Hello, everyone. My name is Ronnie, and I will be your conference operator today. At this time, I would like to welcome everyone to the Coupang 2026 third quarter earnings conference call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question-and-answer session. If you would like to ask a question during this time, simply press star followed by number 5 on your telephone keypad. If you would like to withdraw your question, press star and number 5 once again. Now I'd like to turn the call over to Mike Parker, Vice President of Investor Relations. You may begin your conference.
Thanks, operator, and welcome everyone to Coupang's third quarter 2026 earnings conference call. I'm pleased to be joined on the call today by our Founder and CEO, Bom Kim, and our CFO, Gaurav Anand. The following discussion, including responses to your questions, reflects management's views as of today's date only. We do not undertake any obligation to update or revise this information, except as required by law. Certain statements made on today's call may include forward-looking statements, including statements regarding future financial and operational results. Actual results may differ materially. Additional information about factors that could potentially impact our financial results is included in today's press release and in our filings with the SEC, including our most recent annual report on Form 10-K and subsequent filings.
As we share our third quarter 2026 results on today's call, the comparisons we make to prior periods will be on a year-over-year basis unless otherwise noted. We may also present both GAAP and non-GAAP financial measures. Additional disclosures regarding these non-GAAP measures, including reconciliations of these measures to the most comparable GAAP measures, are included in our earnings release, our slides accompanying this webcast, and our SEC filings, which are posted on the company's investor relations website. Now I'll turn the call over to Bom.
Thanks, everyone, for joining us today. I'll start by highlighting a few things upfront and then go deeper into the quarter. First, momentum continued to build across the business in Q3. Consolidated revenue grew 11% year-over-year in constant currency, exceeding the guidance range we provided, and a step up from Q2, even with the Chuseok holiday timing working against us this year. Product Commerce grew 8% in constant currency, or about 9% adjusting for the holiday shift, an acceleration from Q2. Second, our WOW membership in Korea is at the highest level in our history, with strong growth both year-over-year and quarter-over-quarter. Net additions are running above historical rates, driven by both returning members and strong new signups. We keep investing in expanding the value proposition for WOW members, which in turn drives higher retention and growth in membership.
This quarter, we began rolling out Try Now, Pay Later, a new WOW benefit in fashion that lets members try items on for seven days before paying, one small example of how we keep expanding the value we provide. Our goal is to make WOW the best value membership program anywhere in retail. Third, Product Commerce is tracking towards the mid-next-year recovery we laid out last quarter. The underlying customer behavior remains just as strong as what we spoke to last quarter, and we expect the reported growth and margins to approximate historical levels as we lap the affected periods next year. Fourth, in Developing Offerings, growth accelerated to 27% in constant currency, and the customer signals we track most closely strengthened again. Taiwan and Eats continue to lead the way. We are also excited about extending Eats into non-food on-demand delivery in Korea. More on that in a moment.
Let me take each in turn. On Product Commerce, the drivers of our momentum are the same ones that have powered our growth for more than a decade: our unwavering focus on wowing customers with the lowest prices, broadest selection, and unbeatable service. We are working to widen that experience gap quarter after quarter, which is the compounding engine behind this business. The soft spot in reported growth this quarter remains a small cohort of members who left after the data incident and have not yet returned. The members who did return, despite being away and allocating their retail wallets elsewhere for a period, have not only returned to their prior levels of spend, but they have continued to compound this quarter at normalized rates consistent with what we saw last quarter.
The members who never left are also spending at record levels and growing as fast as they were before the incident. These are among the clearest signs yet of where we stand competitively and the strength of our value proposition across service, price, and selection. New members have been more than filling in the gap in member count, though they naturally begin at the early part of the spend curve. It takes time for them to climb it, as each cohort before them has. As we lap the affected periods next year, the non-returning cohort drops out of the year-over-year comp, and our reported growth converges with the growth of the underlying customer base. That base continued to grow spend at normalized rates this quarter, consistent with the spend growth we spoke to last quarter.
Buyable selection is the lever our teams are focused on because it is the input that we see customers responding to most directly. Our assortment on Rocket grew at strong double-digit rates this quarter. Giving customers more of what they need at low prices and with the best delivery experience is one of the clearest ways we compound growth over the long term. Fresh is a great example of the power of this cycle. Rocket Fresh volumes grew more than twice the pace of overall Product Commerce this quarter. That is the result of years of investment to offer customers the best Fresh selection anywhere, at low prices, and delivered free with Dawn Delivery or same-day delivery nationwide. Fresh is a category customers buy with high frequency, and the delivery experience we provide is deeply differentiated.
It engages customers with Coupang in a way few other categories can, and we see a long runway here. On Product Commerce margins, the three temporary items we described last quarter, supply chain headwinds, fixed cost deleverage against a pre-incident demand curve, and the elevated marketing investment we are making to accelerate the return of departed customers, continue to work through the P&L on the trajectory we expect. The mechanics are the same ones we laid out before. The supply chain dislocation normalizes as our volumes rebuild, and the volume-based savings we are missing this year come back.
Capacity utilization catches up as demand better aligns with the cost base we have built, and the marketing step-up is already beginning to taper, which we expect to continue in the quarters ahead. We do not believe any of these items are structural. Some are near-term investments we have chosen to make, while others are temporary headwinds.
Our expectation in returning next year to Product Commerce margins approximating pre-incident levels is unchanged. Underneath these near-term items, work continues on the long-term margin drivers. Process improvements and deeper automation across our fulfillment and logistics network are producing higher customer service levels and lower cost to serve at the same time. This quarter, for example, we began deploying AI cameras in our packing stations to remove the need for manual scanning, paired with auto baggers that automatically size and customize packaging for each order, cutting both time and packaging waste. Each is small on its own, but they compound with the constant innovation, technology, and process excellence across our operations. Turning to Developing Offerings, where in Taiwan, the response from customers is accelerating. Our owned last-mile network now delivers the vast majority of orders next day, seven days a week.
Dawn Delivery, which we began rolling out last quarter, is already accessible to over 50% of the population. Our WOW membership program in Taiwan has grown more than 100% year-over-year and 14% quarter-over-quarter, an acceleration over the growth we saw in Q2. Our differentiated Rocket Delivery experience, combined with the value of our WOW membership program, are helping drive strong adoption and retention in customers. This is remarkably similar behavior to what we saw in Korea at the same stage. Beyond our fulfillment and logistics network, we are building what will ultimately enable the highest levels of value for our consumers through direct sourcing relationships with brands. The combination of deep selection, operational excellence, and a supply chain that enables everyday low prices and an owned delivery experience no one else matches is what has driven compounded growth for us in Korea for more than a decade.
That same combination is now taking shape in Taiwan. We are encouraged to see the trajectory in Taiwan accelerating and closely following the same path in Korea. The signals we track most closely, customer adoption, cohort retention, service differentiation, are each tracking well. Our teams are focused on scaling the offering by building the foundation for the highest levels of customer experience, while simultaneously progressing towards profitability, not one at the expense of the other. As we continue to execute with discipline, we expect Taiwan to be a meaningful contributor to consolidated cash flow in the years to come. Eats and Rocket Now, our on-demand food delivery offerings in Korea and Japan, continue to delight customers. Our focus is the same as it has been from the start, providing the best combination of service, price, and selection, and customers continue to embrace them at increasing levels.
We're also rolling out non-food on-demand delivery in Korea, extending our Eats offering into product categories where customers can now get free delivery across a wide assortment in just minutes. It's still very early, but we'll continue to iterate on how to expand the value proposition for customers. As always, we continue to take the long view on value for our customers and returns for our shareholders, and Q3 reinforces it. The long-term opportunity in Korea, in Taiwan, and across our Developing Offerings is as compelling to us today as it has ever been. Now, I'll turn the call over to Gaurav to walk through the financials in more detail.
Thanks, Bom. Our Q3 results reflect continued progress in our underlying business in line with the recovery trajectory we have outlined. Consolidated revenues grew at 7%, or 11% in constant currency, above the 8%-9% range we guided to. Before I walk through the segment results in detail, I would note that this quarter we recorded approximately $248 million in expenses relating to the fire at one of our fulfillment centers in July. These expenses were recorded as $95 million in cost of sales and $153 million in OG&A. Where relevant, I will provide results with and without the impact of the fire. Let me cover our segment performance and then move to consolidated results. In Product Commerce, segment net revenues were $8.4 billion for the quarter, growing 5% on a reported basis and 8% in constant currency.
The timing shift of Chuseok holidays we discussed last quarter negatively impacted both our Product Commerce and consolidated constant currency growth rates by an estimated 1 percentage point this quarter. Adjusting for this impact, Product Commerce constant currency growth accelerated versus Q2 to approximately 9%. Product Commerce active customers were 25.2 million, growing 2% year-over-year and quarter-over-quarter. As Bom described, WOW membership continued to build on the recovery we outlined previously, with our current members compounding their spend at normalized rates consistent with the growth rate we spoke to last quarter. Product Commerce gross profit for the quarter was $2.4 billion, and gross profit margin was 28.4%. Excluding the fire-related costs, gross profit margin was 29.5%, a contraction of approximately 260 basis points year-over-year and 100 basis points quarter-over-quarter.
The year-over-year contraction is driven primarily by the factors we discussed last quarter: supply chain headwinds and elevated levels of promotional spend we have deployed to support the ongoing recovery in customers. The quarter-over-quarter contraction primarily reflects the weather-related seasonality we often see from Q2 to Q3, as well as timing of the Chuseok holidays. We continue to view these dynamics as temporary rather than structural. We believe the long-term drivers of our margin expansion remain intact, operating efficiencies across our fulfillment and logistics network, driven in part by continued investments in automation and technology, supply chain optimization, and the scaling of our margin-accretive categories and offerings. Segment adjusted EBITDA for Product Commerce was $385 million, which excludes the FC fire costs, resulting in an adjusted EBITDA margin of 4.6%. This represents a contraction of approximately 420 basis points year-over-year and 50 basis points quarter-over-quarter.
The year-over-year contraction reflects the gross profit dynamics I just described, as well as the headwinds created by the imbalance in our fixed cost structure related to the current demand, which remains a near-term impact from the effects of last year's data incident. The Chuseok timing shift also contributed approximately 40 basis points of year-over-year contraction to both Product Commerce and consolidated adjusted EBITDA margins this quarter. Within Developing Offerings, segment net revenues were $1.6 billion, growing 23% on a reported basis and 27% in constant currency. This represents an acceleration from the 24% constant currency growth rate we reported last quarter, as we saw durable growth across each of our initiatives.
Developing Offerings generated $185 million in gross profit for the quarter, an increase of $29 million year-over-year, and segment adjusted EBITDA losses were $269 million for the quarter, an increase of $50 million over last quarter. The quarter-over-quarter increase in adjusted EBITDA losses was driven by incremental investments as we continue to scale these offerings. Moving now to our consolidated results. Total net revenues were $10 billion for the quarter, growing 7% on a reported basis and 11% in constant currency, with the Chuseok timing shift I referenced earlier accounting for roughly 1 percentage point of year-over-year drag on the constant currency growth rate. Consolidated gross profit was $2.6 billion, with a gross profit margin of 25.7%. Excluding the fire-related expenses, gross profit margin was 26.7%, a contraction of approximately 270 basis points year-over-year and 150 basis points quarter-over-quarter.
The year-over-year contraction is driven primarily by the near-term headwinds in Product Commerce I discussed, combined with our continued investments in Developing Offerings. The quarter-over-quarter contraction primarily reflects the weather-related seasonality and Chuseok holiday timing we see from Q2 to Q3. OG&A expense was $3 billion for the quarter, or 30.2% of total net revenues. Excluding the fire-related costs, OG&A expense was approximately $2.9 billion, or 28.7% of total net revenues, an increase of approximately 110 basis points year-over-year. This increase was driven by deliberate step-up in marketing spends to support the Product Commerce customer recovery and the fixed cost base still carrying capacity sized for the pre-incident demand curve. Operating loss for the quarter was $450 million. Excluding the fire-related costs, operating loss was $202 million. Loss before income taxes was $514 million, or $266 million excluding the fire-related expenses.
Our tax dynamics continue to reflect losses in our early-stage operations in Taiwan and Japan that do not generate offsetting tax benefits at the consolidated level. Net loss attributable to Coupang stockholders was $501 million, or $321 million excluding the impacts of the fire. This resulted in a diluted loss per share of $0.28, or approximately $0.18 excluding the fire. We generated $116 million in consolidated adjusted EBITDA this quarter, resulting in an adjusted EBITDA margin of 1.2%. This represents a contraction of approximately 330 basis points year-over-year, in line with the 300-400 basis points range we guided to last quarter.
The year-over-year contraction continues to be driven by the near-term items we have discussed previously, the imbalance of our fixed cost structure and current demand, the elevated marketing and promotional investment supporting the customer recovery, and the supply chain pressures that come with the lower volumes than planned. In addition, there was an approximately 40 basis points negative impact from the Chuseok timing shift I noted earlier. On cash flows, on the trailing 12-month basis, we generated operating cash flow of $845 million, and free cash flow was negative $409 million. The year-over-year declines reflect the lower profitability we have discussed this year, increases in capital expenditure, cash outlays relating to the fire, and elevated levels of tax payments. During the quarter, we also repurchased 13 million shares of Class A common stock for approximately $201 million. Turning now to our near-term outlook.
For Q4, we expect consolidated constant currency revenue growth of 13%-14%. This reflects our expectation of continued progress in our recovery in Product Commerce. For consolidated adjusted EBITDA, we expect the margin to increase 50-150 basis points quarter-over-quarter in Q4, reflecting continued underlying progress in Product Commerce, as well as reduced levels of marketing spend. We continue to expect Product Commerce adjusted EBITDA margins to return by mid-next year to levels approximating those we generated prior to the data incident. While margins may be uneven quarter-over-quarter, we believe the long-term drivers of our margin expansion beyond the recovery remain unchanged. For Developing Offerings, we previously guided to full-year losses this year of $950 million-$1 billion. We are now increasing the top end of that range by $50 million-$100 million.
This update is primarily driven by our Taiwan build-out, where we are accelerating infrastructure investments to support our differentiated delivery experience and to scale our supply chain. We continue to evaluate each investment in Developing Offerings against the same discipline we apply across the business, rigorous analysis, operational excellence, and our conviction to generating meaningful long-term returns. Operator, we are now ready to begin the Q&A.
At this time, I would like to remind everyone, in order to ask a question, press star, then number 5 on your telephone keypad. If you would like to withdraw your question, press star and number 5 once again. Please limit your questions to two per person. We will pause for just a moment to compile the Q&A roster. Your first question comes from Eric Cha with Goldman Sachs. Your line is open.
Yeah. Hi. Thank you for the opportunity. It was good to hear you reiterate your margin normalization by midpoint next year, but just had a question around the cadence of the margin improvement. Now with three quarters left and giving your guidance on the fourth quarter, it seems like within two quarters, the margin improvement should be quite substantial. Should we be expecting flattish margin throughout and then the margin stepping up all at once at the last quarter, or should we be expecting a continuous improvement up to the second quarter? Why would the dynamics work in such manner? If we were to understand the drivers and what causes that trend, even in a high level, that would be most helpful.
The second question is, just now during the prepared speech, you commented on the DO loss guidance raising the top end by $50 million-$100 million, driven by, I think, Taiwan investments. Would you be able to be a bit more specific about what sort of investment you are making in Taiwan, and what results you're expecting from this investment? I understand that you give full guidance in the fourth quarter earnings for next year, but would be most helpful if you could share your view on the general direction of Developing Offerings loss into next year, whether this year or even next year may be the peak. Thank you.
Hi, Eric. Thanks for the question. On the margin improvement cadence question, let me start by providing some context, and I'll address it more directly. As I mentioned, we expect Product Commerce adjusted EBITDA margins to return to levels approximating pre-incident levels in the middle of next year. On a consolidated basis, for Q4, we expect adjusted EBITDA margins to improve 50-150 basis points quarter-over-quarter, driven in large part by improvements in Product Commerce. We've noted the various inputs or drivers of that margin compression from reducing incremental reacquisition marketing to the imbalance between the pre-incident demand curve and our fixed cost structure continuing to narrow, working through the dislocation in our supply chain. Each of these items are temporary, not structural. We believe Q3 represents the low point for our Product Commerce adjusted EBITDA margins.
The pace of improvement that you asked about, that may vary quarter-to-quarter, given seasonal factors and the speed at which we work through each of these items. But we expect to return sometime in the middle of next year to the Product Commerce margins approximating pre-incident levels. That's really the most important thing from our view. On the second question about Taiwan investments, some more color on that. When we're at this early stage and we're growing as fast as we are, a lot of the infrastructure investments that are required for the level of service that we're trying to provide, they do involve a lot of lead time. So we do build ahead of demand. What you're seeing now is an acceleration of much of that investment in infrastructure and supply chain that really paves the path to continue the exciting acceleration that we're seeing in Taiwan.
The strong adoption that we are seeing gives us a lot of confidence, gives us more confidence by the day to support that expansion. We will keep investing on that basis, and in line as always with our operating principles.
Your next question comes from Stanley Yang with JPMorgan. Your line is now open.
Thank you for the opportunity. I have two questions. The first question is related to the DO loss guidance. Out of your DO loss guidance raise, do you consider the Japan Eats investment increase? Also, I am wondering, your funding dynamics will remain intact going forward, which means your domestic Eats cash flow will be used for this Japan Rocket Now expansion on a combined basis, like a break-even on a combined basis. Do you expect such funding dynamics to continue in 2027? My second question is about the user spending. You mentioned encouraging user spending trend for both existing and return users, indicating the spend of the existing core shows second quarter like-for-like growth of 16% in third quarter as well. Do you see such strong underlying spending growth to remain intact going into mid-next year?
Is it too optimistic to assume your Product Commerce GMV growth to exceed 16% by middle of next year in consideration of additional accretive impact from the new subscribers?
Yeah. Let me take on the second part of your question first, or the second question first. The underlying customer base in Korea continued to grow spend, as we've noted at normalized rates this quarter, consistent with the growth we spoke to last quarter. We believe that the clearest indication about the durability of that, or this continued strong growth comes from the behavior of the customers who are driving that growth. The customers who left us for an extended period and came back are spending at record levels and compounding at high rates. The customers who never left, the majority of WOW members, are doing the same. As we lap the affected periods, the cohort that has not returned stops weighing on the comparison and the Product Commerce growth, as we've explained, converges to the growth of our active customers mathematically.
What gives us a lot of confidence is that the growth that we're seeing, we believe, is driven by the exceptional value we provide across our offerings, and we continue to expand that value. That's really underneath the projections that we're making and the continued strong growth that we're seeing from our active customers today.
Yeah. I'll answer your first question, Stanley. The increase in our full-year Developing Offerings loss guidance primarily reflects the opportunity in Taiwan to accelerate infrastructure and supply chain investments ahead of the corresponding growth. The customer response we have seen recently, including an acceleration over previous quarters, creates an opportunity to pull forward some of our investments to expand the gap in our customer value proposition. We measure these investments and related returns over long term, and at times, there'll be some unevenness in the pace of spend as we lean in where the evidence is strong. Regarding Eats, our Korea and Japan offerings continue to operate at a break-even basis. We don't see that changing in the foreseeable future, but we will provide full-year guidance for Developing Offerings as part of our Q4 earnings release, that we have been doing every year.
We'll have more detail to share on our outlook for Developing Offerings next year when we report the next quarter.
Your next question will come from Seyon Park with Morgan Stanley. Your line is now open.
Hi. Can you hear me?
Yes.
Hi. Thank you for the opportunity. I have two questions. The first is relating to the elevated marketing spend and some of that being targeting returning WOW members. Of your WOW member base, I would assume there is a cohort, which, like myself, just pay the WOW membership on a monthly basis. We are not the ones that are jumping in and out. Then I would assume there are transient WOW members who opportunistically come when they feel they have reasons to purchase for that certain month, and then some will sign out. Given some of the elevated marketing that you have targeted for this segment, I would assume that segment, that number has grown.
But as you mentioned, if you start tapering off on that spend going into next year, is there some risk that those transient WOW members could see a bit of a decline as you do that? That is, I guess, the first part of my question. Secondly, I do get a lot of questions from investors whether this elevated marketing spend has any relations with an overall increasing competitive landscape overall, and whether you are sensing any changes or impact from that, or it is really more focused on the WOW members and, in general, whether it is a smaller part of the overall strategy. That is my first question, and sorry, I will just go on to my second question, if that is okay.
You mentioned a meaningful increase in the Taiwan WOW members, so just curious whether there has been any kind of specific marketing activities or events which led to that acceleration in members over there. Thank you.
Hi, Seyon. Thanks for the question. The elevated marketing spend in Korea really has been aimed at reacquiring the WOW members who left and really focusing on the WOW members who have been spending heavily on Coupang. As we've mentioned before, not only have the majority of WOW members returned, but the vast majority of that spend has returned. It really hasn't been aimed at, I think, increasing the spend. The return to the full wallet, the spend growth, I mean, really the things that are driving the vast majority of the spend on WOW, that's really driven by our customer experience, in our view. We're seeing that the incremental benefit of that elevated marketing spend, which has really been focused on reacquiring customers, diminish as we move further away from the incident, which is something that we anticipated.
That's why we've already started to bring our marketing spend down, the reacquisition spend down, and we expect that to continue. The marketing engine that's underneath remains the same as it's always been. We invest where we see clear long-term returns and let the customer experience do the heaviest lifting in bringing customers to Coupang and keeping them here. Hopefully that clarifies your question. It really is not a competitive response. We don't think the core impact of that elevated marketing spend has been increasing some short-term spend from customers. It's really been aimed at reacquiring the core customers, the core WOW members that left us during the data incident. Again, the vast majority of that spend has now returned to WOW.
Not only returned, it's returned to its full spend and beyond, now spending record levels, compounding very fast at the rates that we spoke to last quarter, which that part, we believe, has really been driven by our customer experience, which we continue to invest in as well. On Taiwan, no, there hasn't been any specific marketing event that's really been driving. I think we continue to build our core value proposition, which is increasing selection with differentiated service at low prices. As I mentioned, we are investing and accelerating investments in the underlying infrastructure and supply chain that provides the level of service that I think is driving that great customer acquisition and engagement.
Thank you.
Your next question comes from the line of Jiong Shao with Barclays. Your line is now open.
Thank you very much for taking my questions. It is great to see your Product Commerce revenue growth has been accelerating for the last three quarters and will continue to accelerate, at least in the fourth quarter. I have two questions, one on Eats, one on the margins. I am going to ask one at a time. Firstly, on the Eats, you talked about Eats now starting to deliver non-food items in Korea, but you also have the Rocket Delivery for the fresh and all that. Could you elaborate a bit on what are the use cases for Eats to deliver the non-food items that is not covered by the regular Rocket Delivery?
Related to that, but I will cheat a little bit, just on the Eats in, or potentially, do you think you need Eats in Taiwan, given you are basically using the same playbook for your e-commerce business in Taiwan? You have Eats in Korea, you have Eats even in Japan. Does it make sense for you to have Eats in Taiwan at some point? That is my first question.
Sorry, is there a second question, Jiong, or?
There is a second question, but do you prefer I just ask the second question at the same time?
Sure. Why don't you give us a second question at the same time and ?
Okay, sure. Yeah, second question is really for the margins. Going back to the margins, maybe this is more for Gaurav. Gaurav, you talked about earlier that the early Chuseok holiday had a negative 40 basis points impact in Q3. I would imagine by the same token, you should have a positive 40 basis point impact in Q4. That is 80 basis points swing. And with the exception of last year when you had a data incidence in Q4, which was not typical, but in the past, your Q4 margins tend to go up 50- 100 basis points just based on seasonality. So I was just wondering if that math is more or less kind of in the right ballpark. Is your 50- 100 basis points sort of a sequential improvement guide, sort of is on the more conservative side?
Just want to see if you can elaborate a bit on that. That is my second question.
Jiong, the first question, there is a lot of context there. I am not sure there is enough time or it is the right context to go into all the details, but the non-food delivery that we are exploring in Eats is not necessarily about selection that we do not provide on Rocket Delivery. I think there are different value propositions. It is a faster delivery time, and of course, from an operational perspective, it is a smaller batch. I think we are exploring. We continue to explore here now, how do we leverage our on-demand food delivery network to provide convenience on non-food items across a lot of merchants offline? We are incorporating a lot of partnerships with convenience stores and local merchants. They do have different value propositions. Price points will be different. The selection is not necessarily different.
In fact, the selection we provide on Rocket Delivery in many cases, in many categories could be broader even than what we are providing on on-demand food delivery. But there is a shorter lead time to receiving the product. All of this is at a very early stage, and we are approaching it thoughtfully, selectively. I think you will see the same disciplined approach that we have applied to new segments that we are exploring. We start small, we test, we look for clear validation. Here we see an opportunity to expand the customer value proposition that we are providing across all of our offerings, and we will do that with a disciplined investment strategy. I think on Taiwan, right now we are really focused on building out our retail offering, the fulfillment logistics network, the supply chain, and our operational teams and culture.
That is where our teams are focused today, and that is where the results you are seeing are coming from.
Thanks, Bom.
On your second question, we are guiding 50-150 basis points improvements in Q4, which is a quarter-over-quarter improvement. This reflects continued improvements against the three drivers that we have discussed that will continue over the next several quarters to get back to the historical PC margin levels. There are a lot of puts and takes in every quarter, but we hope to get back to our margins, entitled margins, by Q2 of next year, as we have guided to.
Thanks, Gaurav.
We will now take our last question from Wei Fang with Mizuho. Your line is now open.
Oh, thank you very much. I have two. First one on Taiwan. I noticed you guys been introducing more global selections to Taiwan recently, and just curious which side of the flywheel right now is becoming more important for the near-term engagement and growth. Is that bringing Korean and global selection to Taiwan, or you continue to building a deeper local Taiwanese marketplace? Then related to your investment in Taiwan, since you're guiding it up, how do you distinguish the portion of the investment that is building ahead of the demand and the portion of the investment that's responding to more near-term demand? My second question, a quick one on your new Try Now, Pay Later. I was curious, what metrics will you be looking at to help you determine whether you want to scale it?
I guess, do you need to see a positive incremental contribution profit before you want to expand more broadly? Thank you.
Hi. Thanks for your question. On Taiwan, that's the way we think about providing our service and selection, low prices, and our value proposition, we're really looking at where the customer finds the most value. So, I think whether it's looking at domestically produced products or internationally produced products, or products in a certain category or other category, we're really trying to provide a wide array of selection, and we respond to where customers find the most value. That's always been our philosophy, and I think we're really trying to capture the We're trying to provide our differentiated service and value across all selection over time in the full wallet. So, that's our general approach, and hopefully that answers the question. On investment in Taiwan, the infrastructure that is required to deliver the level of service we're trying to deliver does require a long lead time.
So, it's part of the design, it's part of our strategy to build a lot of that infrastructure, not only for logistics and operations, but also for the supply chain well ahead of demand. When you're talking about the growth rates we're seeing and the acceleration that we're seeing in Taiwan, it's, I think, part and parcel of the strategy to carry some excess capacity, and that you're really optimizing to build a strong foundation for that growth over trying to maximize capacity utilization at the rates that, for example, we typically have in Korea. Our priorities in Taiwan, just to perhaps give you some background, we're really focused, I should add, on building not only the physical, the operational infrastructure, but really also trying to build the discipline and culture in the organization as well to drive both growth and margin improvement at the same time.
Our priorities today have been the same that have shaped our business from the start. It is wowing customers across selection and price and service, as I mentioned earlier. Specifically right now in Taiwan, we are focused on first identifying and working closely with the right brands and suppliers that are committed to building out the long-term partnership and the right experience for our customers. Second, we are continuing to expand our owned infrastructure and last-mile delivery network, our fulfillment and last-mile delivery network, which now delivers the vast majority of our volume next day, seven days a week. Also, I do not know if you were able to catch it, but we are extending our Dawn Delivery service. We just launched that a quarter ago. Now it covers half of the population in Taiwan.
Finally, we are building, as I mentioned, a culture of operational excellence, which enables us to execute with discipline and innovate with speed. Those are the keys to delivering the best experience and generating attractive returns for our shareholders. It is really the underlying true infrastructure that allowed us for years in Korea, up to the data incident, to deliver both strong compounding double-digit growth and expanding margins. That is our focus today and our priorities.
We are really excited about the long-term opportunity in Taiwan, and we continue to invest with that long-term framework into a business that we believe will be a significant generator of free cash flow and a material contributor to our free cash flow in the near to mid-term. On the Try Now, Pay Later. Sorry, on Try Now, Pay Later, that is just an example of one of many innovations that we are testing out.
We have scores of these across our business. Some of these are in constant iteration. We are testing now. Our philosophy is always to test small, to iterate on the defects, to look for validation on the consumer and our assumptions. Yes, they have to be validated across a number of things. They have to be validated on the customer value proposition front, on the customer adoption front, customer engagement front, on the unit economics front, on the scalability front. I mean, there are many assumptions here that we look for clear validation for. Of course, very importantly, whether this is going to generate the long-term cash flows and the returns from a shareholder perspective that will justify the investment. So, we look for customer, operational, and shareholder ROI lens.
We are looking for validation on assumptions across those types of lenses before we scale any investment in initiatives like Try Now, Pay Later.
This concludes today's conference call. Thank you, and you may now disconnect.