I can think of ways to deal with the Taste & Fit issues but they take a lot of upfront work by the consumer to train the AI. For example, you could take pictures of all your favorite clothes to train the Agent before you send them off to shop. They could then present you with their top finds and you pick the one you want. The Agent then buys it. Given most of these purchases are intermittent, it probably isn't worth it.
In terms of the "customer knowledge" issue, I think the conversational commerce model would help even without closing the loop to the purchase. The purchase is actually the easy part!
In terms of your parasitism point. Agentic Commerce is a modest step beyond Google Shopping: The consumer enters the product they want, and Google already shows them the best options & pricing. AC goes a step beyond that to pick from the options and compete the transaction.
This is one of the most thoughtful and well analysed pieces on agentic commerce I've seen to date. I do wonder if it's worth pondering over what it would take to avoid the bearish outcome where all we get is "agent assisted search". How does one solve the challenges related to autonomous payments to make them a reality?
Also, building on your points around "taste & fit", and "lack of customer knowledge", what is the consumer behaviour unlock we need for autonomous payments to truly become a thing? Any new payment method historically (e.g., credit cards, debit cards, BNPL, UPI, PayPal, contactless cards, Apple Pay, etc.), has won by solving a pain that was previously unsolvable, and by latching onto a habit people already had. New payment behaviour is rarely created; it is borrowed from an existing daily act. It's parasitism of sorts. What might that be for autonomous payments?
Thanks for the comment!
I can think of ways to deal with the Taste & Fit issues but they take a lot of upfront work by the consumer to train the AI. For example, you could take pictures of all your favorite clothes to train the Agent before you send them off to shop. They could then present you with their top finds and you pick the one you want. The Agent then buys it. Given most of these purchases are intermittent, it probably isn't worth it.
In terms of the "customer knowledge" issue, I think the conversational commerce model would help even without closing the loop to the purchase. The purchase is actually the easy part!
In terms of your parasitism point. Agentic Commerce is a modest step beyond Google Shopping: The consumer enters the product they want, and Google already shows them the best options & pricing. AC goes a step beyond that to pick from the options and compete the transaction.
This is one of the most thoughtful and well analysed pieces on agentic commerce I've seen to date. I do wonder if it's worth pondering over what it would take to avoid the bearish outcome where all we get is "agent assisted search". How does one solve the challenges related to autonomous payments to make them a reality?
Also, building on your points around "taste & fit", and "lack of customer knowledge", what is the consumer behaviour unlock we need for autonomous payments to truly become a thing? Any new payment method historically (e.g., credit cards, debit cards, BNPL, UPI, PayPal, contactless cards, Apple Pay, etc.), has won by solving a pain that was previously unsolvable, and by latching onto a habit people already had. New payment behaviour is rarely created; it is borrowed from an existing daily act. It's parasitism of sorts. What might that be for autonomous payments?