Good morning, everyone. Welcome to the Alpargatas Earnings Release video conference for the third quarter of 2023. Today, we have with us Luiz Edmond, CEO, and André Natal, CFO of the company. This video conference is being recorded and translated simultaneously into English. The Q&A session will be conducted according to the following rules: After the end of the initial presentation, we invite those who wish to ask questions to raise their hand using Zoom's toolbar feature and wait with their microphones muted until we give them the go-ahead. Before continuing, I'd like to clarify that any forward-looking statements that may be made during this video conference regarding the company, the company's business prospects, projections, and operational and financial goals, constitute the management's beliefs and assumptions based on information that's currently available. Now, I give the floor to Luiz, our CEO, who will start the video conference.
Luiz, you can go on, please.
Good morning. Good morning, everyone, and thank you for being here for our earnings release. I will start with a short summary of what we have been experiencing throughout the year, and André, after me, will go deeper into the details about the results of the quarter. As you all know, we took on the company in the second quarter of 2023, and we are going through a delicate moment. We have had several consecutive months of bad results and the consumption of our cash. This scenario demands actions which were immediate, some of them quite hard, but able to avoid and prevent a deterioration, a continued deterioration of the company's nature.
The first focus was to preserve cash, so we decided to reduce inventory, compatibilize the production and demand at the moment, and significant reduction of purchase of raw material. We also needed to strongly reduce investment levels, eliminating all initiatives that were non-essential. Another priority was to have the kickoff in the reallocation of the structure of the company and the level of expenses of the company, with some simplification measures to ensure, as much as possible, the preservation of the company's bottom- line, especially in the moment when the top line numbers, especially having the allocation of the numbers in the inventory, is different from what we would like to be. In this third quarter, we have seen some indications that the resumption of sales volume is normalizing.
Sell-out continues to continue to evolve sequentially, going back to the same level as the last year in the same period. Additionally, we started to see some positive trends in our expenses, with reflection as a reflection of a part of the actions which were taken in the second quarter. We are also seeing better sequences of cost of production sold, cost of goods sold, with a higher simplification of the portfolio and better productivity levels in our factory. Overall, we saw a significant improvement in our manufacturing management, whether considering the quality of our products, the losses in production processes and operational levels, which we believe are, is going to continue. We know that in spite of some operational improvements, we still have many challenges to be taken and faced and won. We are away from what we aim.
Our service level still does not reflect what we can really deliver in our international operations, mainly in Europe. They were severely impacted by a series of problems that are still being resolved and will take some time to be. Thus, to ensure that we're going to have the tools and the capabilities needed to reach the results we aim, the creation of a roadmap for the future is necessary. After an important initial diagnosis, we are right now working on a strategic review involving a good part of the company's leadership, and which should direct and guide our future growth, sustainable growth, and with an improvement to profitability. We have in our favor, the brand Havaianas, which remains extremely strong, even facing all the adversities and challenges we have been facing over the past quarters.
We have launched a new campaign that breaks the cycle of three years that we hadn't been exploring the seasonality in Brazil in this time of the year, and takes us back to TV after one year, focusing mainly on digital marketing. Additionally, we represented Brazil once again as the official shoes for the Brazilian Pan- American Olympic team. This action and others have been generating great engagement, but in part of our consumers, showing that our attributes and brands are recognized and in desire. In summary, even if our results are far from what we already delivered in the past, we see an evolution in all the fronts that we started working on. There is a lot of work ahead of us. We keep finding out about new problems and new opportunities to be tapped, but we also have the condition to be going towards the right direction.
I take the opportunity to show my total confidence in the Alpargatas team, motivated by the passion and dedication to our brands.... Our team understands and has been working to deliver the short-term results and simultaneously build on our future. I will now give the floor to André, who is going to explain the dynamics behind our movement so far. Thank you all very much.
Well, good morning, everyone. It's a pleasure to be here with you, sharing part of our results. One thing that we consider very important should be done right now is not only to discuss the picture of the third quarter, but also try to insert the third quarter within a movie that is more comprehensive and where we can understand the movements. That's extremely important to understand the path towards the recovery that Luiz mentioned.
And as he mentioned, it's long and there is a lot of work ahead of us, but on the other hand, we can, if we insert the third quarter in this movie, we can also see a good part of the evolutions that have been made so far, which are actually several, and which go in the direction which we would like to. Even though the company is still not delivering the results we would like it, and we understand it should deliver, considered its potential. Having said that, on this first slide, I would like to recover with you some of the topics that we have spoken about and discussed in the earnings release for the second quarter, when we had a very strong compression, scale compression, where we had almost zero EBITDA results.
Especially because of the level of inefficiency, because of the expenses of the company, but mainly because of the top line that generated all of that dynamic of losing scale in our operations. In that moment, we shared with you what we believe it to be from that moment on, part of the dynamics of reducing our inventory in the chain, and how that would reflect in terms of reducing our own internal inventories and resumption of the scale, and so on. So on this slide, we can see a little bit of that sequence that was the narrative we brought to you in the second quarter of the year, and I'd like to recover it a little bit by reminding you that at that moment, we talked about the need and intention of going back to have a recovery in our sell-out.
So it was the beginning of a sell-out recovery. Up to that point, we had had compressed the sell-out of about 3%, but we believed and everything indicated that we had a gradual improvement of the sell-out situation. That this recovery would be one of the drivers to reduce the inventory levels in the chain, which was already decreasing at the time, but it was still in the middle of this adjustment cycle and had not yet been completely concluded. And as this normalization happened, this would resume the closeness between-- a match between our sell-in and sell-out, so this would allow us to drain our internal finished inventory of products. And this happened-- started to happen in the first, in the third quarter.
These checks you see in these boxes indicated that we went towards the direction. When you look at the inventory of raw materials, at that time, we had already been able to reduce significantly the inventory levels of raw materials, but now we see that this decrease continued to happen along the third quarter, and this happens as to free up working capital in the company. In parallel to all of that, there were measures that we shared with you at the time that would be we would be going to be implemented, that was more visible and more immediate, to be able to capture a little bit of the compression of expenses that we understand that is necessary and possible, and yet implement all the measures for cash burning. That was happening, and I will make more comments later on.
On this slide, we show, showed it to you in the last quarter, and some investors and analysts required us to remain, to keep this slide in a way to keep track of this evolution over time. So this slide summarizes a little bit our business cycle from the purchase of raw material into the sale of our products, and we can see this evolution year to date. Today, up to September, we see the sell-out going down 2%, which means that since we had -3 up to June, the sell-out over the third quarter was flat, actually, compared to the third quarter of last year. Which is important information because we go from a level of -3 to a flat level of a sell-out.
Obviously, inside this quarter, this trend was also upwards in showing an improvement movement. This allowed us to, in the graph, green graph below, to see a positive inventory. You see that the cycle that was midway, now we could say that overall, it has ended. So this is a number, it's approximation. We don't have a complete visibility of the inventory of all of our clients in the supply chain in the entire in all retail. But overall, we can have a reasonable forecast and prediction, and it seems to have ended. When you look at the sell-in, in percentage terms, it gives an idea that there is a mismatch, considering the -2% in the sell-out.
But you're going to see in another graph that the right comparison is, should be in the number of pairs, because in the third quarter of last year was when the entire sell-in was in excess to the sell-out, and the biggest part of the sell-out was created. So what we said in last quarter, that the sell-in and sell-out should be in a match, is seen when we see the growth of our sale, which was led by the growth of sales to the final consumers. Our inventory of finished products, which had been balanced up to June, over the third quarter, started to decrease, which we understand to be important for the release of our working capital, and to resume some leverages inside our management and inside our inventory and inside our commercial policies.
The production cycle are still below last year's, and this is just because we wanted to keep control of our inventory levels. The cycle of reduction in raw material had been happening already, and we were—we managed to accelerate that as we saw an increased acceleration in the raw material purchase, which you can see on the left here in -52%. So we purchased much less, so we could effectively drain this inventory that was in excess last year. It was in too much excess last year due to all the volatility of the commodity. So this is a summary of all the dynamics of the volumes and release of inventory, and this dynamic was very important.
It is behind, maybe underlying the numbers, but here we try to make it very clear to understand what evolution the company is having in this sense. This graph shows and reflects what I have just said about the discompass between sell-in and sell-out. Just to remind you that the sell-out last year, the yellow bar, was much bigger than the blue bar. In other words, we sold much more to clients than it was sold to consumers, so it accumulated an inventory in the retailers and the clients that was too large. This inventory, throughout the first and second quarter, was drained, and this is shown. You can see numbers of pairs on this slide. There is a mismatch between the first quarter and second quarter, between the yellow and blue bars.
They are quite far apart from each other, and this was what generated the pressure, the scale pressure in our operations. But it was important to try to adjust the business cycle along our chain, production chain and sales chain. As you can see in the third quarter, it's very easy to see visually that the yellow bar is very close now to the blue bar, which means that at this moment, we seem to have ended the adjustment cycle. There were still some adjustments in the first months of the third quarter, but we leave the third quarter with a inventory level much more in, like, leveled up with the, what is expected for this moment. On this slide now, we also show you on another front, the measures we have taken and we have been implementing, considering efficiency gain and simplification of processes inside the company.
We had told you that we were starting, at that time, a process, a structured process, to detail the understanding of our opportunities for cutting costs and expenses, and this is still ongoing. It, it's not yet over. We're still ongoing with this project, which will go into the end of the year. But at that moment, we also said that we wouldn't wait for the end of this process to start to implement part of the actions that we understood to be possible to be implemented at the moment. So we have started them, and this is a breakdown. We are talking about all of the packages of our zero OBZ packages.
So, for example, here, you don't have the write-offs or severance, but you have—we have the real expenses, so we can have a real tracking of what to have a reasonable metrics of the efficiency of the company. So when we look at that, is anchored on the first quarter to have a effect of comparison. We can see clearly the first and second quarter in a leveled, but already a considerable decrease effect in the third quarter, especially because of the small time that there was between the quarters. We see a important compression of expenses that is being, and will be, very important as a way to attenuate the effects of the top line in our results. Which should help us, over time, to resume the profitability of the company.
On this next slide, I'd like to show in a more visual way what I mentioned before. The company had been, up to April this year, in a route of burning cash. So the cash burning added up to BRL 1.2 billion in these 13 consecutive months without exception. And of course, this generated a pressure in the leverage of the company up to that point. But luckily, and happily, due to all the measures to hold the investments, CapEx and working capital, we have been able to revert this, reverse this, trend. So we are going to the sixth consecutive month of net positive cash generation. We understood that was a priority before anything else, to preserve liquidity and to preserve this financial structure of the company.
So here we have concrete evidence that this is happening right now. I'm going to give the floor now to Rafael Estides, who is going to give you some details about each one in the breakdowns of the business areas and the financials for this period.
Thank you very much, André. I'm going to talk a little bit about the results concerning the results of the quarter, starting with Havaianas Brazil, BU. We saw a decrease of 20% in the volume year-over-year, and 15% decrease of net revenue. Still affected by the increase of volumes that happened last year. As André mentioned before, the sellout was flat in the third quarter. Therefore, there is an important sequential evolution month-over-month in the quarter.
Even with this decrease in volume of 20% in Brazil, we managed still to have a good reduction in the inventory levels, accelerating the inventory decrease of finished goods in this quarter. Moving on to Havaianas International, BU. We saw a volume decrease of 40% year-over-year against the third quarter, with a net revenue decrease of -24%. In Europe, as we had mentioned before, with the exchange of our logistics partner, all the consequences that arose from that in the beginning of the season, we ended up having not many products available for sale for consumers, and which generated cancellation of orders, and this impacted the sell-out throughout the period in the region.
Obviously, the entire replacement of inventory that happens in the middle of the season, at the end of the season, and it's recognized in the third quarter. So the third quarter was still very impacted by all the consequences that we had in our logistics issues in our service levels. In the U.S., we started to see some evolution, both in volume and revenue, especially due to our B2B and off-price accounts and key accounts. It's important to say that we saw a growth in the revenue per pair that was reasonable.
In the markets where we operate through distributors, especially in Southeastern Asia and Latin America, we still see a very strong decrease in volumes, especially in Southeastern Asia, due to all the inventory that was created at the end of the second half of second year, last year. We are still ongoing this normalization of inventory levels. As you saw, this normalization of inventory happens in Brazil a few months. In our distribution distributors markets, it's the same, we are going to have this year as a year for normalizing the inventory level, so next year we can start with a more normalized level. Going to the P&L, talking about our gross profit, there was a decrease of 26% in Brazil and 35% international.
The gross margin went down 5.5% compared to the third quarter last year. But there was an important evolution when we look at the sequence compared to the second quarter of this year. It was a 9.7% increase compared to the second quarter 2023, as a result of the improvements that we have been made to the manufacturing processes, and also to as a consequence to the reduction of, of, the cost of our raw materials. It's important to emphasize that we had in this quarter some write-offs, some additional write-offs above the, the usual level, which, added up to about BRL 9.4 million, especially concerning finished products and concentrated in the categories that we call, we used to call, we call Beyond the Core.
It's important to emphasize that we are going to say these write-offs did not, did not happen only in the third quarter, and they were not going to also pop up in the next quarters in any reasonable amount, especially related to these categories of sandals and slides. As for the external warehouse that we have been hiring, and that had a strong impact in our, in our previous quarters, due to our high level of inventory of finished products, we can see an improvement quarter-over-quarter in this external hiring of warehouses. We had a strong reduction. It was at about a level of only BRL 1 million expenses for that purpose. We are still not normalized level, but we felt a smaller impact of this hiring of external warehouses.
In the international of Havaianas International, we saw a 52.3% gross margin impacted by the deleveraging, operational deleveraging, and also for the high production costs that we had in the second half of last year. So it's important to remember that a good part of what we sold in the third quarter is still an inventory of the products that were manufactured in that period when our manufacturing costs were higher. So we can see this still is impacting the international operations. Going to expenses, we saw a decrease in 6% year-over-year, considering the third quarter in our SG&A, overall SG&A.
It's important to emphasize that our SG&A, excluded the marketing, went down specifically because of the reduction in payroll of -8% year-over-year, but also with the expenses in distribution of BRL 4 million and additional cost of BRL 3 million with PDD. We also saw a reduction in marketing of 24% year-over-year. It's important to emphasize that this reduction is much more connected to the quarterly phasing of our marketing budget that was different this year compared to the last year. And also with a very strong reduction of what we call non-working marketing, which are expenses that are not effectively reaching the final consumers.
As Luiz mentioned in his opening remarks, we have a very strong performance in campaign, marketing campaign, this third quarter to stimulate the fourth quarter results in Brazil. Going now to our EBITDA. We saw an EBITDA of BRL 77 million normalized EBITDA for Havaianas. An important decrease considering the third quarter of 2022, which was very influenced by the decrease in volume that we saw, but also by the increase in the cost per pairs. And as we said in our SG&A, excluded the marketing and other expenses. I'm sorry, the graph does not show all the information that I need, but the PowerPoint presentation that you are going to find in our website, you can see all the numbers properly. Moving on to the next slide. Here, also about the EBITDA.
We start with the Havaianas EBITDA, BRL 7 million, and we reach our corporate EBITDA of BRL 77 million normalized, BRL 7 million, the EBITDA for Ioasys. And we had extraordinary items of BRL 18 million. Extraordinary items are specifically effects of our simplification, so there are many costs related to payroll expenses reduction, and some other small expenses that add up to BRL 2.8 million, reaching our corporate EBITDA of BRL 59 million. As for our net financial position, we closed the quarter at a net financial position of BRL 109 million, better than the last quarter, with cash generation concentrated in the operational flow, as André and Luiz mentioned previously, especially concerning our decreasing working capital, which was due to our inventory level decrease. And also with a very low CapEx level that we had seen in previous quarters.
So this way, we managed to generate BRL 109 million in operational operating income. Going to our working capital a little bit more, we saw a decrease of 13% of our inventory, of our overall inventory level, considering good decrease in our finished products of BRL 99 million, and also BRL 75 million in reduction of our raw material. It's important that we just say that we see, we see the medium-term raw material prices normalizing, going the direction, in a clear direction of improvement. As for our accounts receivable, we saw a reduction of 12 days reduction in the year-over-year net receivables.
We have been focusing at the top line in the cash preservation, so we are not really making any contingencies of deadlines, so we are being actually very conservative, even at moments of harder top-line results. And in this account of suppliers, we had a reduction due to the reduction of raw material, and also to the reduction of the cost of raw materials. We saw a very relevant decrease and reduction in the third quarter, and as our productions resume growth, so we can meet future demand, once we have the normalized inventory levels, we should see the supplier account to grow in the future. Moving on to the next slide. Talking about Rothy's. Rothy's saw a quarter with a net revenue of $30 million, -1% year-over-year, compared the third quarter.... explained by the e-commerce.
The e-commerce operation resulted in this decrease. The direct e-commerce decreased year-over-year in an environment of a hard top line result. On the other hand, a very good and important evolution of gross margin. We saw an evolution of 7.3% in gross margin year-over-year, and the gross margin of the quarter is already above 60%. So that's an important level for evolution of our—of the Rothy's gross margin. Explained by the reduction of the freight costs from China to the U.S. Also improvements to manufacturing costs in China, and also reduction of costs inside the logistics operation within U.S. There is also an effect of the mix of products, which also accounts for this overall improvement of gross margin. Rothy's closed the quarter with a minus $6 million in EBITDA.
And we also had told you that the seasonality is lower in the third quarter. In the U.S., we probably would expect it to have a negative EBITDA, but it was an EBITDA better than the EBITDA found reached in the last year. And we also had a loss net of $7 million.
Now we are moving to the Q&A. I'm going to invite people in the order that they raise their hands to ask their questions. And we will open your microphone, so you can ask your questions. So the first question from Guilherme Vilela with JP Morgan.
Rafael, hello, how are you? Good morning, everyone. I would like to talk a little bit about what you mentioned about the sell-in expenses, the marketing expenses, excuse me.
How do you see this in, you know, from now on? As we saw in the release, it was the account that had the largest increase, you know, compared to the other accounts considering the expenses. Please correct me if I'm wrong, and I would also like to hear from you, seasonally, the importance of the fourth quarter. So what do you have any visibility for the sellout that is going to be expected for the fourth quarter? Thank you for your time.
First of all, thank you for your question, and good morning.
I think it's important to mention about the marketing expenses, is that the entire diagnosis that we have run so far, looking the long-term history and all of the marketing history records we had and how this evolved over time, considering the evolution of our revenue. All of the diagnosis points to the direction that we are actually should have more investment than lower investments in marketing. And especially, we should be working in how we use this marketing budget, what percentage and to what we are dedicating this marketing budget. I'd like to mention that the direction we see when you're looking at actual numbers along the history of the company, this number doesn't decrease over year- over- year. Usually, there was a small, slight increase year over year, but certainly it has been compressed over years because of the revenue.
We don't agree with this path. We understand that this path would be much more to support everything that we are doing. So we should be should have more relevant investments in marketing, and this is basically what we see from now on. Everything that we're talking about, expenses reduction, including the expenses that we have been able to compress, they don't include the marketing account. We are not having any kind of... The compressions that we understand that are due in terms of efficiency, are much more compared in the other, all the other accounts in the SG&A, especially to prioritize the marketing from now on.
When we talk about the future, just to wrap up this part, the marketing budget for the future should be to remain stable or to grow over the next years in the future. This should be the direction we are going to see. But we don't have a number or a guidance of how this is going to happen or what the marketing budget will be for the next year. We're still. These numbers are still being discussed. Over, as for the fourth quarter, we see that the dynamics of the reduction of inventory levels in the chain has been seen, has been happened, and we don't have any large difference that will impact our sell-in.
The path that we see for the future, and in the third quarter, it was a sell-out path that was an improvement year-over-year. So September, better than August, August better than July. So this is a trend that is somehow positive and inspiring. On the other hand, we don't have a visibility of what is to come in the future. So we trust the potential of the campaigns that we mentioned that have been launched recently, and we trust the brand and our collections and all the commercial actions we are carrying out to incentivize a lot. However, on the other hand, we still don't have even the numbers of October to have a first impression of how this trend has been proved to be kept or not, considering the month of September.
Of course, we're going to have the seasonality effect at play, at play, and it's important to remember that we have our operational challenges. So we continue with our rewire project and the coordination of our planning team and the management of our inventory, our delivery cycle, et cetera, to meet the demand that will appear again, and we'll have to face that. We see a positive trend year-over-year in this category. So when we have this flat sell-out right now compared to the third quarter last year, this is because of a category that is growing. So when we look, we lost some share compared to the end of last year, but now we understand that we should grow together with the category, and the category shows trends of continuous improvement.
So I would say that we have a positive trend for the category and for our own growth. Once since we shared we lost share, we should go back, we should resume our growth together with the growth of the category. But how much it is going to be materialized in the fourth quarter is hard to forecast right now. We believe that the main pain that prevented us from resuming growth was the mismatch between sell-in and sellout. As the sell-in is more, you know, on level with the sellout, we are going to be able to uncompress our scale significantly over time. I don't know if you have any comments you would like to mention.
Complementing what André said, we are talking about the dynamics of the Brazil operations. The answer was based on Brazil's operations.
Going back to the marketing discussion, marketing has two, like, big components: the working money and non-working money. The working money is the one you spend to create content with artists that participate, photographers, influencers, and so on. It's the money that the consumer doesn't actually see that. And then you have the other piece of that, which is the one that you spend on purchasing space in the media. So nowadays, in Brazil, we have the perception that we are investing less than we should be, given the benchmarks that we already know. So we see an opportunity to be tapped, since we have the right campaigns, campaigns that are proven to have an effect on the perception of the brand and on its impact—and that has an impact on volume. These campaigns will continue.
I think the first big movement this year was to improve the financial results. We didn't invest over the last years in marketing. This year, we decided, in spite of the pressure that we have in the results, which are much below the levels of what we expect, we decided to keep the investments in marketing as it had been planned originally. Outside Brazil, I think there are three dynamics that are quite different. In the U.S., the U.S. has been delivering everything that it was bound to deliver throughout the years. They made all the corrections they needed at the beginning of the year. It went back to grow in the e-commerce, understanding better the leverages to better perform in the e-commerce, both through the Havaianas.com, as well as through our e-commerce partners.
In Europe, we still believe that next year is going to be a hard year. Europe works with longer cycles, and right now, we are taking orders for next year, which really affected us, is since our sales is seasonal in the summer, and since our clients are very large and organized, if we did not serve them inside the service levels we, we offered, we are perceived as, as an operation that cannot be trusted, so we're, they did not dedicate the room and the space to our brand that we had before. So our priority right now is to regain the confidence from our clients through improvement service level. So I don't see big bets or big transformations Europe for 2024. It's quite the opposite.
We want to operate well, execute well within good service levels to establish a good basis for growth, and we have confidence that Europe is a little bit shows more readiness right now, considering 2024. And you have all of the distribution markets, Asia, Africa, Latin America, they are going through a profound revision of which is the role of each country in terms of growth, profitability, in terms of what do they present, in terms of capabilities to execute from our distributors. So we are going to go in, into something more continuous with different levels of expectation, so we can create the Europes of and, and USAs of the future in some of those bets, in some of those countries. So there is a lot of work ahead of us to be done.
We are learning, we are finding out a lot, we are discovering a lot. There is a lot of operational inconsistency, so we have made a decision to, in the second half, not to start sell-in crazily to the market without having this diagnosis very well defined and studied. We are seeing a little bit of the deleveraging of this inventory. They have very large inventory levels for some years, and I think we're going to be able to give you some more information over the next months about how the strategy for these markets is going to evolve. I don't expect big improvements to these distribution markets in the upcoming months.
Thank you very much for your question, Guilherme. Next question from Felipe Cassimiro with Bradesco BBI. Felipe, go ahead.
Good morning, everyone. Good morning, Luiz, André, and Rafael. It called my attention, André, your comment on growth with the brand. This called my attention because now, excuse me, grow with the market. Nowadays, the volume is at 187 million pairs, and this is a relevant gap considering the records. You know, removing 2021, we'd had a 2020, which had a relevant 231 million pairs sold. So just trying to understand, what is your idea of a normalized annual amount of pairs to be sold? Because we discuss a lot with everyone, how quickly you can go to this level above 210, 200, 200, 210 million pairs sold annually. Can you give us some light, some understanding on this topic?
My second question is, the simplification sounds one of the most present words in the head of the management, in the mindset of the management. So simplification of portfolio, where is the—where are you with concerning SKU simplification, and where are you at the simplification of payroll and structures? You know, are there more things to come in the future? Is this all finished? How do you see that?
Felipe, thank you for your question. I will start from your last question, and Luiz will help me to take on part of your questions.
So about the structures, as I mentioned before, we have been doing a work, an important work, with the support of a consulting company that had already been started in the company of understanding the structures that we have to reach an optimal structure, and now we are accelerating these works, concentrating with benchmarking and external support from this consulting company. This takes some months to be carried out, so this should be ended by the end of the year, with some actions to be mapped in this planning and which should be implemented next year. On the other hand, as I said in the previous quarter, and I repeat it again in this quarter, we understood we shouldn't be waiting until the end of this diagnosis, so we could find the low-hanging fruits and better opportunities to make the first adjustment.
So this is about the organizational structures, but this is also about other discretionary expenses. So all the factors have some opportunities, and some of them are obvious, and these are the ones that we attack at first, and some of them require previous work before we could reach a higher level of simplification, centralization over time. So I would say that, yes, we have already started part of this adjustment that you mentioned and that you can see in the third quarter's figures. This arises from simplification of structures, in opportunities that we mapped, but this is not over. We don't have a silver bullet for that. We have several actions in different fronts that are already ongoing, and certainly, we will go deeper into this work. We are studying in more detail right now, the responsibilities.
It's my responsibility and the responsibility of controlling area, and that's a super important and sensible topic to ensure our effectiveness, agility, and so on. And we are going deeper into the other fronts that we call much matrix expenses that are going to deal with all the other expenses, you know, ranging from printing paper to engineering works or any other types of expenses, you know, consulting and IT and so on. All of this is being discussed right now in this work that is ongoing, and we don't have a guidance right now on...
We do have a reference for a first, preliminary assessment, without actually having a goal, a clear goal of where we want to get at, but this was mentioned to give and share some of the potential idea that of the things we identified initially in a superficial assessment. So for the time being, nothing has changed, and we don't have any guidance, and we will not offer any guidance on that, but we will go deeper in our studies to get there. Luiz, would you like to comment?
Well, as for the portfolio simplification, the cycles of innovation, sales, delivery are long.
So we are talking about a cycle that an innovation cycle that starts two years before the product actually reaches the market. So these are trends, fashion trends that we have to follow, or product development in the sell-in that varies a lot from country to country. Brazil has shorter, slightly shorter cycles between the innovation and the product to be found in the market. So the USA is also different due to seasonality, obviously. So these portfolio measures for structuring portfolio take some time before you see the practical consequences in the results. On the other hand, our studies show that we had a quantity of products or items or SKUs with minimal impact level. So this is the old story of looking at detail.
When we compare the impact of volume versus innovation, without much detail at the very beginning, we think that now we are ready to introduce to you in the future a theoretical structure of a better work portfolio framework that is educating the portfolio that's going to be released for the 2024, 2025 collection. It presents an important reduction in the complexity that we used to have. We are always careful to look at the market trends first, to understand where we have market share opportunities, especially in Brazil, where we have opportunities to grow the brand. In these places, the portfolio and has a very important role to develop the brand in these countries where we are relatively less relevant than compared to Brazil.
At the same time, this structure allows us to balance whatever is recurring and shouldn't be changed year over year, and which provides an important base for our business, which needs to be new, you know, more renovation than innovation. We have an excess of innovation in the portfolio that didn't actually generate incremental results, with much more substitution than incrementality. So we are compressing a little bit the intensity and frequency of those renovations, and we are clearly developing the new batch over time, where we are not going to have big bangs, you know, all at once of a new product in the market. First, we're going to learn in pilot markets, so in the next year, we can scale it up and follow what works. So in addition, like considering what works, we're going to add new variations.
If you remember, for example, the glitter product was launched for the Carnaval. We documented that, and we created a portfolio and created a pattern of how this should be developed. And at the same time, we are adequately this per each market. It doesn't make sense to make the launch of a product that has high cost globally. There are markets where are much more adequate to receive products of higher price. I don't want to mention any specific countries, but clearly, the organization of portfolio versus the organization of the roles of the countries, of the stage we have in each one of the countries which we're calling brand growth model, is going to be very important from now on. So the portfolio is being simplified.
It will be much more simplified for the upcoming year, but the simplification maybe is the year that everyone is bearing in mind. But we are actually being more accurate considering the opportunities we have. So we're being more accurate and less erratic, you know, less trial and error, you know, creating complexities and complexities and investment that are not justifiable in our results. As for your question concerning the volume, we don't have any volume guidance. Clearly, in Brazil, we are always going to depend on macroeconomic factors. This is a product that is hyper-distributed and penetrates all the social levels, so it depends on the population growth, depends on inflation, depends on families' indebtedness, and all of this has an impact on the consumption of our product. So but the numbers, you saw the numbers of sell-out.
The sell-out is reasonable, reasonably stable. We started much, actually, much worse at the beginning of the year because of our share position or because of the market performance. But the 200 million pairs sold is what should be, in the same basis, the starting point for us for every year. The difference is that we clearly had an adjustment, a very large adjustment in inventory. So if you normalize that over the past two or three years and understand that the market didn't grow that much, you are going to reach the conclusion that you are going to have a match between sell-in and sell out. And what we expect as in practice is that there is a convergence between the sell-in and sell out over the next quarters.
We have made many of the adjustments that were due. It doesn't mean that all of them have been completed, especially because of the very high cost of capital right now, not only for us, but also for our clients. I don't want—we don't want to build up high inventory levels right now. Our manufacturing efficiency has been improving, allowing us to make a faster catch up with the demands that arise. I would expect the market to behave within the 200, 200-plus million pairs. Always having in mind the effect of the real price over the past years, due to because of the price, the cost pressures, there were the adjustments were much above the inflation.
I believe this—I hope this is not going to be necessary again. So we believe that our price is going to be in line with inflation, maybe small impacts because of the evolution of our mix of products. So we are going to have much less pressure on the capability of our consumers to purchase our products. So without giving you any guidance in the short term, I would say that in the short term, we are going to see a convergence of a sell-in and sell-out. And naturally, the sell-out is not being impacted by—won't be impacted by these huge inventory variations that we have.
Thank you, Felipe, for your question. Next question is by Danny with XP. Danny, go ahead.
Good morning, everyone. Thank you for taking my question. I have two questions.
First, considering the operational adjustments, I'd like to explore some of the points. The first one is about the pricing or price positioning. How you're thinking about the pricing price positioning, especially for 2024 and onwards, considering that this has been a very important lever to compensate the volume decrease. We probably will see a normalization of this volume, right? But I want to know about this balance between volume and price, and the footprint in the international operations, if you're looking at a possible closing of operations in any markets in the future. And my second question, Luiz mentioned about the service levels that have somehow affected negatively the image of the company and the trust in the company's performance.
You talked a lot about on time in full improvements, but recent feedback is that these service levels have actually decreased, maybe due to the adjustments you have been made. I'd like to understand how this OTIF will happen, and how this will be improved or not, or will worsen in the upcoming future.
Well, let me first tackle the pricing about the international markets. We won't give you any guidance on pricing, but as a policy overall, the pricing should be in line with inflation and benefited from the mix portfolio. The improvement of our mix in the revenue doesn't mean better profitability. This is one of our analysis. We found out that our products, even that have a higher price, not necessarily have translated into better margins for the company.
So this is something we are being careful about, and we are following a little bit more to understand the effect on the margin, and not necessarily on the top line, on the revenue. But having said that, as a philosophy and policy, we want that the affordability is maintained, so the prices should be in line with the inflation. One of the things we know we have to do in the long term is to make our portfolio more, more premium. However, at the same time, we have left some reasonable gaps in the entry points in the first two price segments, in which our competitors have positioned. So there are some initiatives that are ongoing right now to develop products and solutions to tap these two slots, where the affordability is very important and attended.
Especially because these are two slots that are, you know, where products are purchased by people who are less privileged in a financial point of view. So in an intelligent way, we have been finding solutions for some segments in some markets. As an overall policy, we would like to have the pricing ongoing, you know, hand in hand with inflation. As for our internationalization, it goes through the success in Europe. And in Europe, it doesn't mean Europe. There are four countries that actually make a difference in Europe for us, where the brand is really relevant and where there is a real potential, and this translates into the positioning of Havaianas and the way that Havaianas is worn.
We had a usage and attitudes large study, where we understood not only the size of the population, but also how this population interacts with the several types of products in the open shoes market and similars, and similar products. Nowadays, we are sure that those four countries in Europe have a very large fit with Havaianas, its brands positioning, and with our products themselves. After that, there is the U.S. There are no companies in the world that can be called global or international if they are not relevant in the United States.
We are satisfied with the evolution that we have had so far this year, but we also know that we need to invest in the U.S., we need to learn in the U.S., and we need to be very consistent to keep evolving in the US and continue maintaining our relevance in that market. These are large cycles that at the end of the day, they matter, and they have an effect in the short-term relevance. The other 120 countries where we are present, as I said, we are creating clusters, and we know the work is not over yet. We know that some countries are easier, some countries are more profitable, some are less profitable.
But basically, we're doing what we said we would do back then, and maybe we have not yet completed the work that we said that we're going to go from the market in. The company is an exporting company in a way that we push the products into this company, and we want to have the commitment to develop the market, not only find the distributors to push the product into the markets and, you know, God help us. So for this purpose, we're going to concentrate our bets in less countries where there is a brand and product fit. And this work has not yet been done. As soon as it is completed, we are going to give you more transparency so you can understand better our findings and decisions.
André, I think there was one more question, wasn't it?
Hello, Danny, how are you? There was one more question which concerned the deliveries, the OTIF in Brazil, the service levels in Brazil. I understand you have your own sources, but I think there are two sides to your to answer your question. First of all, we can say that we have not reached the operation and consistency of deliveries, and we are away from the efficiency level that we believe we have potential to have. But on the other hand, looking at the half full glasses, that's our OTIF level service level numbers have been improving.
So there is a sequential improvement since the beginning of the rewire and now with the arrival of our new VP in the area of supply to organize this receipt of orders, of placement of orders, in the sequence of the orders, in the organization between factories and order-taking directed to the inventories that we have. So there is a series of different operational challenges that, at the end of the day, are going to hit our deliveries needs and, considering the demands we have in this seasonal peak seasonal demand. So I think you're far away from having this transformation of our operational cycle completely finished. There is still a lot to cover, a lot of ground to cover, but we have seen our on-time in-full OTIF levels improve sequentially.
So this is a positive, trend. In the last, earnings release, I, I commented in one of your questions that we have established policy, policies for certain products, and we are involving now in certain predictive, models, so we can have a, a more accurate, forecast and less disarrangement and mismatch. Maybe one of the most critical, factors that increase our, capacity to deliver is when we sell what's not in the inventory. So when you, you sell that, you add dozens of days to our operational cycle, and at the end of the day, this is going to hit the delivery time, the lead time for our clients. So we have been carrying out several actions to reduce the cycle and to have the best and most accurate possible in terms of, the sequence of production and delivery and in logistics.
We have been making adaptations to our mixing center. There is a structural adaptation that we started in the beginning of the third quarter to be able to have a level of delivery that is more, you know, on par with the demand. But we are also in a ramp-up period for the mixing center, so we are in the first year of operation. So we have this expectation that the mixing center itself is going to be able to dispatch, to ship more efficiently over the next months because of the actions that we're carrying out.
You know, all together, we should see continuous improvement in our levels and capacities of delivering in our OTIF, and at the end of the day, this is going to also impact the cost efficiency of our operation. But as I said, we all said, there is a lot of work to be done. The symptoms are good, the trends are good, but there is still a lot to be done. Andreas mentioned several important points. I just would like to clarify one important thing about your statement for Danny. I don't know where your sources are, but in fact, our own on-time in-full level, OTIF level, has been improving significantly over the year. But this is an average, a large average.
So if you go to certain stores in some regions, you are going to see that you are going to have more noise. In our own stores and franchise stores, they have a limited capacity for inventory, and since we had a lot of inventory that was not being sold as much as it should be sold, because of the huge complexity of portfolio and huge quantity of innovations, the frequency of deliveries that we should have, the frequency and the prediction we should have for specific items to avoid the out of stock items. This has created some complications to us, specifically in these channels. In the distributors channel, in the food channel, we have not seen this same level of noise that you mentioned.
But maybe your source is more connected to the channels I mentioned. And part of these problems, they come much more, not from the simplification, but from the mixing center. These were gigantic changes. Maybe you don't have an idea of how much it was, but we concentrated the entire operations of picking, warehousing, and shipment of all the products in an only place, in an only warehouse. A new warehouse with new layout, new personnel, new challenges to be faced. And its operation started at the very beginning of this year. So there is this learning curve that we see. And the warehouses that we used to have in our factories were used to create allocation of equipment and to increase our manufacturing capacity. So the learning curve was somehow a struggle to us.
As André said, we are ongoing, totally this correction, both from the operational and from the structural point of view of the original project of the mixing center, and it's still not over. It's probably going to be over closer to the beginning of next year. To add to that, we saw a second quarter where we saw a huge adjustment in the internal and retail inventory level, so the operation cooled down. From overnight, we saw the sell-out improve a lot. The inventories were adjusted, so now the rhythm of the operations has to keep up, has to catch up.
We see a challenge now over the past two or three months, a challenge to catch up with the and to have productivity, manufacturing gains, to be able to meet the demand per region and per product correctly, especially now at the end of the year, when it's our sales peak in Brazil. It's not a worry, it's a point for.... It's a point of attention, and there are several action plans ongoing to correct this topic. But I wouldn't at all connect this to the simplification. It's quite the opposite. Part of the results we have been are arising from simplification. We are reinvesting them to make sure that the execution of the brand, the execution of our operation is better made.
Thank you for the questions, Danny. And now the last question, João Soares from Citibank.
Thank you all for taking my question. Two quick questions. The first one is about the cost per pair. I'd like to hear a little bit about the cost per pair, margin. As you absorb cheaper raw material, I'd like to understand this curve a little bit. And the second one is about the expenses. You mentioned about you mentioned the marketing expenses and focusing much more on the working money rather than on non-working money. I would like to understand a little bit better, like, looking at the big picture. You know, there were some new expenses, there were some distribution expenses that weighed the SG&A, but thinking in a curve, how this should happen in the future?
Thank you, João, for your questions. Well, for your first question, you are right. We see an evolution in the cost per pair.
I think that if we look sequentially, we are having a significant improvement in the cost per pair, and this is not only due to the factor you mentioned of the raw material. This factor of raw material is actually going to impact us even more, because remember that we were pretty much not buying raw material at these cheaper prices. We had a lot of inventory of raw material, right? We had assembled a very large inventory of raw material at higher prices. So we were draining from stocked raw material prices that were much higher than the ones that are in market right now.
It's too bad that the main commodities we buy went to a 20-year low price, but we couldn't actually take advantage of this opportunity because of the inventory levels and because of the working capital levels we had. So at that moment, throughout the first, second and third quarters, we had a preference for much more to have liquidity rather than preserving our cash rather than trying to gain more inventory at a higher price. That it wouldn't make any sense because of the high level of inventories we already had. So we opted for not making this movement, and as a result, you see very little of this reduction, this average reduction in the raw materials costs in our products.
But there are other effects that actually are being seen right now in these effects and have had a large impact. All the measures we took in the manufacturing aspect to adequate the expenses and the payroll and the expenses overall, costs to the manufacturing level we have right now, this obviously has created a partial, at least a partial adjustment to our scale effect. In the first moment, when the volume goes down a lot, you have a scale economy. But as the year, you know, elapsed, we gained a lot of productivity in processes, and we managed to balance better the use of our factories, and as a result, we managed to find this decreased cost per pair. Especially when we look at the second and third quarter compared to the first.
We see the COGS converging to lower production costs in a significantly lower level than the average in the past. The COGS per pair saw an evolution of 1.8 BRL less per pair when you compare the second and the third quarter of this year. If you look at the year-over-year, there is still an economies of scale that has an effect, plays an effect here. But year, like, quarter-over-quarter, we have seen a reduction. And as times go by, we expect it to remain in this new level of productivity with this balanced level of production, starting now to go back to acquiring to purchasing raw material at a much reduced level. Much reduced costs to the ones to the cost of the raw materials we had in our inventory.
We will keep on following new opportunities for cost reduction, and we still see certain difference between the different raw materials, the different costs for products produced in different factories. So which means that there is optimization to be addressed. How to find out the way to reduce the production costs is by addressing these internal manufacturing issues. And as for the raw material prices, I think this is a factor that's going to impact us positively from now on. And now, in the second question, I'm not sure exactly what you want to explore, but I understand you want to discuss the overall marketing costs. Can you clarify your second question, João? João's connection just... João lost his connection. He's reconnecting.
Well, I will start answering the question as I understood the question. If that's another question, you can follow up with me and clarify. When you look at the marketing expenses level in the company overall, for many years, it never decreased. We got the last 15 years of history, and we have seen a growing level in marketing investments. On the other hand, when we look at other metrics, for instance, marketing compared to the revenue, yes, there was a compression over the years. And this compression took this investment in marketing below, to a number, to a figure below what we believe it's reasonable. This is one of the diagnoses we have made. So, the SG&A grew in other accounts, disproportionately to the volume, and somehow this was affected by marketing.
Marketing was a discretionary account that was the simplest one to be adjusted. So over time, this seems to have been one factor that led the marketing expenses to decrease compared to the revenue. This is one of the first points we believe that we should reverse from now on. Secondly, when you look at this number, this figure of marketing expenses, the non-working money had a disproportionate and unreasonable figure. So from now on, we are going to work more in the working money that is going to reach consumer and help the image of the brand.
So we understand that not only the investment level, the aggregated investment, the added investment level, should be more towards growth than improvement, but also the allocation of the marketing expenses in the working money and being supportive to the brand growth and to the attributes we want to reinforce with our project is the ideal for the future. This is adjustment. It's not something only related to Brazil. These are diagnoses that we are running right now across the world in all the geographies. The diagnoses for Europe and the U.S. also contribute to this understanding. We see that this movement, we have been reducing the marketing investments compared to what is reasonable in the market, and we were allocating in a way that was not supporting the brand.
It was more related to trade, marketing, and, you know, and not the front line. So we would like to reverse this trend. And overall, this is what you are going to see as a trend from now on. We don't have any guidance. We won't tell you what level of marketing investment is going to exist for next year, but I would like you to understand that we won't have any kind of adjustment that is going to sacrifice the brand. Quite the opposite. We tend to increasingly support the brand.
No, thank you for answering the question, Luiz. Now, back to Luiz to have his final remarks.
Thank you, André, Rafael. I think in summary, we are not satisfied with the results. There are a lot of improvements to be made.
But on the other hand, the evolution from the diagnoses that we made, that were very well made concerning the short-term priorities that we made, they were clear. And this work is more invisible for you because it's not going to result, to generate, to yield results in the short term, results. But we are working heavily on the planning for the future, in the planning the three- to five-year plan for each one of the geographies. What is the role of each one of the international markets? Which is the, the portfolio to be addressed in each one, the profitability work that had started, and also the manufacturing processes work that had started.
Above all, we continue to have super committed employees and excellent professionals that have been supporting us, and every day inspire me about the future and the potential of the company for the future. There is a hard and long path ahead of us. The results are below what we expected, but the improvements are very evident internally. Thank you all very much for your participation. We will remain available to you, and I wish you all a good day and a good week.