future of commerce, recognizing that no matter how critical the payments industry might be, it fundamentally exists to facilitate the flow of value between parties engaged in commercial activity. The payments industry exists solely to serve commerce; people buying stuff from people selling stuff. The impact of AI on commerce are undeniable, and its a simple exercise to look at our own experiences to see how this may play out. Taking myself to the shops to buy groceries is necessary because there are a bunch of human decisions that need to be made during this process. What do I want to eat? What am I allergic to? What are my health goals? What do my children eat? When is the right time to try out something new? Its super complicated and there is no heuristic system capable of doing this for me, and so between my wife and I we need to solve this problem on a daily basis using our innate human decision making skills. As a result, we invest a significant portion of our valuable time in this task. While some may find the act of shopping, cooking, and experimenting with food exciting, for most of the world, it remains a chore that's difficult to outsource without significant compromises in terms of money, health, or time.
Grocery shopping is one obvious example, but there are many many more. Hunting for my kids next birthday present, a pair of "Blue Nike Dunks", is an investment in at least a few hours of web surfing, looking for availability, styles, sizes and then finally also trying to ensure we're paying the right price, and that the takkies will be delivered to the right address and on the right date. It's a huge investment in high friction activity for something seamingly so simple.
AI will form a part of the system that address these inefficiencies by automating these decision-making processes with the same or better outcomes versus human made, faster and with more information available. For instance, AI systems can learn individual preferences, dietary restrictions, health goals, and even family preferences, to automate shopping lists and make purchase recommendations. Additionally, AI can manage complex logistics, ensuring products are priced optimally, available when needed, and delivered accurately. Eventually we will find ourselves wanting to fully outsource the entire function of "shopping", as these AI systems become better at managing our own lives than we are.
In the specific case of searching for items like shoes, AI could streamline this process by quickly scanning multiple websites, comparing prices, checking availability, and even making the purchase automatically based on predefined mandates, or reach out and ask clarifying questions based on new information. This capability not only saves time but also enhances the overall shopping experience by reducing the friction associated with these tasks. Over time, these AI agents will only improve as they receive more feedback—leading to the possibility that one day I may only ask "please get get my daughter something special for her birthday" and fully trust the AI to make the most optimal decision based on the circumstance. It is scary, but something we are very likely heading toward.
Given the above, it's clear that humans will always have a role to play, but there will be a shift from the "human executing the payment" to the "human providing a mandate to execute the payment", realised in whatever form that AI agent may take. Building this type of interaction ontop of the existing e-commerce and terminal payment rails can certainly work, but I think will present a number of challenges around interface, risk, legal and information accuracy issues—all of which represent significant opportunity for the payments industry and especially those leading the charge in the short-term.
Our role in this
In the coming years, every business worldwide will face a critical juncture: evolve or perish. As AI reshapes industries, introduces new business models, and shifts consumer behaviors, companies will be forced to adapt if they are going to continue to thrive in this new future.
Working across payments and software gives us an interesting view of this shift. We've already seen how applying AI to a specific problem can change the way a business operates.
A prime example of our capability is the notable success achieved within a telematics company managing over a million vehicles. In less than six months, we implemented machine learning technologies that enhanced the company's response to life-threatening vehicle events, resulting in a twentyfold improvement in emergency contact centre operational efficiency. This achievement not only has the potential to boost the company’s operational performance but also represents substantial financial gain in just this one very thin functional area within the business.
Combining that experience with an understanding of payments opens up some interesting possibilities. While the exact trajectory of the convergence between Payments and AI remains uncertain, the potential extends well beyond fraud detection and predictions of consumer behaviour.
In the Payment Terminal sector, the capacity to proactively predict market trends and consumer behaviors is becoming a critical strategic asset. This ability to instantly predict behaviour can significantly influence pricing strategies, allowing for optimization of revenue, reduction in delivery costs, discounts based on loyalty or competitor pricing pressure and even adjustments based on stock quantity levels. Such capabilities are particularly powerful as they become more democratized within the grassroots of our merchant ecosystem.
Even more leading edge is the exploration of a range of more sophisticated opportunities, including the development of AI-driven Consumer Buyers in an e-commerce setting. These sophisticated AI agents could revolutionize the shopping experience by autonomously managing purchases based on user-defined criteria. For example, "I would like to buy a new pair of blue Nike Dunks and I am not willing to spend more than R1500." The AI agent would then take over, handling everything from asking clarifying questions to making the purchase itself, all within the set budget. This is the Uber model for Commerce. Something else does your shopping for you - why should we care how it gets done.
These two simple ideas sit on opposite sides of the spectrum, one with short-term and the other with long-term value. Both are opportunities to rethink how commerce works and what the payments industry needs to make possible.
Article
{ slug: "future-of-commerce"
, originalUrl: Nothing
, title: "Future of Commerce"
, date: constructDate 2024 4 30
, keywords: ["payments","commerce","AI"]
, cover: { alt: "Code", caption: Nothing, src: "/images/covers/code.png" }
, summary: "AI agents, shopping mandates and the future of payments."
, author: { bio: "Towards an AI-first, hyper-scaled, functional reactive universe!", email: "tom@tomwells.org", name: "Tom Wells" }
, body:
[ FlowRichParagraph [ InlineText "The ", InlineEmphasis "future of payments", InlineText " is inextricably linked to the ", InlineEmphasis "future of commerce", InlineText ", recognizing that no matter how critical the payments industry might be, it fundamentally exists to facilitate the flow of value between parties engaged in commercial activity. The payments industry exists solely to serve commerce; people buying stuff from people selling stuff. The impact of AI on commerce are undeniable, and its a simple exercise to look at our own experiences to see how this may play out. Taking myself to the shops to buy groceries is necessary because there are a bunch of human decisions that need to be made during this process. What do I want to eat? What am I allergic to? What are my health goals? What do my children eat? When is the right time to try out something new? Its super complicated and there is no heuristic system capable of doing this for me, and so between my wife and I we need to solve this problem on a daily basis using our innate human decision making skills. As a result, we invest a significant portion of our valuable time in this task. While some may find the act of shopping, cooking, and experimenting with food exciting, for most of the world, it remains a chore that's difficult to outsource without significant compromises in terms of money, health, or time." ]
, FlowParagraph "Grocery shopping is one obvious example, but there are many many more. Hunting for my kids next birthday present, a pair of \"Blue Nike Dunks\", is an investment in at least a few hours of web surfing, looking for availability, styles, sizes and then finally also trying to ensure we're paying the right price, and that the takkies will be delivered to the right address and on the right date. It's a huge investment in high friction activity for something seamingly so simple."
, FlowParagraph "AI will form a part of the system that address these inefficiencies by automating these decision-making processes with the same or better outcomes versus human made, faster and with more information available. For instance, AI systems can learn individual preferences, dietary restrictions, health goals, and even family preferences, to automate shopping lists and make purchase recommendations. Additionally, AI can manage complex logistics, ensuring products are priced optimally, available when needed, and delivered accurately. Eventually we will find ourselves wanting to fully outsource the entire function of \"shopping\", as these AI systems become better at managing our own lives than we are."
, FlowParagraph "In the specific case of searching for items like shoes, AI could streamline this process by quickly scanning multiple websites, comparing prices, checking availability, and even making the purchase automatically based on predefined mandates, or reach out and ask clarifying questions based on new information. This capability not only saves time but also enhances the overall shopping experience by reducing the friction associated with these tasks. Over time, these AI agents will only improve as they receive more feedback—leading to the possibility that one day I may only ask \"please get get my daughter something special for her birthday\" and fully trust the AI to make the most optimal decision based on the circumstance. It is scary, but something we are very likely heading toward."
, FlowParagraph "Given the above, it's clear that humans will always have a role to play, but there will be a shift from the \"human executing the payment\" to the \"human providing a mandate to execute the payment\", realised in whatever form that AI agent may take. Building this type of interaction ontop of the existing e-commerce and terminal payment rails can certainly work, but I think will present a number of challenges around interface, risk, legal and information accuracy issues—all of which represent significant opportunity for the payments industry and especially those leading the charge in the short-term."
, FlowSection "Our role in this"
, FlowParagraph "In the coming years, every business worldwide will face a critical juncture: evolve or perish. As AI reshapes industries, introduces new business models, and shifts consumer behaviors, companies will be forced to adapt if they are going to continue to thrive in this new future."
, FlowParagraph "Working across payments and software gives us an interesting view of this shift. We've already seen how applying AI to a specific problem can change the way a business operates."
, FlowParagraph "A prime example of our capability is the notable success achieved within a telematics company managing over a million vehicles. In less than six months, we implemented machine learning technologies that enhanced the company's response to life-threatening vehicle events, resulting in a twentyfold improvement in emergency contact centre operational efficiency. This achievement not only has the potential to boost the company’s operational performance but also represents substantial financial gain in just this one very thin functional area within the business."
, FlowParagraph "Combining that experience with an understanding of payments opens up some interesting possibilities. While the exact trajectory of the convergence between Payments and AI remains uncertain, the potential extends well beyond fraud detection and predictions of consumer behaviour."
, FlowParagraph "In the Payment Terminal sector, the capacity to proactively predict market trends and consumer behaviors is becoming a critical strategic asset. This ability to instantly predict behaviour can significantly influence pricing strategies, allowing for optimization of revenue, reduction in delivery costs, discounts based on loyalty or competitor pricing pressure and even adjustments based on stock quantity levels. Such capabilities are particularly powerful as they become more democratized within the grassroots of our merchant ecosystem."
, FlowParagraph "Even more leading edge is the exploration of a range of more sophisticated opportunities, including the development of AI-driven Consumer Buyers in an e-commerce setting. These sophisticated AI agents could revolutionize the shopping experience by autonomously managing purchases based on user-defined criteria. For example, \"I would like to buy a new pair of blue Nike Dunks and I am not willing to spend more than R1500.\" The AI agent would then take over, handling everything from asking clarifying questions to making the purchase itself, all within the set budget. This is the Uber model for Commerce. Something else does your shopping for you - why should we care how it gets done."
, FlowParagraph "These two simple ideas sit on opposite sides of the spectrum, one with short-term and the other with long-term value. Both are opportunities to rethink how commerce works and what the payments industry needs to make possible."
]
}