Hello, everyone, and welcome to the Mercado Libre Earnings Conference Call for the Quarter Ended September 30th, 2019. I am Federico Sandler, Investor Relations Officer for Mercado Libre. Our senior manager presenting today is Pedro Arnt, Chief Financial Officer. Additionally, Osvaldo Giménez, CEO of Mercado Pago, will be available during today's Q&A session. This conference call is also being broadcasted over the internet and is available through the Investor Relations section of our website. I remind you that management may make forward-looking statements relating to such matters as continued growth prospects for the company, industry trends, and product and technology initiatives. These statements are based on currently available information and on our current assumptions, expectations, and projections about future events. While we believe that our assumptions, expectations, and projections are reasonable in view of the currently available information, you are cautioned not to place undue reliance on these forward-looking statements.
Our actual results may differ materially from those discussed in this call for a variety of reasons, including those described in the forward-looking statements and risk factors section of our 10-K and other filings with the Securities and Exchange Commission, which are available on our Investor Relations website. Finally, I would like to remind you that during the course of this conference call, we may discuss some non-GAAP measures. A reconciliation of those measures to the nearest comparable GAAP measures can be found in our third quarter 2019 earnings press release, available on our Investor Relations website. Now, let me turn the call over to Pedro.
Thank you, Federico. Hi, everyone, and thanks for joining our third quarter 2019 earnings call. I want to kick off by saying that we've delivered another quarter of robust results. We've made significant progress in executing against our strategic roadmap to not only deliver innovative technology solutions to our customers, but also to continue democratizing e-commerce and digital financial services. With that said, let me start with our fintech progress report. During the quarter, we reached an important historical milestone, as for the first time ever, total payment volume away from Mercado Libre's marketplaces surpassed TPV on marketplaces during a full quarter, and did so not only on a consolidated basis, but also in our largest market, Brazil.
As of this third quarter, two of our main countries are processing more payment volume away from our e-commerce platforms than on them, as Argentina and Brazil TPV off marketplaces reached 63% and 52% of total TPV respectively. Driven by the successful execution of all our off-platform initiatives, TPV off-platform U.S. dollar growth accelerated to 140% year-on-year for the quarter. On a service by service level, starting with our merchant services business, the business continues to maintain momentum. During the quarter, merchant services once again surpassed $1 billion of TPV on a consolidated basis. Argentina and Brazil were highlights during the quarter as our gateway solution scales, as well as we execute on market fit solutions for both larger merchants and long-tail merchants.
Moving on to our mPOS service, the largest of our payments businesses off platform in terms of TPV, we're pleased to report that it continues to grow strongly, with quarterly active merchants already surpassing the two million mark on a consolidated basis. Additionally, transactions per unique active device continue to increase as usage grows in both Brazil, Argentina, as well as in Mexico. In Brazil, for mPOS, we also launched our Point Pro device during the quarter, which has been gaining traction driven by our recent branding campaigns as we begin to observe meaningful improvements in top of mind and spontaneous awareness in surveys we have recently performed post-campaign. Our Point Pro device will enable us to pursue slightly larger merchants, while in parallel, we have also launched a third device, the Point Mini, which will allow us to better serve the needs of our micro merchant segment of users.
In Mexico, our mPOS business is also gaining traction as we scale out branding and communication campaigns. Devices sold in Mexico grew by 58.5% quarter-over-quarter and almost 200% year-over-year, while active collectors during the quarter continued to grow north of 300% year-over-year. Moving on to our wallet services, I am pleased to report that we have also reached an important milestone during the quarter. For the first time since launch, wallet TPV almost reached $1 billion on a consolidated basis. On an FX neutral basis, TPV grew by a factor of over 4x year-over-year. Additionally, consolidated quarterly active payers on the wallet grew a robust 35% quarter-over-quarter, reaching 6.1 million active payers.
Argentina was a highlight during the quarter for our wallet business as we see positive results across most key KPIs. One of these is frequency of payment per unique payer. This metric reached almost 11 payments per unique payer during the quarter, driven by successful execution in the rollout of our in-store QR network. A testament to this is that in spite of the FX headwinds that occurred during the quarter in the country, spend per unique payers per quarter continues to rise in U.S. dollars, even reaching $200 per quarter. A significant part of this wallet success is driven, as I just mentioned, by in-store QR payments. During the quarter, we doubled the amount of QR payers and QR collectors versus the prior quarter, reaching two million and almost one million respectively. Additionally, in-store QR payments in Argentina already represent over 50% of total wallet TPV for the quarter.
As we advance with our payment strategy, we aspire to replicate the success we are seeing in Argentina across other markets in the region that have launched at a later date. Still on fintech, I want to take a moment to update you on our credit business, Mercado Crédito. Overall, the merchant credit business continues to grow at a healthy clip in terms of originations and is contributing to diversifying our revenue streams while generating more value-added services to our merchant base. We continue making inroads in growing our merchant credit business in Brazil and in Mexico. During the quarter, as originations grew 118% year-on-year and 216% year-on-year respectively. Additionally, in Mexico, as we strengthen our scoring algorithms and have more data on our merchants, we have increased loan terms while still maintaining very low levels of uncollectability.
In Argentina, as a consequence of our strong FX headwinds and increasing local benchmark interest rates that followed during the quarter, our overall merchant loan portfolio is down on a sequential basis when measured both in U.S. dollars as well as in terms of credit originations. On the consumer credit side, on-platform consumer credits in Argentina continue to show resilience, giving us confidence that our scoring models continue to strengthen, as we do not observe increases in non-performing loans despite the deteriorating macro conditions and higher benchmark rates. In Brazil, however, we are still observing higher levels of bad debt as models continue to improve and older vintages start to seek repeat loans. Additionally, I am pleased to report that during the quarter, we launched our consumer credit business in Mexico, which should help us to further improve our value proposition to buyers in that country.
We've also hit strides in funding sources for Mercado Crédito as we continue advancing in lowering funding costs and driving scale. In Argentina, we issued a public market securitization under the SME FIGER, which has allowed us to lower funding costs, while in Brazil, we launched our first consumer credit trust, also generating savings and funding that we will be able to pass through our buyers and sellers in the form of lower interest rates. Before we delve into the rest of the prepared remarks, I'd like to first make a few comments in regards to the increase in bad debt stemming from our credit business. During the quarter, bad debt increased by $13.3 million, an increase of 86% versus last quarter. This increase is explained for the most part by two of our more recently launched products in Brazil, consumer credits and mPOS merchant credits.
Given that we are at such an early stage in both of these products, the higher level of loan losses are within our expectations. It's important to highlight that both in consumer credit and mPOS credits in Brazil, we are taking the appropriate measures to improve these loan losses going forward. We have adjusted pricing, taken advantage of incremental information we collect on non-performing payers to strengthen our churn and behavioral credit scoring algorithms, and have refocused on performing payers. That wraps up the fintech progress report. We still have much to execute in the back half of the year and beyond. It is imperative that we continually innovate and deepen our customer engagement in order to maintain and try to further widen the distance between our value proposition and that of our competitors.
In keeping with that focus, the third quarter saw significant progress in consumer and merchant engagement on the Mercado Pago platform. Let's now move on to some of the high points from our Marketplace business. Consolidated GMV accelerated for the third consecutive quarter to 37% year-on-year on an FX neutral basis, driven by solid execution in Argentina, Mexico, Colombia, and Chile. Despite macro headwinds, Argentina delivered the best quarter since Q2 2015 on an FX neutral basis and continues to grow meaningfully above inflation. During the quarter, Argentina accelerated GMV on an FX neutral basis to 80.2% year-on-year. Moving on to Mexico, it continues to be one of our fastest-growing markets, accelerating GMV growth on an FX neutral basis to 47% year-on-year. Continuing to gain market share and growing well above market growth rates.
Additionally, we continue to see higher net promoter scores as we improve our user experience and bring down delivery times and costs year-over-year as our managed logistics network expands. Colombia FX neutral GMV grew at the fastest pace over the last 28 quarters, accelerating to 50%, while Chile also accelerated to deliver the best quarter since the beginning of 2018, delivering FX neutral GMV growth of 32% year-on-year. Incremental marketing investments, greater traffic, and improving conversion rates explain the results delivered in the latter countries. Lastly, a recap of Brazil during the quarter, where results were uneven. FX neutral GMV growth came in at 24% year-on-year. The deceleration was driven in part due to a Correios strike, which created approximately 200 basis points of quarterly headwind during the month of September.
Initiatives to cap free GMV to 2% of GMV versus 5% last year also had a negative impact on growth. Despite the deceleration, incremental GMV added during Q3 2019 when compared to GMV of the prior year during the same quarter, was still a robust $378.6 million, among the highest in recent quarters. In line with that, we're also encouraged to see unique buyers accelerating for the second consecutive quarter in Brazil to 25% growth year-on-year, while new buyers deliver the fastest pace of growth in over a year. In Brazil, selection continues to deepen as we reached 120.4 million listings during the quarter, another indicator that we continue to execute well on key KPIs.
Demand metrics for other countries continue to trend well as unique buyers reached 26% year-on-year growth on a consolidated basis, while Argentina, Mexico, Chile, and Colombia all accelerated sequentially year-on-year to 22%, 50%, 38%, and 53% growth respectively. Also, new buyers are accelerating in all geographies as we begin to invest more aggressively in marketing. On a consolidated basis, new buyers grew a robust 10 percentage points versus last quarter. On the supply side, we also continue to focus on deepening selection available to our buyers through our cross-border initiatives. As cross-border trade seller base and pipeline continues to grow and we integrate our international sellers into our logistics platforms. Fulfillment by MELI for cross-border is operational in Mexico, and we continue to strengthen partnerships with international carriers. Additionally, we are starting to deploy our international seller supply into Brazil with early positive results.
Moving on to another key strategic priority for us and enabler of our enhanced marketplace vision, logistics. Our own logistics network continued to grow during the quarter on a consolidated basis, reaching 1/3 of all items shipped and gaining almost 20 percentage points of adoption versus last year and 7 percentage points versus last quarter. In Brazil, reliance on drop shipping network continues to come down, reaching 71% of items shipped versus 90% last year. A clear reflection that our own network continues to grow and gain share of items shipped. As of Q3 2019, this managed network reached almost 30% of items shipped, while fulfillment penetration grew to 7%, up from last quarter. In Mexico, fulfillment penetration continues to move along full steam. During the quarter, it reached 35% of items shipped, growing penetration 6 percentage points versus the prior quarter and almost 30 percentage points versus Q3 2018.
This successful execution is also allowing increase by 2 percentage points the amount of shipments in two days or less in Mexico versus last year. Argentina is making progress as well on the logistics front. Mercado Envios penetration reached almost 70% of items shipped, while fulfillment reached 5% of items within a quarter of launching the service. Another positive data point from Argentina is that our flex logistics platform continues to fire on all cylinders and reached 10% of items shipped during the quarter, meaningfully contributing to increase by 23 percentage points the amount of shipments in two days or less in the country versus last year. Now that I've covered the main highlights and business KPIs for the quarter, let's move on to financials, where we have continued accelerating the pace of investment in our growth initiatives, mainly in marketing, as we move into the second half of 2019.
During the third quarter, gross billings continued to maintain strong momentum, growing on an FX neutral basis above 60% for the third consecutive quarter, while also accelerating in U.S. dollars to 45.4% year-on-year, despite FX headwinds in some of our main countries. On a by country basis, gross billings delivered excellent performance as well on an FX neutral basis. Brazil growing at 43.5% year-on-year, Argentina at 118.1%, and Mexico at 104.4% year-on-year.
Consolidated net revenues grew faster than gross billings, both on an FX neutral basis and U.S. dollar basis, growing to 91% year-on-year and 70% respectively, and reaching $603 million as we continue investing behind free shipping, loyalty, and optimized subsidies, which have minimized contra revenues. Argentina was a highlight as net revenues accelerated sequentially over 20 percentage points in dollars to 38.8% year-on-year during the quarter, despite a devaluation of 13% of the Argentine peso versus the prior quarter.
On an FX neutral basis, revenue accelerated again to 118.9% year-on-year, demonstrating the resiliency of our business in that country. Net revenues in Brazil and Mexico also continue to grow at a very good clip, both in U.S. Dollars and on an FX neutral basis. Net revenues in Brazil accelerated to 76.6% in dollars, while on an FX neutral basis it grew 77.3%. Mexican net revenues grew triple digits in U.S. dollars for the fourth consecutive quarter to 146.4% year-on-year, an equally impressive 152.5% on an FX neutral basis. Gross profit was $284.3 million, representing 47.2% of revenues during the quarter, and relatively flat versus 47.8% a year ago. This 62 basis point margin compression was driven for the most part by shipping subsidies and warehousing costs of our managed network, which was partially offset by collection fees, sales taxes, and hosting fees.
On a sequential basis, gross profit was $284.3 million, representing 47.2% of revenues during the quarter versus 50% during Q2 of 2019. This 281 basis point margin compression is explained for the most part by incremental inventory costs from the robust sales of mPOS devices during the quarter and increased shipping subsidies to promote adoption of our logistics network. We've included a detailed breakdown as we do every quarter of these, and also the OpEx margin evolution I am about to cover in the slides that accompany this presentation. As reported, operating expenses ascended to $366.3 million or 60.7% of revenues versus 50.9% during the third quarter of 2018. On a sequential basis, operating expenses increased $81.4 million, which resulted in sequential margin compression of 848 basis points, mostly attributed to incremental marketing expenses and bad debt, as I've previously discussed.
From these 848 basis points of sequential margin compression in OpEx, 689 basis points are explained by the incremental marketing expense, while 194 basis points are explained by growth, mainly in bad debt from our credit business in Brazil. Both of these partially offset by 101 basis points of scale in salaries and wages in G&A, given that a significant portion of our G&A headcount is in Argentina. The step-up in marketing is explained almost entirely, 90% of it, to the launch of our branding campaigns in our main countries as we continue to strengthen our e-commerce brand. Perhaps more importantly, as we begin to build the Mercado Pago brand and begin to communicate the benefits of our payments ecosystem to our users through more traditional marketing channels.
It's important to highlight that although the volume and revenue returns on these branding initiatives will play out more in the midterm, we've already began to see increases in unaided brand awareness of MELI in the most recent brand tracking surveys that we've conducted. If we break down that $51.3 million of incremental marketing QoQ during this quarter, the vast majority, $46.2 million, were deployed to branding initiatives, 60% of those for the marketplace and 40% of those for payments. As a result of these incremental investments I've walked you through, operating losses ascended to $81.9 million. The 848 basis point contraction that I just explained, plus the 281 basis point gross margin contraction covered earlier in COGS, explain the sequential decline and also the difference between positive EBIT and negative EBIT.
Moving down the P&L, we saw $14.5 million in financial expenses attributed for the most part to interest accrual on our convertible note due 2028 and financial guarantees in Argentina. Interest income increased by $229.5 million year-on-year to $28.5 million, mainly attributable to the proceeds from the convertible note issued in August 2018 on our follow-on offering earlier this year. Which both generated more invested volume and interest gain, and also due to higher float in Brazil and Argentina from our payments business. Forex gain was $987,000, primarily as a result of strengthening the U.S. dollar over our Argentine peso net liability position in Argentina. We recognized a valuation allowance on deferred tax assets in Mexico and Colombia, which accounted for $91.5 million and $7.2 million respectively. We still anticipate the possibility of eventually being able to use these tax credits in the future.
Should we be able to use the valuation allowances, they will be recorded as P&L gains in future periods. As a consequence of all this, net loss for the quarter ascended to $146.1 million, explained for the most part by the incremental investments in marketing and the aforementioned tax valuation allowances. On a per-share basis, all this resulted in a basic net loss per share of $2.96. Free cash flow, defined as cash from operating activities less payment for the acquisition of property, equipment, and intangible assets, net of cash acquired, was $124.1 million versus $74.4 million in the same period last year. Cash, short-term investments, and long-term investments totaled $3.1 billion. Reflecting on the first three quarters of 2019, we remain very encouraged by the performance of our business overall and remain excited about the opportunities that lay ahead of us.
We believe we are building superior experiences and products for our consumers and merchants, and the sustained momentum we see in the business gives us confidence to continue investing. We look forward to keeping you updated on our progress next quarter, and for now, we can take your questions. Thank you.
Thank you. As a reminder, to ask a question, you will need to press star one on your telephone. To withdraw your question, press the pound key. Please stand by while we compile the Q&A roster. Our first question comes from Stephen Ju with Credit Suisse. Your line is now open.
Okay, thanks. Pedro, there's a ton of stuff to focus on, given you guys are marching in multiple different directions. I'll just pick one. The Pro mobile point-of-sale device that you guys just introduced, can you talk about whether the merchants that you're targeting there are existing Mercado Libre merchants, or are these a completely different set of new merchants? Just wondering what the acquisition cost for these merchants might be on a go-forward basis. When you release the mini version also, I guess this is probably the version that's more appropriate for the merchants that would otherwise be on the marketplace. Okay, thanks.
Thank you. Our next question comes from.
Hi.
Hi, Stephen, this is Osvaldo. Thanks for the question. With regards to Point Pro, what we are addressing is really small businesses. In the past, with our Point Mini device, we were targeting mostly the end of the long tail, which were mostly individuals. Now we're reaching to slightly larger businesses. Here, even though we are promoting it in our platform, I would say most of the new users are not necessarily already on the Mercado Libre platform. We are using a sales force, and we're reaching out through marketing to these specific merchants. We have not, and we are not disclosing the acquisition cost, but we are very comfortable that the price is very appropriate, and the payout period is reasonable.
We also introduced a Point Mini chip, which is a device that is targeted on the smaller merchants, but that already has a telephone chip in it, so that it does not need to be paired with a telephone. We also introduced that during the last quarter. These are addressing two different segments of the market. As I was saying before, the smaller individuals and then slightly larger merchants.
Thank you. Our next question comes from Deepak Mathivanan with Barclays. Your line is now open.
Great. Thanks for taking the questions. Pedro, can you talk about some of the initiatives that's driving use case and frequency growth on the wallet, perhaps in terms of the merchant type on the in-store side? Without going into specifics, how are you planning to attack this aggressively in 2020 and beyond? The in-store QR wallet payments is growing really fast, particularly in Argentina. How do you think about the monetization plan for this long term? What are the primary merchants that's seeing the usage right now? Thank you.
Hi, Deepak, this is Osvaldo, and let me take the first part of the question. With regards to use cases and increase in frequency of users, I think the two main drivers have been, on the one hand, adding use cases in each of the countries where we operate, and the other one is adding capillarity for the QR code networks. Networks have been growing in all of Argentina, Brazil, and Mexico. The things we try to do is encourage users to
Do more transactions, but also to use more than one different flow. We find that retention rates are higher when users engage in more than one of the use cases. Towards the later part of the question, you ask about monetization plans, we have already made public to our users in Argentina that we will start charging 0.6% for stored balance transactions and for debit card transactions, starting during Q3. 0.6%.
If I can just follow up on that, what is the size of that business right now, roughly, and can you provide on kind of growth characteristics of it? Thank you.
Great. I think what we disclosed in the script was that that business was approaching $1 billion during the quarter. It's growing at a very robust rate. It's growing over 4x.
Okay. Thanks, Pedro.
Thank you. Our next question comes from Robert Ford with Bank of America. Your line is now open.
Hey, good evening, and thanks for taking my question. Osvaldo, can you talk a little bit about how rapidly your lending algorithms are evolving for consumers and mPOS clients in Brazil, and how you think about the growth of the credit book in Brazil as a whole?
Great. You have the disclosures around originations. We'll get you the number in a second. Let me just address the evolution of the algorithms. Whenever we launch something new, obviously there's a little time that it takes for the algorithms to improve. We began to see strong improvements in the third quarter that hopefully we see them reflected as we move forward in the year. We are seeing, and we're comfortable with how the algorithms are performing. In terms of originations, when we look at Brazil, originations have been fairly stable. Overall originations QoQ, slightly down, and the same applies for the consumer book. In terms of incremental originations, it's been similar size Q2, Q3 in the range of incremental $30 million, Brazil.
Pedro, you dictate that, right? This is not a demand issue. You're determining how quickly you grow those originations, correct?
We do, and if you look at our history, we've been very disciplined. Either if macro conditions change or if we think the algorithms still need to get more robust, we can slow down origination rhythms, and we have done that in different quarters. This last quarter, for example, in Argentina, you'll see a strong slowdown for obvious reasons. We continue to take a long-term view on this. We continue to be very, I think, cautious in making sure we're being efficient, and we entirely determine how much we open the spigot or close the spigot in terms of the origination volume.
Understood. I think Wish is one of the fastest-growing sites in Brazil. How are you thinking about the cross-border business there, and how are you balancing user experience with assortment growth and then GMV growth?
Yeah. I think on CBT, really what we've been more focused on recently is how do we offer a CBT offering that's differentiated from some of the other global players, and a lot of that is focused on the user experience, both for sellers and buyers. We're working on how do we really streamline the whole process in terms of shipping, customs, trying to figure out if there are ways to get inventory into country faster and therefore to your doorstep faster once you've ordered. I think that's been one of the reasons why growth there hasn't been explosive, because we've really tried to focus on building out the right user experience. We've made consistent and solid strides on that.
I think hopefully when we look back in a few years, that will have become a sizable business with a user experience and net promoter scores that clearly differentiate it from some of the faster-growing Wish or other players that focus more just on getting demand from China to Brazil and not so much on the user experience on the way.
Very helpful. Thank you.
Thank you. Our next question comes from Mike Olson with Piper Jaffray. Your line is now open.
Hey, good afternoon. I know you don't provide specific go-forward guidance, but Q3 was a fairly dramatic change in operating loss. Just wondering, how do you think about kind of the near to medium-term operating loss? Should the quarterly loss run rate kind of be more like Q2 or more like Q3 or somewhere in between? Would you say the company's still focused on profitable growth as you've described before, and if so, what kind of timeline to get back towards that?
What we need to make sure that we balance is a sustainable business with making sure that we're also capturing opportunities that exist in front of us and that we're aware of what the competitive dynamics are. I think if you look at the quarterly disclosures, you'll see that we've tried to give a lot of indication on Q3 on where the incremental margin compression comes from. You'll see a lot of that is either increased investment in customer acquisition and marketing for the entire wallet and payment strategy across multiple geos, which is something that we do want to invest in because we see an enormous TAM and a very large opportunity. In some cases, with longer payback times on merchants we acquire, or in the case of wallet, an enormous opportunity with a very good proof of concept in Argentina, but still in a market launch monetization model.
On the marketplace, I think we've identified an opportunity to increment our brand awareness and our brand equity. You'll see in the disclosures that there has been almost the entirety of the incremental spend sequentially is on marketing. Marketing is something that's very easy for us to control and that we also think scales well going forward if we sustain these levels of revenue growth, while still being a material marketing budget to continue to consolidate our leadership.
Thank you.
Thank you. Our next question comes from Marcelo Santos with JP Morgan. Your line is now open.
Hi. Good evening. Thanks for taking the question. The first question I'd like to make is could you please give some description of the more or less ballpark number and profitability levels for the different businesses you have, like general terms? I think you did this in the past, and just wanted an update now. The second question is regarding the mPOS. Do you plan or do you offer already some sort of software together with payments? Is that something that's on the pipeline to help the merchants together with the payments? These are the two questions.
Okay. Very quickly. Relative profitability by the different businesses, we don't disclose. I think at times we've given overall indication, but given how competitive the situation is getting, I think you have very good disclosures per segment, not by business line. Obviously these businesses are in very different states of growth and development. I would argue that the current P&Ls and what they look like is by no means an indication of what those businesses could deliver in terms of margins and P&L at scale. Again, just to reiterate, if you look at many of the fintech initiatives right now, it's more about growth in TPV and customer acquisition and making sure that the customer engagement and customer user metrics are going in the right direction. When we hit a certain level of scale, we start introducing the monetization model.
If you look at Argentina, as Osvaldo just mentioned, we're going to be doing that this quarter. I think that reiterates our commitment to monetizing the wallet and the fintech initiatives when the time is right. That right time could be in the relatively short term as what we're doing in Argentina.
Thank you.
Thank you. Our next question comes from Ravi Jain with HSBC. Your line is now open.
Hi. A couple of quick questions. The first one on Brazil on the GMV growth. Do you think is there room to kind of accelerate that in 2020 in terms of especially given the competitive landscape is getting even intense, local peers are raising money and international players are trying to also expand that offering. How do you see MELI's positioning and your strategy? Do you think you want to accelerate your logistics build-out? Do you think that's going to be the key thing or do you want to just continue investing in branding and that's how you plan to attack the competition? The second question is, do you see some synergy or the potential for bringing the two businesses together?
Are the users of the wallet becoming more and more buyers on your e-commerce platform or should we think that the e-commerce buyers are your first adopters of the wallet? How do you think about cross-selling between the two businesses? Thank you.
Okay. Sorry, just one thing. I got cut off before finishing the answer to the previous one. Very quickly, in terms of software for the mPOS, I think when you look at the mPOS devices that we distribute and the merchant base and the multiplicity of services and products that we're pursuing, currently we don't have as a high priority software or ERP-like solutions. I think we're much more focused on the multiple other financial services that we're offering and not the ERP business. We do look to integrate with existing ERPs, like the Linx partnership we announced last quarter. In general, it's not a focus of us to build our own core ERP. Brazil. I don't think it's an either/or question. We're very focused on the rollout of our logistics platform.
We're actually extremely pleased with the results we're seeing there in terms of migrating more and more volume onto our own network and also the sophistication and the results we're seeing on that network. That will be a core component of our differentiation and our value prop in Brazil. That drives a lot of OpEx through the P&L. I think because we're seeing incremental improvements in net promoter scores and engagement metrics, that also gives us greater encouragement to also invest more in marketing to communicate some of those new services and new brand attributes, but just in general to attract more visitors and buyers to the category. I think we're being more aggressive on the marketing front, incrementally and sequentially. It's not in detriment of investments that we're making in user experience or technology or fulfillment.
Again, we don't give forward guidance on growth rates, but I do think that we hope that all these investments could lead to a better user experience, and hopefully that leads to more incremental growth. In terms of ecosystem and platform, obviously, we believe that that's one of, if not the biggest differentiator that we have and we need to focus on. I think when we look at it today, there's probably more that we leverage the existing user base of e-commerce to drive wallet and fintech usage. However, I think it's roughly split 50/50. 50% of monthly active payers have some sort of activity on the marketplace, and 50% are just net new fintech users that hopefully over time we are able to bring onto the marketplace.
As we launch our revamped Mercado Puntos loyalty program over the next few quarters, that should be instrumental in driving more and more cross-usage and cross-selling of our entire ecosystem.
Thank you. That's helpful.
Thank you. Our next question comes from Gustavo Oliveira with UBS. Your line is now open.
Hello, Pedro. Thank you for taking my question. I wanted to understand if still the Brazil GMV growth, you've been reducing your shipping subsidies as you grew more comfortable with your algorithm. Do you have any intention in increasing again the subsidies, or you think you already found the optimal level that you want to work with going forward? Whether you prefer to invest more of that resource in the brand investments?
Yeah. Look, obviously, we are always innovating and always thinking of ways to drive better user experience, more volume. I think MELI is always about potentially changing things. Right now, I think we have reduced, versus prior years, the level of subsidies, as a percentage of revenue. We also feel that the subsidies we offer now are a lot more intelligent and a lot more targeted. If you look at it sequentially, there was actually a slight increase in subsidies, not so much on transportation, but subsidies aimed at getting merchants to send more inventory to our fulfillment centers. That's had very positive results in terms of growth of the managed network, fulfillment in Brazil has began to grow again, as you saw in the numbers. Again, I want to stress, I don't think this is a trade-off between marketing or shipping.
I think a lot of the marketing spend has been incremental, and that's what's driven the change in the P&L profile, and we're confident that that's the right thing to do for this phase of the business, where we need to invest in growth of fintech and user acquisition, and where we see an improving user experience on the marketplace, and we want to invest behind that. That's, I think, the way we're looking at the incremental marketing investment that was in the P&L in Q3, going forward, at least for this phase.
Thank you, Pedro.
Thank you. Our next question comes from Irma Sgarz with Goldman Sachs. Your line is now open.
Yes. Hi. Thanks for taking my question. Regarding the managed network where you made a lot of progress over a quarter basis, as you look out further into the future and you continue to sort of deepen your footprint, there's two questions here. Firstly, what do you think in terms of what you need in terms of footprint? Is it more a question of getting additional distribution center space or maybe increasing more hubs and increasing potentially even the level of automation in your network?
Secondly, as you sort of think, again, further out and look through your merchant base, do you have any plans for also offering additional services where you help your merchants directly target their local clients through sort of click-and-collect initiatives within their stores, where you could just offer the interface on the marketplace and connect buyers and sellers, also a solution on the logistics, but where it doesn't really go through your fulfillment or your cross-docking, but offers basically a local transportation solution? Thank you.
Okay. I think most of the elements you included in your question at the beginning, in terms of what other additions to the managed network we believe we will continue to see in terms of incremental warehouses, incremental service centers and hubs, and increased automation, I think the answer is yes. Remember that in our model, warehouses and incremental services centers, those are OpEx, they're not CapEx. Incremental automation, depending on what it is CapEx, but that's a very manageable number. All of those are part of our network plans for all the countries where we're building out the managed network. In terms of click and collect, Irma, the functionality does exist, so we can work with select retailers on click and collect.
Having said that, however, I think our focus, not being a bricks-and-mortar retailer who has sunk cost into buildings, is much more on building out the fastest and most efficient network to get packages to your doorstep, with potentially the overlay, and we began to do some of this in Brazil, of drop-off points and eventually pickup points, but not so much storefronts of our merchants, but rather nodes within our network where drop-off and pickup can occur. Our focus, I think, is more of a pure native e-commerce player for now that isn't trying to leverage existing physical stores and more on building out the efficiency and the speed of our own network. We can work with select retailers who want to on offering click and collect on what they sell on the platform, but that hasn't been a focus.
Great. Thank you.
Thank you. Our next question comes from Marvin Fong with BTIG. Your line is now open.
Hi. Thank you for taking my question. My first question is just on the marketing spend. I'm looking at the slide in your presentation, which is very helpful, just breaking it down between branding, performance, and promotion. I was a little surprised perhaps that so much as about 60% looks like is on branding, the majority of that on the marketplace, given you're such a well-known brand already. Could you just talk about the decision behind your marketing allocation, and do you find that in terms of trying to drive first-time users onto the digital wallet, that promotions and performance marketing aren't good ways to do that? Thanks.
Yeah. A couple of things. First of all, these include both commerce and fintech branding investments. Bear in mind that you should look at the branding investment probably on an annualized basis to have a better sense of overall percentage of revenues. We've concentrated a lot of the investments in Q3 and a little of that in Q4. That doesn't necessarily signal that that is an ongoing quarterly amount. Although we have a very strong brand, as e-commerce becomes more and more mainstream, we still see opportunities to drive more user and more top of mind behind the Mercado Libre brand. Like I said earlier, part of what we're communicating there are also some of the newer attributes we have. The speed of our delivery network, the prevalence of free shipping, and other newer benefits of the marketplace as it's really improved its service.
On the fintech piece, if you were to look, we do have to build out the Mercado Pago brand. If you look at some of these markets, Mercado Libre is incredibly well known. Mercado Pago as a brand was historically on marketplace, there is room to start building the standalone knowledge behind Mercado Pago. It's the first time we've ever done any brand building for Mercado Pago. It's always been known and used primarily on the marketplace. We do agree with you that for customer engagement and customer acquisition, promotional and targeted discounts are very effective, that's really where most of the promotional budget comes in, is for the fintech piece.
Because we've never generated any awareness around the brand and there are other competitors that have invested brand marketing behind their brands, I think there is room to do that as well, and we're pleased with both the quality of the campaigns and some of the initial results we're seeing in terms of incremental brand awareness and top of mind.
Great. As my follow-up question, just on the decision to start charging a transaction fee on the digital wallet in Argentina, I believe you said. Could you just help us with that thought process? Why do it now when you're still at an early stage of adoption? Do you feel like you can start charging the transaction fee and it won't slow growth? Or is it more that you've decided to increase or drive more profitability and slow down growth a little? If you could just comment on that'd be great. Thank you.
Hi, Marvin, this is Osvaldo. I think it's two things. On the one hand, we were offering an unbelievably good value proposition. We continue to believe that after this price increase, this will be a hell of a good value proposition. It will continue to be the cheapest electronic payment method in the market. We will charge 0.6% for both store balance and debit cards, and this is cheaper than the going rate everybody pays for debit cards in Argentina, which is 0.9%. We still believe it's a huge value proposition, but we also think that it was worth doing it as a proof of concept and to test the market and see how they are okay with starting to pay transaction fees for QR code payments.
Yeah. Just complementing that, and I think this is important, is we're committed to investing behind a business if we know that there's actually an attractive business behind that. We felt that in Argentina we had enough traction with the free product that it was time to start monetizing and also to make sure that we start building out a business that's sustainable and profit over the long term. I think that's a reflection of how we've always approached our businesses, and that hasn't changed. What's changed, I think, is just our desire to use the scale we have and the capital we have to invest aggressively to really gain users, and then to, as rapidly as we can without hurting that long-term growth, beginning to monetize and actually build out a sustainable business. That's what we do in credit.
It's what we're beginning to do in QR. It's what we do in mPOS. It's what we've always done in merchant services.
Great. Thank you, guys.
Thank you. Our next question comes from Tom Champion with Cowen. Your line is now open.
Hi. Good afternoon, guys. Thanks for taking the question. Just to ask, can you comment on any marketplace buyer changes in Brazil since the launch of Prime in mid-September? Also, what's the status of the Correios strike? Has that been resolved or does it remain ongoing, or did it flow into 4Q? Then on the payment side, the QR network appears to be a really important frequency lever with the wallet in Argentina, and just curious if it's available at this point in Brazil. Thank you.
Look, I think we haven't seen any changes in our business attributable to anything that's happened on the Prime front. It's very early stage. I would say we compete with them very aggressively and head-on in Mexico. I think we like what we're seeing from how our business continues to perform there. Brazil, I would say there's a very big difference between the business we have and the business they're running. I don't think we attribute anything to whatever was launched by a competitor during the quarter. Correios strike is over. Obviously, it had an impact on our business. I think we said in the ballpark of 2% of GMV growth in the quarter. It was only a few days. Obviously the impact in the month of September was larger than that.
I think the silver lining to this is unlike the last time Correios had a strike, our managed network, we were able to move volume away from Correios towards the managed network. That's one of the reasons why we saw the strong improvement in managed network adoption in Brazil during the quarter. Increasingly going forward, every time this happens again, I think we will be better and better prepared to simply move volume away from whoever has an operational complication to other carriers. We can already do that in Argentina and Mexico quite well. In Brazil, there's still significant reliance on Correios, but that's waning at a very consistent pace month-on-month.
In regards to QR code network, as you mentioned, Tom, definitely has been a frequency lever in Argentina, and we have seen that across the Mercado Pago ecosystem, that those users who use QR codes do more transactions every month. In Brazil, what has happened so far is, remember as we launched during the second quarter, during the third quarter, we are very excited with the acceleration of transactions and of monthly active payers and monthly active sellers we have seen. We are very excited by that acceleration. Most of that acceleration has been driven by new users and new sellers joining the network rather than by an increase in frequency.
If it evolves as it did in Argentina, the first things that need to happen is for this capillarity to increase, for there to be more payment options available, and then we should see an increase in frequency.
Got it. Thank you, guys.
Thank you. Our next question comes from John Coffey with Susquehanna. Your line is now open.
Hi. Thank you for getting me in. As I remember from previous calls, and maybe I'm forgetting something, that your cross-border transactions were fairly minimal. I thought I recalled there were some maybe transactions between countries like Brazil and maybe China. I know you've called it out a little bit more in this call. I was wondering if you could just walk me through a few basic use cases of some cross-border transactions you would have today or are planning in the near future. The last question, just also a pretty short one. As far as your marketing spend in Q3, should we think of this as a little bit more bursty, or could this be a new trajectory that we might see going ahead in the next few quarters? Thanks.
Just one clarification. When we refer to cross-border here, the focus is really on cross-border commerce. It's more on our retail business. What we're trying to do there is to offer a cross-border solution that, unlike the more prevalent ones, really focus on companies that have global merchant bases and then push product globally, and how we can leverage our existing marketplace and the benefits of that marketplace locally to bring in inventory, not so much intra-Latin America, but primarily from Asian merchants and North American merchants, but deeply integrated into some of the assets that we've built on the marketplace.
When I mentioned earlier, I think Bob's question around tying cross-border trade to the user experience, it's how can we leverage the assets we've built in logistics to get those cross-border items to your doorstep a lot faster than we would if you were buying on some of the other global platforms. How can we use local teams we have to really facilitate and expedite tariff and customs processes so it's less of a hassle for people purchasing on the platform? The idea then is to get global inventory to the doorsteps of Latin American consumers. It's not so much focused on payments. Look, marketing, again, we don't guide. I think we've given you guys a thought process on why we think the timing is right to pick up the pace of marketing investments.
Very obvious on the fintech space, we're attacking many different fronts simultaneously, QR network, mPOS, cards, merchant services. We see tremendous opportunity here, and we want to make sure that we invest behind it, and we need to build out the brand. On the marketplace, because of all the improvements we've made, we think it makes sense to strengthen some of the brand attributes, communicate some of the new brand attributes. Going forward, I think a lot will depend on the performance and what we see in the data around these investments. Brand investments are not performance-like that you can see immediate impact. We will be tracking key numerical KPIs through surveys, through direct-to-site organic traffic and other ways. We need to give it a little bit more time to see what the residual impact of those brand investments are.
I think based on that, we'll have a clearer sense of how we continue to invest going forward.
Thanks, Pedro.
I think we'll keep you guys posted every quarter on what we did during the quarter rather than forward-looking.
Thank you. Our next question comes from Richard Cathcart with Bradesco. Your line is now open.
Hi, guys. Good evening. I wanted to ask about the managed network in Brazil, and specifically kind of what behavior you're seeing from consumers that are receiving their products via the managed network. Clearly, I think they're getting a better service than via Correios. I think what I'm trying to get at is if you're seeing any kind of improvement in frequency or conversion as consumers increasingly receive products from the managed network. Thanks.
Yeah. Look, we're beginning to see some of the numerical flow-through of the benefits of the managed network in Brazil. Lead times are becoming faster than the rest of the other alternatives. Transportation costs are coming down. It's becoming cheaper to deliver packages across the network, although it's still far from at full capacity. Perhaps most importantly to your question, net promoter scores are now higher on the managed network than away from it, at this point, roughly six points higher and tracking to widening that spread. Conversions are a little bit more difficult to measure because there's a lot of other stuff that impacts conversion. The tests haven't been as conclusive there. Again, I think we're increasingly convinced on both the immediate and also long-term benefits of the managed network versus the old drop ship network.
Thanks very much. If I may, just a quick follow-up for Osvaldo, I think. I wanted to ask about the promotions, the discounts that you've been offering to consumers to use the Mercado Pago wallet. Can you just give us a little bit of color about kind of what you're learning from those discounts and promotions and kind of how that is impacting users' behavior. Thank you.
Sure, Richard. Let me tell you about our experience in Argentina, which probably is what we're using as a base for what we're doing now in Brazil and Mexico. Every time you went to a given store, you would get a promotion. We switch off to every time to these stores or chains, which we call lighthouses, which are the larger merchants we work with. We were giving you a discount the first time you paid in each of those. We are at that stage now in Brazil. Eventually in Argentina, we were able to diminish the amount of promotion we did. The first month we did this, we stopped doing promotions. There was a little bit of a slowdown, but a month after that, we were growing at the same rate as we were before.
We're at very similar rate. We are following that playbook in Brazil. Early on, we just want to add as many sellers and as many buyers as possible, payers as possible, and eventually we will focus on improving and making more efficient the discounts.
Okay. Thanks very much, guys.
Thank you. Our final question comes from Rodrigo Nistor with Itaú. Your line is now open.
Hi. Thank you for taking my question. Regarding logistics, specifically in Brazil, if you could give us more color on the steps you're taking to increase the fulfillment penetration and which are the main obstacles you're facing there. Thank you.
Okay, great. I think like we said, the level of fulfillment penetration really began to pick up again from a low base, but nearly doubled sequentially this quarter. I think it's been a combination of solving some of the friction around sending inventory to us, building better tools for sellers, building more efficient pickup routes, more frequent pickup routes, and then combined with economic incentives that we did offer. I think we were kind of stuck for two quarters there without too many improvements in adoption. We begin to see that pick up again. I think our level of confidence right now that we will continue to scale out our fulfillment network, and also our cross-docking network, is pretty good.
Thank you.
Thank you. This concludes today's question and answer session. Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.