Latest / Business Strategy Talks with Fexingo: Competitive Advantage, Market Positioning, and Growth Planning / How Coupang Built a Moat on Logistics Speed
Transcript
- Lucas: So Coupang. If you live outside South Korea, you probably know it as the Amazon of South Korea, which honestly undersells it. In 2025, the company delivered roughly three out of every ten online orders in the country, and it did it with a network of over a hundred fulfillment centers that put it within a few hours of nearly the entire population. Luna: And that speed is the moat we want to talk about today. Because it's not just about being fast — it's about how that speed creates a self-reinforcing loop that rivals can't easily match. Lucas: Exactly. The starting point is what Coupang calls Rocket Delivery — next-day delivery as the default, same-day for orders placed before midnight in major cities. That might sound like table stakes for Amazon Prime, but in South Korea, it's a different beast. The country is dense, with over fifty million people packed into a landmass about the size of Indiana. That density makes speed logistically feasible, but it also means competitors like Naver and eBay Korea are fighting for the same customers. Luna: So how did Coupang actually build the network to pull that off? I mean, they didn't just wake up one day with a hundred fulfillment centers. Lucas: Right. The key decision was going all-in on their own fulfillment network early on, even when it was a huge capital drain. In the mid-2010s, while competitors were leaning on third-party logistics, Coupang invested billions in building its own warehouses, its own delivery fleet, and its own routing software. The bet was that controlling the entire chain would let them hit delivery windows that third-party partners couldn't guarantee. Luna: And that's a classic 'high fixed cost, high barrier to entry' play. But what made it work in the long run? Was it just the capital, or was there something smarter in how they deployed it? Lucas: Good question. The smarter piece was starting with a limited geography. Instead of trying to cover all of South Korea from day one, Coupang focused on the Seoul metropolitan area, which is home to almost half the country's population. They built their first fulfillment centers there, nailed the delivery experience, and then expanded outward. That let them prove the unit economics before committing to a nationwide build-out. Luna: So it was almost a franchise model — prove it in one market, then replicate. And the density of Seoul meant that early on, each delivery route was super efficient. Lucas: Exactly. And that efficiency is what makes Rocket Delivery profitable, or at least not a pure money-loser. If you look at Coupang's gross margin per order, it improved steadily as volume grew, because the fixed costs of warehousing and software got spread across more orders. By the mid-2020s, they were consistently generating positive operating cash flow, which a lot of skeptics didn't expect. Luna: But here's the thing: other companies could copy the logistics model. What stops a big player like Amazon itself from coming in and doing the same thing? Or a local player with deep pockets? Lucas: That's the core of the moat, and it's not just the physical network. It's the data and the network effects that come with it. Coupang's delivery data tells them exactly what customers in each neighborhood order, when they order, and how often. That allows them to forecast demand so precisely that they can position inventory in the right fulfillment center before you even click 'buy.' That shortens the last mile and shaves hours off delivery. Luna: So the more orders they get, the better their forecasting, the faster the delivery, which attracts more orders. That's a textbook data network effect. Lucas: Right. And it's reinforced by the fact that once you experience Rocket Delivery, alternatives feel painfully slow. There's a reason Coupang's customer retention is so high — people don't go back to waiting three days for a package when they're used to next-day, often before 7 a.m. Luna: Wait, before 7 a.m.? Is that a real thing? Lucas: Yeah, Coupang offers 'dawn delivery' — orders placed by midnight can arrive by seven the next morning. It started as a test in a few Seoul districts and expanded because the demand was insane. You order a pack of diapers at 11 p.m., and it's on your doorstep before you wake up. That's the kind of experience that becomes a habit. Luna: That's a killer hook. And it's not just about convenience — it saves customers a trip to the store, which for busy parents is a real value. Lucas: Exactly. And that value lets Coupang charge a premium or earn loyalty in other ways. They have a membership program called Rocket Wow, which is similar to Amazon Prime — free shipping, streaming, and other perks for a monthly fee. In 2025, they had over fourteen million members in South Korea, which is roughly a quarter of the adult population. That's a sticky base that drives repeat purchases. Luna: And those members order more often, which feeds the data loop again. It's a flywheel. But I want to push on something: what about the costs? Building a fulfillment network is capital-intensive, and labor costs in South Korea aren't cheap. How does that not eat the whole margin? Lucas: It does, if you're not disciplined. But Coupang's approach is to treat warehousing and delivery as a technology problem, not just a labor problem. They use automation heavily — robotic sorters, ai driven slotting, and route optimization that cuts unnecessary miles. Their fulfillment center in Daegu, one of their largest, processes tens of millions of packages a month with a fraction of the manual handling you'd see in older operations. That keeps labor costs per package down even as wages rise. Luna: So it's a mix of capital and software. But competitors like Naver, which has a massive logistics partnership with a company called CJ Logistics, could theoretically invest in the same automation, right? What's the actual barrier? Lucas: The barrier is time and ecosystem. Even if Naver threw ten billion dollars at logistics tomorrow, they'd still need years to build the same footprint, tune the algorithms, and — crucially — accumulate the order data to make the network efficient. Coupang has been iterating on this for over a decade. That's a decade of learning how to predict demand in a market with very specific cultural quirks — like the fact that Koreans order a lot of fresh food online, which requires cold-chain infrastructure. Luna: Cold chain is a whole other beast. And it's a great example of how local knowledge becomes a moat. You can't just copy a model from the US or China without understanding the specifics. Lucas: Right. And Coupang is doubling down on that local advantage with services that bundle logistics with entertainment and payments. Coupang Play, their streaming service, is included in the membership, and they've been investing in original content. That's a classic 'ecosystem lock-in' move. If you're watching their shows and getting your packages delivered by their trucks, you're less likely to switch to a competitor for either. Luna: So it's not just a logistics moat — it's a customer relationship moat. And that's harder to replicate than just building warehouses. Lucas: Exactly. So when we talk about Coupang's moat, we're really talking about a compound moat: logistics infrastructure, data, network effects, and ecosystem. Each piece reinforces the others, and together they create a barrier that's not about any single factor but about the integration of all of them. Luna: I think that's a really useful way to think about moats in general, not just for e-commerce. It's not one thing; it's the combination that matters. Lucas: Yeah, and that's why I wanted to talk about Coupang specifically. It's a case study in how a company in a relatively small market can build a defensible position by focusing on operational excellence and customer obsession, rather than chasing global scale. Luna: And if you're listening to this and thinking about your own business, the lesson is to find that one thing you can do better than anyone else and build everything around it. For Coupang, it's speed. Lucas: Before we wrap up, let's talk about the risks. Because no moat is permanent. What could erode Coupang's advantage? Luna: I was just thinking about that. Regulatory pressure on labor practices, for one. There have been reports about overworked delivery drivers and warehouse workers, and that could lead to stricter rules that raise costs. Lucas: That's a real one. South Korea has strong labor protections, and any misstep could become a major reputational and financial hit. Also, there's the question of whether the Korean e-commerce market is reaching saturation. With a population of fifty-one million, Coupang can't grow forever on organic customer acquisition alone. That's why they're expanding into adjacent services like food delivery and even overseas markets like Taiwan. Luna: Taiwan is interesting because it's a similar market in terms of density and tech adoption, but it's also a different regulatory environment and a new competitive landscape. It's not a slam dunk. Lucas: Right. But if they can replicate even a fraction of the Korean playbook, it could be a meaningful growth engine. Still, the core moat is built on their home turf, and that's where they have the deepest data and the strongest brand. Luna: I think that's a good place to leave it. We've talked about how Coupang built its moat, what makes it hard to copy, and what could threaten it. Thanks for listening. Lucas: Yeah, and if today's episode gave you a useful framework for thinking about competitive advantage, we'd love to keep bringing you these conversations. One way to support that is by buying us a coffee at buy me a coffee dot com slash fexingo. It helps keep the show ad-free, and we genuinely appreciate it. But no pressure. Let's get back to the topic — actually, we're already done, so catch you next time.