What this covers

Drew and Matt trace Coupang from its origin as a Groupon-style group-buying site. Bom Kim pulled a planned IPO a few weeks before it happened, having decided the deal model was unsustainable and that the company needed to copy Amazon's full logistics stack. Unlike Amazon, which started on FedEx, Coupang built its warehouses and Rocket Men couriers in-house, because Korea's small-parcel market was underdeveloped. Local conditions help the story: a Fair Trade Act bars disclosing MSRPs on electronics, which kept store prices high, and Korea's density lets one courier serve a single high-rise for a day, so a 24-hour network costs far less than it would across the U.S.

The comparisons with rivals are where it gets sharpest. NAVER's alliance with CJ Logistics is set against Alibaba's Cainiao, which had to take equity stakes in its shipping partners. SSG.com's failure at fast delivery in Seoul, Etsy sellers' petitions that moved no sales, and Amazon's white web page losing to Taobao's busy layout each illustrate a different way that copying a tactic falls short. The Merchants as Bees, Customers as Flowers framing explains who holds leverage on a marketplace.

The last stretch turns to economics. Drew argues for measuring Coupang by GMV because 1P books full item prices as revenue while 3P books only commissions. He also says a high take rate says little about profit. Advertising on a high-intent platform adds margin, and grocery through Rocket Fresh builds weekly habit. With about 14 million Rocket WOW members among 22 million households, growth has to come from wallet share. Drew closes by valuing the company with a Reverse DCF.

Sharpest takeaway

Drew argues Coupang overtook incumbents despite entering late because it identified consumer preferences for reliability, speed, and trust that marketplaces left unmet, then rebuilt its whole business around first-party logistics to serve them.

  • Marketplace models sold impulse discounts and could not guarantee reliable delivery, so trust and consistency were unmet.
  • Owning warehouses, couriers, and returns created a self-reinforcing flywheel of volume, data, supplier terms, and habit.
  • Japan's quick-commerce failure showed that copying a tactic without the full integrated system does not work, while category extension in mature markets is a lower-risk growth path.

The argument · threads15 threads · 82 claims
0.85

Owned integrated logistics delivers reliability and speed that rivals and marketplaces must take years to match.

11 pointscentrality 5/5
  • Shoppers value reliability, consistency, and trust above price and selection, and marketplaces fail on these because they often lack inventory before selling, so buyers keep backup options in mind; Coupang's ownership of logistics and fast returns removes that hesitation and wins share.

    “whenever you're buying on one of these platforms, you're always wondering in the back of your head if something is going to go wrong and that creates hesitation”

  • NAVER's shipping alliance with CJ Logistics is hard to coordinate because merchants and logistics partners have different incentives, as Alibaba's Cainiao experience showed when it had to take equity stakes in its six shipping partners, while Coupang can optimize a single integrated system.

    “everyone has different prerogatives. And if you look at Alibaba, who tried to do something similar, their shipping service, they put together what's called Cainiao”

  • Coupang removed about 85 percent of the cardboard it used to use by switching to plastic containers and reworking upstream sorting, which saves space so each truck carries more items and one courier delivers more, making each truck more profitable.

    “they got rid of 85% of all the cardboard they used to use in their operation”

  • Coupang's flywheel starts with a large warehouse footprint and inventory, which lets it buy better from suppliers and attract customers, and volume produces data that allows inventory placement and logistics optimization, after which third-party merchants can be added because consumers are already onboard.

    “I mean look at how their model, how their flywheel is set up, it all starts with the warehouse.”

  • Shinsegae's SSG.com failed to deliver very quickly in Seoul, even though Seoul is the easiest place to attempt it, because it was trying to fulfill from a store footprint rather than a dedicated fulfillment network, so partial rearchitecture cannot deliver the experience.

    “You can't do it if you're holding all your inventory in a department store”

  • Market Kurly, a 2015 competitor, offered overnight grocery delivery for orders placed by 11 p.m. and delivered by 7 a.m., which set the bar Coupang later pushed past with its own delivery promise.

    “if you place an order by 11:00 p.m. the night before, we will get you that delivery by 7:00 a.m. the next day”

  • Coupang moved from a roughly 8-hour overnight grocery delivery window to a 7-hour window for millions of items under Rocket WOW, then added Rocket Fresh grocery in 2019, and by 2018 had 99 percent of deliveries done in less than a day.

    “By 2018, 99% of all deliveries are done in less than a day”

  • Coupang's couriers are known for conscientious service like handwritten notes and knowing residents, which emerges because dense Korean neighborhoods let a single courier serve the same people repeatedly.

    “you get to know the area as like your local deliverer”

  • Unlike Amazon, which initially relied on FedEx before building its own logistics, Coupang built its warehouses and Rocket Men courier network entirely in-house from the start, because Korea's small-parcel logistics market was underdeveloped.

    “It's not like Amazon piggybacking off of FedEx initially before they built their own logistics footprint.”

  • Drew says a new winner in e-commerce must build a large warehouse footprint and first-party inventory, win customers, and then add third-party sellers, because each step depends on the previous one and incumbents with stores or marketplace-only models cannot shortcut it.

    “If you want to go and compete against them, you've got to get this huge warehouse footprint”

  • As of 2022 Coupang had over 47 million square feet of logistics space and more than 100 fulfillment centers, and it is not expected to keep building fulfillment centers indefinitely.

    “They have 47 million square feet plus of logistics infrastructure space, and that number itself is from 2022. They have over 100 fulfillment centers.”

0.60

NAVER is the main competitive threat, though cultural fit shapes success and Korea can support more than one winner.

7 pointscentrality 4/5
  • Amazon entered China with a clean white web page similar to the U.S. site, which consumers found boring compared with Taobao's dense, busy layout, showing that foreign platforms fail when they miss subtle cultural preferences.

    “they came in with this very clean web page, which was white and basically very similar to the website you would see in the U.S.”

  • South Korea never developed specialized chain retail the way the U.S. did, so there is a large selection gap, which is why e-commerce penetration is high at about 35 percent of retail and e-commerce demand is structurally large.

    “South Korea never had the same development in specialized retail that we have in the U.S.”

  • Coupang is priced in line with or much cheaper than physical retailers but loses the most price-sensitive shoppers to Temu, AliExpress, and NAVER third-party merchants, while winning on cheap everyday items and expensive items where return assurance matters.

    “on very low items, the cheap items are always going to Coupang. They don't care about a small price difference.”

  • NAVER is the most significant latent risk because it is where many Koreans begin product search, offers price comparison, NAVER Pay checkout, and Smart Stores catalogs, and about 20 percent of surveyed consumers use it more than anything else for e-commerce.

    “of the consumers we spoke to, about 20% said they use NAVER more than anything for their e-commerce purchases.”

  • A Korean Fair Trade Act prohibits disclosing MSRPs on electronics, which keeps in-store electronics prices high and gave e-commerce a price advantage it could use to undercut physical retail.

    “There's something called the Fair Trade Act, which actually makes it so you're not allowed to disclose MSRPs on electronics”

  • Drew sees NAVER more like Shopify in the U.S., a platform that enables an alternative channel for merchants that still needs to be controlled, and he says there could be more than one winner in Korea.

    “There definitely could be more than one winner. I'd see NAVER as being a little bit more like Shopify is in the U.S.”

  • NAVER, Korea's Google equivalent, is the main long-term threat to Coupang, because about 20% of surveyed consumers use it most for e-commerce, while about 60% use Coupang as their number one platform.

    “about 20% said they use NAVER more than anything for their e-commerce purchases”

0.57

Coupang is now profitable, with rising margins from operating leverage, so runway concerns are overstated.

11 pointscentrality 3/5
  • Coupang raised a couple hundred million in initial VC rounds, then a $1 billion SoftBank investment, after which it was more or less self-funded; it ran negative cash flow for early years, hit profitability around 2023, and still had about negative $1.5 billion in 2021.

    “They did a $1 billion investment. And then it was more or less self-funded from there on out.”

  • Using the mature margins of Walmart, Costco, and Target at roughly 3 to 6 percent as a framework, Speedwell expects Coupang's mature margin to land near 5 percent, with advertising adding a couple of points on top, and EBIT mature margin of 5 to 8 percent.

    “everyone focuses around on the same mature margin framework, which will put you somewhere around 5%.”

  • Coupang moved from about negative $1.5 billion EBIT in 2021 to roughly positive $500 million today, showing operating leverage because the infrastructure spend was front-loaded and the margins should keep inflecting upward as build-out slows.

    “they were losing $1.5 billion in EBIT. And we fast forward today, and they're doing about plus $500 million.”

  • Analysts should measure Coupang by GMV rather than revenue because in 1P sales the full item price is booked as revenue while in 3P only commissions are recorded, so a rising 3P mix distorts reported revenue economics.

    “if you're selling something on 1P, 100% of the item is recorded as revenue, if it's 3P, it's just the commissions.”

  • Speedwell estimates Coupang's GMV at around $38 billion, with a range around that figure.

    “We estimate GMV is somewhere around $38 billion altogether”

  • Coupang earns roughly 25 percent gross margins, and Speedwell estimates first-party sales at about 55 percent of the business and third-party at about 45 percent, triangulated from segment reporting and an accounting change; gross margins should rise as the mix shifts toward 3P.

    “they're doing about 25% gross margins, and you would expect those to go up more”

  • Coupang grew revenue 18 percent in 2023, accelerating somewhat with FX adjustment, and generated about $1.5 billion in free cash flow, roughly ten years after committing to a fully integrated logistics operation, which is much faster to profitability than Amazon.

    “they did 18% revenue growth in 2023. They accelerated revenue growth a bit.”

  • A high take rate is not necessarily informative about profitability, because logistics fees contribute less to EBIT than advertising does, and as platforms take on more of the transaction value (storage, fulfillment, customer finding) they capture a larger share of it.

    “it's not necessarily the most informative because you could have a very high take rate and a large portion of that could be not very contributive to EBIT”

  • Drew says he cannot know what the market is pricing, but revenue growth was 18% in 2023 and accelerated, and foreign exchange effects make the reported figures understate growth, so the question mark over runway may be overstated.

    “they did 18% revenue growth in 2023. They accelerated revenue growth a bit.”

  • Coupang is profitable on an EBIT basis and is free cash flow positive, after reaching profitability around 2023 following years of negative cash flow.

    “they are profitable on both an EBIT basis as well as free cash flow positive”

  • Management has said its long-term adjusted EBITDA margin should be around 8% to 10%, which Drew translates into an estimated 5-8% mature EBIT margin.

    “I believe they said long-term adjusted EBITDA margins are somewhere in the vicinity of 8% to 10%”

0.52

Deal-based group buying is unsustainable, so Coupang had to build its own logistics.

4 pointscentrality 4/5
  • Bom Kim, after launching Coupang as a Groupon-style group-buying business in South Korea, pulled a planned IPO a few weeks before it, because he saw the deal-based model as unsustainable and decided the company had to copy Amazon's full logistics stack instead.

    “he decides to pull the IPO. And he says at this time that they are not providing the 100x experience that they want to for their consumers.”

  • A third-party marketplace sells overstocked, spontaneous-purchase items, but high-frequency essential goods require first-party inventory and warehousing, which is why Coupang moved from 3P toward 1P.

    “they decide they need to do 1P if they really want to get some of these more essential high frequency purchased goods.”

  • Coupang is basically the Amazon of South Korea with a similar first-party logistics operation, has signed up about half of Korea, is the #1 e-commerce player with about $25 billion in revenue, has grown at a 45% CAGR since 2018, and is both EBIT profitable and free cash flow positive.

    “Coupang is basically the Amazon of South Korea, very similar first-party logistics operation.”

  • Groupon-style group buying drives spontaneous, discount-driven purchases rather than essential purchases, so despite strong early revenue it is not a sustainable long-term business model.

    “It's very impulsive, it's spontaneous, and it's not the same very quality essential sales that a lot more commerce usually focuses on.”

0.44

Owning urgent, high-intent purchases makes Coupang the default, habit-driven destination for shoppers.

6 pointscentrality 3/5
  • Coupang already has about 14 million Rocket WOW members out of 22 million Korean households, with 21 million active buyers, so growth must come from deeper wallet share and category expansion rather than new customer acquisition.

    “14 million members are these Rocket WOW members.”

  • Advertising is especially valuable on a high-intent platform like Coupang because shoppers arrive with specific purchase intent and spend less time browsing, so top search placement is monetizable, and this is why advertising can add margin beyond the retail baseline.

    “People come there with very high intentionality of what they want to purchase, and they tend to not want to be there that long.”

  • Coupang holds about 25% of Korean e-commerce but only a high-single-digit share of total retail, so most of the opportunity is in shifting retail spending online.

    “right now, they have about 25% penetration of just the e-commerce market, but it's a high single-digit penetration of the overall market.”

  • The most valuable e-commerce transaction is one where a shopper needs an essential item now and goes directly to a platform they trust, and Coupang aims to own that moment of intent.

    “That's going to be Coupang, and that's what they're trying to win.”

  • Coupang's purchase flow lets a shopper open the app, search, and buy in under 30 seconds, which Drew argues is the kind of frictionless habit that makes the platform the default.

    “You could buy something on their site in under 30 seconds.”

  • Achieving top-of-mind positioning, where consumers stop hesitating and go straight to the platform, requires heavy capital expenditure that rivals using a marketplace model do not make.

    “You have to spend all of this CapEx just to get to the top-of-mind positioning in the consumer's head”

0.38

A low take rate and nascent advertising leave substantial monetization headroom.

4 pointscentrality 3/5
  • Coupang's merchants did not object to advertising until a couple of years ago, when sponsored ads began appearing above organic search results, and the advertising business is still relatively nascent.

    “the merchants we spoke to didn't get upset about the advertising until a couple of years ago”

  • A fully loaded take rate for an Amazon merchant using all services, such as advertising and fulfillment, may be in the vicinity of 35 to 45 percent, which is far above Coupang's roughly 11 percent consolidated take rate.

    “their fully loaded take rates on someone who's using all these services, may be call it anywhere in the vicinity of 35% to 45%.”

  • Coupang's consolidated take rate is roughly 11 percent of GMV, built from selling commissions of 5 to 11 percent, advertising of roughly 5 to 15 percent for merchants, and logistics fees around 15 percent, and this low level suggests substantial room to grow advertising and fulfillment monetization.

    “their fully loaded take rates on someone who's using all these services, may be call it anywhere in the vicinity of 35% to 45%.”

  • Merchants are like bees and customers are like flowers, so whoever controls customer traffic holds the leverage; marketplace merchants cannot protest effectively because the platform controls search position, while strong brands that customers search by name retain more bargaining power.

    “merchants are like bees and customers are the flowers. And so the bees always go to wherever the flowers are.”

0.36

Korea's density and low theft let Coupang build fast delivery far more cheaply.

3 pointscentrality 3/5
  • South Korea's population density lets a single courier serve sometimes a single high-rise building's deliveries for the day, so a fast 24-hour delivery network can be built with far less capital than the same network would require across the much larger U.S.

    “a single courier is able to do sometimes at extremes and today, just a single high-rise building, maybe all of their deliveries for the day.”

  • Very low crime and theft in South Korea let Coupang pick up returns left outside doors without packaging and leave grocery cooler bags outside during the day, which enables service innovations that would be impractical in higher-theft markets.

    “there's very little crime or theft in South Korea, and so you are able to just leave an item outside your door when you return it.”

  • Coupang trucks are designed for small parcels with a side door, so they save seconds or minutes per delivery, which compounds into efficiency gains that competitors using rear-door trucks do not capture.

    “if you look at a Coupang truck, it is literally designed for these small parcel packages and you open up the door on the side”

0.30

Near-term growth should come from deepening existing platforms rather than launching new geographies.

6 pointscentrality 2/5
  • Coupang tried quick commerce in Japan with small fulfillment centers, motorcycle couriers, and about 300 items such as bento lunches to meet 30-minute demand, but consumers did not care, so the company abandoned Japan as a failed experiment.

    “that turned out to not be so successful.”

  • Southeast Asia is a potential target, but its hundreds of islands, multiple countries and languages, and lower household incomes make a full integrated logistics build-out very hard and expensive, so it is unlikely for some time, while Singapore is too small to move the needle.

    “You're talking about hundreds of different islands. You're talking about multiple different countries, languages.”

  • Growing by extending categories and converting good customers into great ones is easier and lower risk than launching a new geography and hoping it sticks, so near-term growth comes mainly from deepening the Korean and Taiwanese platforms.

    “getting a good customer to become a great customer, in some ways, it's been easier avenue of growth than launching a whole new market”

  • Grocery is the most valuable e-commerce category because it is bought weekly or biweekly, is essential with no real substitutes, and carries large order values, so winning it through Rocket Fresh habituates customers who then buy more across categories.

    “grocery is such an important category for an e-commerce player, and it's the reason why Amazon and Walmart are tripping over themselves”

  • South Korea's population is flat to slightly shrinking, so growth must come from penetration, category depth, and rising incomes rather than from more people.

    “South Korea, though, the population is not growing. It's flat to slightly shrinking over time.”

  • A company that dominates its market and grows by adding categories represents lower-risk growth than one opening a new market that depends on success there, so market saturation may cap upside while still protecting downside.

    “you may also feel that a company that has such dominance in a market and is able to grow by just adding more categories is a lower risk growth”

0.30

Valuation depends on judging risk against price, with financial disclosure gaps still unresolved.

6 pointscentrality 2/5
  • Capex to D&A is over 2x, but without separating maintenance from growth capex it is unclear how much spend is truly one-time build-out, since trucks have finite life while warehouses last longer and do not require new land.

    “the issue there is we don't know what growth versus maintenance CapEx is.”

  • Coupang's accounts payable are about 300 percent of inventory, meaning roughly $3 of payables per $1 of inventory, and it turns inventory almost 10 times, which together reflect how fast it moves product and how much it is financed by suppliers.

    “For every dollar of inventory they hold, they have $3 in accounts payable.”

  • Drew values Coupang using a reverse DCF, choosing assumptions he is comfortable with on growth, market share, and EBIT as a percent of GMV, then computing the implied return to judge whether the risk is worth it relative to the price paid.

    “So I approach it the way I always do, which is the reverse DCF. And so you see what assumptions that you're comfortable with and then you see what the implied return of those assumptions would give you.”

  • Coupang's IPO happened at a much higher valuation, around $60 billion by Drew's recollection, when the company was losing money, lacked the same customer grip, and had fewer Rocket WOW members, so today's profitability reflects much more execution.

    “I believe it was around $60 billion, they were losing money.”

  • Coupang claims a retail TAM of about $480 billion, but Speedwell estimates $400 to 420 billion after excluding travel and restaurant spend; e-commerce penetration is about 35 percent, implying an e-commerce TAM of about $145 billion.

    “Coupang will quote a number that it's $480 billion. Our estimates are a little lower.”

  • Coupang reports only a single operating expense line item that bundles all costs together, which limits cost transparency and is a legitimate criticism of the company.

    “they only report one operating expense line item and they bucket absolutely everything together in there.”

0.30

Third-party sellers fill selection gaps while brands and sellers hold limited leverage over the platform.

6 pointscentrality 2/5
  • Branded merchants like Nike or major cosmetics labels have more bargaining power than generic sellers because shoppers search for the brand specifically, though even brands risk losing sales if they are absent from a platform where customers expect next-morning delivery.

    “If you are looking at a brand like Nike or some well-known cosmetic brand, where people are searching specifically that brand, they do have more ability to command what they want to command.”

  • Etsy merchant petitions against policy changes made noise but had no measurable impact on sales, because most shoppers never go beyond the first page of search results, so sellers lacked leverage.

    “it never impacted sales whatsoever because at the end of the day, you're never usually even going into the second page of the search engine results.”

  • Coupang's apparel and cosmetics/beauty categories are weaker, which it is addressing through increased third-party selection and the roughly $0.5 billion acquisition of Farfetch with about $4 billion in GMV, though Drew says the purpose of the deal is somewhat unclear.

    “They did recently do an acquisition of Farfetch”

  • Drew argues that a founder-owned, operator-led company is distinctive because Bom Kim cuts failed experiments openly, as with Japan, rather than making excuses, which is what an owner-operator should do.

    “They just cut the spend there. They said it didn't work and they moved on, which is exactly what you would want to see owner-operator do.”

  • Amazon acquired Diapers.com early on to fill out its baby categories, whereas Coupang built its category depth organically.

    “I know Amazon in the early days acquired Diapers.com to fill out some of the baby categories”

  • Coupang has emphasized third-party sellers over the past couple of years to fill selection gaps, with an estimated split of about 55% first-party and 45% third-party in its business.

    “we estimate that 1P is about 55% of their business and 3P is 45%”

0.27

Membership and deepening cohort spend drive growth beyond new customer acquisition.

5 pointscentrality 2/5
  • About 60 percent of consumers use Coupang as their number one e-commerce platform, giving it a clear lead over NAVER on delivery speed, consistency, trust, order ease, selection, and to a lesser extent price.

    “Coupang was used by about 60% of consumers as their #1 e-commerce platform.”

  • After switching from quick commerce, Coupang in Taiwan replicated the full Rocket WOW model, opened fulfillment centers, and reached the top shopping app, filling a niche because Taiwan lacks an Alibaba, JD.com, or similar first-party logistics player.

    “instead of doing this quick commerce, they decided to do the full Rocket WOW experience.”

  • Coupang's Rocket WOW membership has about 14 million of 22 million Korean households paying roughly $5 to $7.50 per month for free delivery, low grocery minimums, Coupang Play streaming, and 5 to 10 percent discounts on Coupang Eats, and members who use more services churn less and spend more.

    “14 million members are these Rocket WOW members. They're paying about, it used to be about $5 a month, and then it was just increased to about $7, $7.50 a month.”

  • Coupang's easy returns, handled directly with the company rather than through individual merchants, remove pre-purchase research and hesitation, which raises conversion, and many customers simply keep items because returns are cheap to avoid.

    “they are able to really remove a lot of the frictions from returns, which then when you go to buy something, it means that you're not going to be hesitating to buy it.”

  • Every single cohort of buyers that joined Coupang has increased its spend over time, with buyers now averaging a little over $1,100 in revenue a year, about 3x since 2018, meaning growth comes from deepening existing customers as much as adding new ones.

    “every single cohort of buyer that joined Coupang, you see that they increased their spend over time.”

0.25

Coupang's Korean rise came through aggressive competition and heavy investor backing.

4 pointscentrality 2/5
  • Coupang's competition falls into three buckets: department stores and big-box retailers such as Lotte, Shinsegae, Hyundai, and Costco; pure-play marketplaces such as Lotte On, eBay Korea, SSG.com, Market Kurly, 11Street, Timon, and Wemakeprice; and internet platforms Kakao and NAVER.

    “We have the department stores and big-box retailers, which include LOTTE, Shinsegae, Hyundai and Costco.”

  • When Coupang launched as a Groupon clone in South Korea, it faced about 30 competitors, and Kim responded by blanketing the market with Facebook ads, reportedly showing 72 ads to every person in Korea.

    “they're showing 72 Facebook ads to every single person in Korea.”

  • Coupang's early success in outcompeting Ticket Monster, Wemakeprice, and others helped it attract Sequoia after the pivot and SoftBank a couple years later, with SoftBank making a $1 billion investment.

    “they were able to get Sequoia involved and a couple of years later, SoftBank as well.”

  • Bom Kim, born in South Korea and raised in the U.S., sold a magazine called Current as an undergraduate at Harvard, sold a second magazine called 02138 near the financial crisis, then dropped out of his Harvard MBA after a year to pursue Groupon-style group buying in South Korea.

    “At the end of the first year of his MBA, he drops out, and he drops out because he wants to pursue this new and novel business model”

0.22

Late entrants win by serving unmet consumer preferences rather than by being first.

3 pointscentrality 2/5
  • A late entrant can still win leadership as long as it fulfills consumer preferences that remain unmet, because the first-mover advantage is about preference fit rather than being early to the category; Coupang entered in 2014, almost two decades after Amazon.

    “you could be a pretty late entrant, and as long as you are fulfilling consumer preferences that are still unmet, you still have an opportunity for this top leadership position.”

  • Coupang entered in 2014, almost two decades after Amazon, which gave local competitors time to build a lasting presence in South Korea and leaves Coupang with lower market share than Amazon had at comparable penetration.

    “Coupang really got into this business in 2014. That's almost two decades after Amazon.”

  • A company should explicitly choose which consumer preferences to optimize for, because winning comes from identifying unmet preferences and organizing the whole business around serving them better than anyone else, not from a generic first-mover advantage in e-commerce.

    “you could be a pretty late entrant, and as long as you are fulfilling consumer preferences that are still unmet, you still have an opportunity for this top leadership position.”

0.08

Korea's conglomerate history and pro-worker protections shape Coupang's operating environment.

3 pointscentrality 1/5
  • Korean commerce was historically directed by large government-friendly conglomerates and department stores, and Coupang and other e-commerce players are an anomaly in breaking through that structure to become major parts of the economy.

    “most of it has been directed by these larger, more government-friendly corporations.”

  • Coupang's Rocket Men couriers are employed directly rather than as contractors like DoorDash drivers, but after a warehouse fire brought scrutiny over whether couriers were overworked, the company pushed back by citing the jobs, pay, and benefits it provides.

    “It's not like the DoorDash relationship or some of these other food delivery platforms.”

  • Korea is a fairly capitalistic economy but has historically been less so and has introduced pro-worker protections that can shape outcomes for platforms like Coupang.

    “It's a pretty capitalistic economy at least as of late, but that definitely always hasn't been the case.”

0.05

Handling returns directly removes friction, though Coupang's return rates remain undisclosed.

2 pointscentrality 1/5
  • Because Coupang handles returns directly rather than through a variety of merchants with different policies and post-office packaging, it has removed more return friction than NAVER, and it goes further than Amazon on this dimension.

    “one of the big advantages Coupang has versus them is that it is much easier to return items because you're dealing directly with Coupang.”

  • Coupang does not disclose its return rates, so any comparison to Amazon's return experience is inferred rather than measured.

    “they've never disclosed return rates”