Dialogue with Pattern: From Multi-Platform Operations to AI Shopping, What's Next for Global Brands?

[Ebrun Original] On September 17, Pattern brought its Accelerate conference to China for the first time. "The future for brands is not a single-choice question, but how to make different channels work together to drive growth," said John LeBaron, Chief Revenue Officer of Pattern, at the Accelerate 2026 China Cross-Border Ecosystem Acceleration Conference. Now, managing a market is rarely about a single platform. Meanwhile, the rise of AI could further change existing operating models. In his speech, John further mentioned that in the past, brands mainly competed for rankings in search results, but "in the future, brands will also need to compete for visibility in large language model responses." This means that brands now have to consider not just where to sell, but also how to allocate resources across channels, manage inventory, coordinate advertising, content, and fulfillment. Pattern itself is adapting to these changes. The company started in the Amazon ecosystem and now operates across over 70 e-commerce platforms globally. At the conference, Pattern showcased its capabilities in data, advertising, content, logistics, fulfillment, and AI. Compared to its previous Amazon-centric operating system, Pattern aims to extend these capabilities to more platforms and markets. However, which of these Amazon-proven capabilities can be replicated, and which need to be rebuilt? When brands enter multiple channels simultaneously, how do they decide where to go next, and what operating and supply chain capabilities do they need to add? As AI begins to impact product discovery and even transactions, how should brands reshape their business models? During Accelerate 2026 China, Ebrun Power engaged in a dialogue with Pattern's Chief Revenue Officer John LeBaron and Chief Operating Officer Rob Hahn. Compared to the conference's discussions on trends and opportunities, this conversation focused more on the specific practices and judgments behind these questions.

01

From Amazon to More Platforms: What is Pattern Replicating?

Ebrun: Pattern developed a comprehensive set of operating capabilities within the Amazon ecosystem and now covers over 70 platforms globally. However, platforms differ significantly in traffic logic, advertising systems, and consumer behavior. When entering a new platform, which capabilities can be replicated from Amazon, and which need to be rebuilt?

John: From a high-level perspective, most of the capabilities we developed on Amazon can be applied to other platforms. The key factor determining the extent of replication is the platform's technical maturity. Internally, we assess platforms based on their technical capabilities. On a scale of 1 to 7, with 7 being the most mature, Amazon currently sits at the most mature end. Specifically, we look at the product information the platform collects, such as weight, dimensions, price, description, and images. We also evaluate what data the platform opens to merchants—whether it requires downloading tables or can be accessed via APIs—and whether the payment, logistics, and returns systems are robust. Another critical factor is what we can input back into the platform. If we can adjust ad bidding, update images and videos, and modify titles and descriptions via APIs, we can replicate more of our established algorithms and automation capabilities on other platforms. If the platform does not support these operations, we need to invest more manual effort, which reduces efficiency. Currently, the gap in technical maturity between platforms remains significant. Walmart has improved many technical capabilities over the years, and TikTok provides interfaces that allow us to access data and execute more operational actions. However, on some platforms, many tasks still require manual work.

Ebrun: If a platform's technical capabilities are insufficient, does Pattern only adapt to it, or does it also participate in building the platform's capabilities?

John: More and more platforms are willing to collaborate with us on improvements. Sometimes, we directly point out that their current technical capabilities limit what we can do. These platforms themselves want to attract more brands, consumers, and products, so they have the incentive to improve. For example, Coupang once sent developers to the U.S. to work directly with Pattern to improve platform capabilities. Walmart's advertising platform was not as mature a few years ago but has made many improvements since. During this visit to China, we just had discussions with JD.com and will continue technical development next, including algorithmic advertising capabilities. TikTok already has API interfaces that connect with Pattern. So, bringing a set of capabilities to more platforms is not just about Pattern adapting unilaterally. The extent to which platforms are willing to open data and interfaces also affects how much we can ultimately achieve.

Ebrun: Beyond technical and operational capabilities, why is Pattern continuing to invest in its own warehousing and fulfillment system? FBA and third-party logistics are already quite mature.

Rob: What we aim to build is a platform-neutral supply chain. For example, if a Chinese brand sells on Amazon, Walmart, and TikTok simultaneously, following each platform's system could require sending goods to different warehouses. We want brands to send goods to our system first, then allocate them based on different market and platform needs. We are not competing with FBA. Instead, we can replenish FBA stock or restock other platforms. We focus on inventory forecasting and allocation across platforms. Our goal is to shorten the time from production to entry into various markets and platforms. As channels grow, this cross-platform supply chain capability will become more important.

Ebrun: As Pattern connects more platforms and gains cross-platform data, operations, and supply chain capabilities, has the company considered building its own consumer-facing e-commerce platform?

John: There is currently no such plan. First, building a platform is extremely difficult and costly. Pattern does gain a lot of data from various markets and platforms, which helps us make better decisions. But at this stage, we prefer to continue as an accelerator rather than creating a new e-commerce platform.

Rob: A key value of Pattern is helping brands connect with consumers, wherever they are. If we build our own platform, it could conflict with brand interests. For example, we might prefer brands to sell more on our platform, but that might not be the best choice for them. Now, we remain channel-neutral, helping brands decide where to sell based on where consumers are. At the same time, we are not in competition with platforms like Amazon, Walmart, or Coupang; we help brands connect with them.

02

Multi-Platform Globalization: What Capabilities Do Brands Need First?

Ebrun: In your speech, you mentioned that brands need to shift from single platforms to multi-channel operations. But adding channels increases costs, so for brands with limited resources, where is the boundary for multi-channel operations? How does Pattern judge whether a brand should enter a new channel or deepen existing ones?

John: We don't make decisions for brands; we help them make their own judgments. Each brand has different capabilities, budgets, and resources, so it's impossible to enter over 70 platforms at once. Multi-channel operations don't mean covering many platforms; the key is deciding where the most valuable next step is. We help brands evaluate from multiple angles, such as the revenue opportunity of a channel, its strategic value, and the opportunity costs of entering or not entering a market. Different products suit different channels. For example, if a product requires more explanation and education and has a moderate average selling price, TikTok might be suitable, as creators can handle some product education and sales. For other brands, entering a market like Canada might be a more appropriate next step than the usual European or U.S. markets. So, we focus on what is best for the brand at its current stage, not just increasing channel numbers.

Ebrun: For a brand that has succeeded on one platform, what problems are most often underestimated when entering new platforms or markets? Why can't original success methods be directly replicated?

Rob: Inventory management is a major issue; brands struggle to determine how much inventory to place on each platform. But beyond that, I think compliance is often underestimated. Different countries have different regulations. For example, selling health products, supplements, or other regulated goods in Canada has different rules than the U.S. or Europe. Brands often underestimate the time, cost, and complexity of compliance when entering a new market. Another key is consumers. Even if a brand sells well on Amazon in the U.S., it doesn't guarantee the same success on Walmart because the consumers on each platform are not identical. Consumer needs and purchasing habits vary across countries and platforms. So, before entering a new market or platform, brands must reassess whether the product fits local consumers, whether the price is right, whether the target consumers are there, and whether the supply chain can support this operating model.

Ebrun: We observe that more Chinese new brands are considering globalization at early stages. What capabilities should these brands prioritize building?

Rob: The most important is the product itself. Don't spread resources too thin at the start. Choose one market, get consumer feedback quickly, and confirm whether the product truly solves a real problem and achieves product-market fit. Brand, supply chain, and advertising strategies can be adjusted, but the underlying product is hard to change through operations. So, I'd advise brands to validate quickly whether they have a good product that fits the target consumer.

John: Global expansion is actually very difficult. Often, the money and time lost in globalization can outweigh the gains. Brands that do globalization well are typically disciplined and clear about what they want. For example, whether to do it in-house or find partners; whether entering a market is for short-term revenue or long-term strategic goals; the local regulatory, tax, and geopolitical environment; the actual market size; and whether distribution is online-only or also offline. Globalization sounds attractive, but brands that do it well often think through these issues before entering.

Ebrun: Compared to Western brands, what advantages have Chinese brands formed that you've seen?

John: One clear advantage is the cost structure. Chinese brands are often closer to manufacturing, giving them some advantage in production and labor costs. Additionally, Chinese brands often have a deep understanding of e-commerce platforms. Sometimes, their knowledge of U.S. e-commerce platforms exceeds that of some American or European brands. There's also social commerce. Chinese brands have accumulated extensive experience in social commerce, while many U.S. brands are still learning how to use TikTok. So, in content, social commerce, and new channel judgment, Chinese brands often have better intuition.

Rob: I'd also add product iteration speed. Chinese brands are closer to the manufacturing system, so when they have new product ideas, they can adjust and iterate faster with production. If used well, this capability could allow Chinese brands to make better products faster.

03

When AI Enters the Transaction Chain, Where Will Pattern Stand?

Ebrun: Traditional e-commerce operations heavily revolve around search keywords, rankings, product detail pages, and ad placements. If more consumers start using AI to express needs and let AI compare and select products, how will brand operations and marketing change?

John: First, it's important to note that LLM-related shopping behavior is growing fast but is still very small in volume. So, in the short term, brands don't need major adjustments. Most e-commerce transactions today, especially in China and other major global markets, still occur through traditional e-commerce platforms, so brands still need to advertise and optimize products around search. But looking ahead, product discovery will indeed change. AI is still a form of search, but it returns fewer results. Traditional search might have many pages of results, but large language models typically give consumers only two or three options. This means brands need to continuously monitor which brands are more frequently recommended by AI in their competitive categories and why.

Ebrun: When AI recommends only a few products each time, how can brands increase their chances of being recommended? How is this different from past search optimization?

John: We now use GEO to help brands assess their visibility and evaluations in large language models, and then analyze why those results occur. But this isn't just about optimizing product titles or descriptions. Large language models can read an increasing number of information sources, including consumer reviews, Reddit forums, TikTok videos, and even company-related information. They judge a product or company from multiple sources. So, to get recommended in the future, brands need not only better product information but also genuinely good products. To some extent, how the company itself is doing will also become increasingly important.

Ebrun: You mentioned that Pattern has no plans to build its own e-commerce platform. But if AI in the future not only recommends products but also participates in ordering, payment, and fulfillment, would Pattern take on a new role?

John: This is a direction worth watching. There is no real agentic marketplace yet. Today's e-commerce platforms are more beneficiaries of AI trends; after consumers discover products via AI, transactions still flow to brand sites or existing traditional e-commerce platforms. But if large language models don't establish a neutral transaction platform themselves, new space could emerge. Pattern currently has both technical infrastructure and logistics infrastructure. Theoretically, there could be a layer of intermediate capability that connects brands and consumers, handling transactions, fulfillment, and even returns without the massive investment in building a full platform like traditional e-commerce. However, this is more of a thought about future white space, not a current business initiative.

In John and Rob's view, Chinese brands do not lack product iteration speed or platform operation experience; the real challenge is taking these advantages to different markets. Pattern is trying to connect channel selection, advertising, data, inventory, and fulfillment capabilities to serve brands moving from single platforms to more markets. For Chinese brands accelerating globalization, these cross-platform, cross-market capabilities are increasingly important.

Ebrun will continue tracking this intelligence. For more related information, please scan the QR code to follow the author's WeChat.

This article was first published on Ebrun Power's official website.