Composable Commerce Meets AI Agents: A Survival Guide

The Gist

• Product data matters. Poor product data limits visibility to autonomous agents and impacts product discovery.

• Dynamic pages win. Generative AI can create product pages tailored to specific user queries in real time.

• Composable commerce essential. API-first and modular commerce systems are needed to support AI-driven shopping.

Over the past few months, I’ve spent much of my time in deep contemplation about the future we’re all likely to see come to light over the next few years as it pertains to the invasion of AI in essentially everything we do.

From “Where should I eat tonight?” to “Why does my knee hurt after playing a soccer game?”, there’s no question that LLMs like ChatGPT are becoming part of daily life. They provide answers to curious questions, offer advice on dating and inspire our dreams and aspirations.

However, there’s something a bit more foundational occurring that you may have considered, as well.

These human-like AI interfaces are learning about us, who we are, what we like and what we feel. All of this will be used for what’s to come. This is a shift. It’s a new way of living where our collective consciousness starts to interact with machines.

Of course, I’m talking about autonomous consumer agents. An autonomous agent refers to the capability for tools like ChatGPT to become proactive in their decisions and recommendations. We see a glimpse of this today with OpenAI’s operator, which, as OpenAI describes it, “can be asked to handle a wide variety of repetitive browser tasks such as filling out forms, ordering groceries and even creating memes.”

Ordering my groceries for me? Sign me up.

While many of us have started integrating tools like ChatGPT into our daily lives, ChatGPT has been learning about us. This information will prove to be vital as it starts to purchase items or services for us autonomously. This pivot represents a massive shift in how brands need to rethink their ecommerce shopping experiences as more autonomous agents come to complete tasks and purchases on behalf of their human overlords.

What does this mean for marketing and customer experience leaders? Here are three ways your brand can prepare for the shift toward autonomous agents that may arrive sooner than expected.

Table of Contents

Product Data Needs an Overhaul

Have you ever heard the saying, “Garbage in, garbage out”? That applies here. Everyone is jumping at the opportunity to incorporate AI into their brand. In doing so, most are forgetting that the key lifeblood to AI is data, specifically product data. If a consumer agent is being sent to your site from a tool like ChatGPT to find a product, there’s no way in hell it will return results if your product detail page is lacking key information like enhanced descriptions, inventory levels and reviews.

Now, I know what you’re going to say. “Our product descriptions are solid.” But how can you make sure your descriptions align with all your personas? Someone might now find you through tools like ChatGPT, and you may not know their use case.

Related Article: Are You Ready for Agentic AI Shoppers as Customers?

Generating Product Pages in Real Time

To solve this potential mismatch between product and use case, consider using AI. It can generate product descriptions and pages on the fly for each individual search or query that comes in.

Have you lost your mind?!

Not quite. Hear me out. Tools like ChatGPT can be directed to certain pages or URLs where brands can use generative services to build a new page with unique content, images, banners and pricing each time a human enters a question into the tool.

Here’s an example. If I ask ChatGPT to show me point-and-shoot cameras under $300 that can be delivered tomorrow for a trip to the Sahara desert, it will search the web for results that match key elements. The price is under $300. The product can be delivered by tomorrow. And the camera is likely reviewed or built for shooting desert landscapes.

Today, most product detail pages may have two of those three elements, but most don’t have the actual desert use case as part of the page. The solution is using generative AI tools to create pages on the fly that can be directed into the query mentioned above. That’s the future. Every time a LLM like ChatGPT is trying to find products for its user, it should be directed to a brand site that is set up to accept the question and generate a page that directly speaks to the needs of the user. 

Related Article: Can Traditional Web Development Survive AI?

Why Composable Commerce Is Critical

The only feasible way to achieve much of the future state above will be through API-first composable commerce solutions. Let me break that down for you.

API-first essentially means that brands need to build connections to tools like ChatGPT and others before building any other required code to make the solution successful. This approach makes the connection more appropriate and clear.

Composable commerce focuses on breaking down components of an ecommerce solution into bite-sized pieces that can be worked on independently of the overall solution. Breaking out elements like the cart, checkout or product detail pages gives brands the ability to do tasks like I mentioned above. They can write product detail pages on the fly, for example. 

As time progresses, other AI channels like phygital will come into play. Phygital refers to the digital and physical worlds blending to create new experiences, new shopping opportunities and a new world for us to interact with. Brands must focus on creating commerce experiences that aren’t tied to a screen or a storefront.

Composable AP-first commerce will give brands the firepower needed to build shopping experiences wherever, whenever and however their customers desire.

The Shift Brands Can’t Ignore

There are not many times in history where you can say, “We’re at a massive turning point,” but I believe I can safely say that we are quickly approaching that pivotal point in history. Similar to the dot-com boom, this AI boom will create a massive wave of doing things in new ways.

Consumers will soon rely on these autonomous agents to perform tasks like making purchases, and these purchase opportunities for brands will only be there if they can pivot to generate data in ways these consumer agents can digest it. Speed, accuracy and relevance are all key elements that need to be validated quickly.

As Bob Dylan once said, “The times they are a-changin,” and your brand will need to change to adapt to this new reality of shopping and product discovery. The last thing you want is for consumer agents to visit your site looking for products, yet the experience yields complete unknowns.