The briefing will be live-streamed only, and also the session will last until 5:35, and after that, from 5:50, on a different Zoom channel, we will have a Q&A session for institutional investors. Let me introduce the presenters of today's session: President and CEO Kotaro Sawada, and Executive Vice President and CFO Koji Yanagisawa. Thank you.
So, we have two presenters today. Now, Yanagisawa will take you through the earnings results.
Good afternoon. I will take you through the FY 2024 second quarter earnings results. The presentation material can be found on our IR site, so please take a look. I would like to jump right in and share with you our quarterly results. In FY 2024 second quarter, the GMV increased by 7.9% year-on-year to JPY 279.1 billion. GMV excluding other GMV increased by 7.9% year-on-year to JPY 260.8 billion. OP increased by 5.3% year-on-year to JPY 30.4 billion, and OP margin was 11.7%, decreasing by 0.3 points year-on-year. Those are our results. We have achieved 45.6% of our plan with respect to the GMV excluding other GMV and 47.5% of the OP plan.
With summers being hotter than last year, fall winter products started off slow, thus impacting the GMV. However, all of the businesses are, for the most part, on track. OP is also positive in comparison to the plan due to the unused variable costs as a result of the delay in promotion execution and fixed costs such as employee payroll and consignment fees. As of the end of the first half, we have reached record high GMV and OP. On page 7 of the material, here are the consolidated quarterly results.
During the second quarter, the GMV excluding other GMV increased by 8.3% year-on-year. September this year was even hotter than last year, so the sales of fall winter items remained sluggish, but we saw great results in July and August, especially during summer sales. With respect to OP, the GMV increased in the advertising business group, and we changed the shipping policy. This resulted in a higher gross profit, which surpassed the increase in costs from the new logistics center and higher shipping costs due to the changes in the economic condition.
Thus, OP increased 11.6% year-on-year. The OPM was 11.4%, up by 0.4 points year-on-year. Let me now take you through the business results in detail. Please turn to page 8. Here is the OP increase-decrease analysis as of the end of the second quarter and year-on-year comparison. From JPY 28.93 billion in FY 2023 second quarter, the OP has increased by JPY 1.54 billion to JPY 30.4 billion. The OP increased due to three factors. First one, it increased JPY 5.15 billion due to the increase in GMV of the ZOZOTOWN business and LINE Yahoo e-commerce.
It also increased JPY 0.98 billion as a result of the growth in the advertising business and JPY 2.13 billion due to an increase in gross profit resulting from the increase in shipping income arising from a change in the shipping policy. On the other hand, OP decreased due to four factors. It decreased by JPY 2.59 billion due to the rise in fixed costs attributable to an increase in employees and logistics centers, JPY 2.76 billion attributable to increasing variable costs such as costs that increase in correlation to the GMV growth and increasing shipping fees.
The OP decreased by JPY 0.81 billion as a result of increasing actual promotion-related expenses for customer acquisition initiatives and promotions using points, and JPY 0.56 billion due to the increase in other expenses such as cloud server fees. Now, if you turn to page 21, I would like to take you through the breakdown of the SG&A expenses. The SG&A GMV ratio was 23.6%, up 0.7 points year-on-year. Here are the reasons why the SG&A expenses have increased. There are three reasons. First, although the AOV has increased year-on-year, Yamato's shipping fee increase came into effect on April 1st this year. Thus, the shipping expenses have increased by 0.4 points.
Second, depreciation of material handling equipment for logistics center ZOZOBASE Tsukuba 3 has begun. Thus, depreciation has increased by 0.3 points. We also began incurring rent for ZOZOBASE Tsukuba 3 and DPL Tsukuba 2, so the rent has increased by 0.2 points. On the other hand, the SG&A has decreased due to the following two reasons. Automation in logistics centers has enabled us to have fewer people working in the centers. Thus, the logistics-related payroll expenses have decreased by 0.1 points. Second, the payment collection fee has also decreased by 0.1 point due to changes in the payment option composition. Now, on page 24, here are the trends in actual promotion expenses.
This quarter, the percentage of actual promotion expenses, which include advertising and promotional spend and point-related expenses, against the GMV was 3.9%. From the first quarter, we have pushed back our promotional spend, and we still have surplus budget in the second quarter. This percentage against the GMV has gone up by 0.1% year-on-year because we have spent more on web advertising and ran more free shipping campaigns for purchases over JPY 12,000. Now, if you turn to page 20, this is the quarterly trend for OP and OPM. This quarter, GMV and advertising business growth, as well as the increase in gross profit resulting from the changes in the shipping policy, have outpaced the increase in costs from the new logistics centers and shipping expenses due to changes in the economic condition.
Therefore, the OPM was 0.4 points higher, landing at 11.4%. Now, I would like to talk about the KPIs for ZOZOTOWN, starting with page 25. These KPIs do not include the results of LINE Yahoo e-commerce or the B2B business. First, starting with the total buyers, the total buyers increased by 80,000 from last quarter to 11.87 million. Of the active members increased by 110,000 from last quarter to 11.02 million, and guest buyers decreased by 30,000 from the previous quarter to 840,000. In the second quarter, as was the case with GMV, we experienced severe late-summer heat.
Thus, the growth in September was sluggish, but we were able to grow steadily from the first quarter. Next, page 28, the number of shops. As of the end of the second quarter, the number of shops were 1,621, with a net increase of 16 shops from the end of the previous quarter.
In the second quarter, we welcomed 34 new shops, including Sweden's fashion brand Acne Studios, as well as Kao Group's Kanebo and luxury brand Byredo. So they have joined our fold. Next, moving on to page 30, the average retail price. The average retail price was JPY 3,629, up 1.1% year-on-year. The brands continue to raise their prices. However, the launch of fall winter products in September was sluggish, so the increase in the average retail price remained limited. On the other hand, on page 31, the average order value was JPY 8,196, up 3.8% year-on-year.
More items were purchased per order. Thus, the AOV increased more rapidly than the average retail price. The reason why the number of items purchased per order increased was because, as I mentioned before, we ran more free shipping campaigns for purchases over JPY 12,000 in comparison to last year, resulting in people buying more items together. On the day of these campaigns, and this drove the number of items purchased per order upward. Lastly, on page 33, here are the consolidated earnings forecasts and dividends. We have not made any changes to these plans. This concludes my explanation. Our President, Sawada, will take it from here.
Good evening. Thank you. So, usually I talk about the strategy and the progress. But in recent years, using AI and technology, we do receive a lot of questions about that. We want to talk about how we intend to use AI and how we intend to automate our processes. That's what I'll be covering today. This is AI technology automation initiatives. Basically, we are embarking on various initiatives, but we're focusing on these three segments. We utilize AI to raise immediate sales and also automate our processes. That's in the short term. You see these initiatives in the short term, and also improvement of operational efficiency. We're utilizing AI and LLM. I would like to introduce some examples of our technology usage in that respect.
Over the long term as well, we're thinking about how we can use AI and automation technology. I would like to cover that as well today. We'll start from short term first, introducing some case examples. First, within ZOZOTOWN, how are we using AI? This is just one example. We have two screens. This is the same top page, ZOZO app top page. This just shows how different it will be for individuals. Of course, we target the masses. Our products cater to the mass. But of course, these are fashionable products that we handle. Somebody who's in their 50s, like me, like a gentleman, might have different preferences.
I mean, if I saw some fashion that was more catered to younger men, I might think that that's not for me. We knew that that has been a challenge. We wanted to personalize for each individual. We've been focusing on personalization. By doing so, we believe that this will improve our retention. That's our strength. With respect to the logic behind this, we are, of course, making minute adjustments every single day. If you have time, please compare your top pages to see how different they are in relation to others. Next, in recent years, climate or weather has had a significant impact on the fashion industry. In order to alleviate that impact, this is something that we've done.
Simply put, depending on where the person is looking at the products from, we automatically show the right products for that weather. We're using Weathern ews API. We actually use original information, location data of that person who's looking at our site, and we recommend the products that best fit that weather in the area. But in recent years, it's been quite warm. There hasn't been that much of a regional difference, but it is becoming cooler now. We hope that we expect that these functions will have a bigger impact. Next, the WEAR front page as well, where we are implementing AI here significantly. This spring, we renewed the WEAR service and on the top page, we will show styling photos that we believe that person will like.
This is not just something that is similar. It's not a simple technology as that. There's a lot of information that is included in the styling photos, and fashion genre is very complicated. We try to incorporate different information. We can define some style as girly, or we define these fashions ourselves. Based on that definition, we use the browsing history as well to show images of styles that we believe that our individual users would like. The more the user clicks on images, the smarter the AI will become. It will learn. It will continue to learn the user's preferences. We hope that you will give this a try and continue to use it. Next, this is not directly going to impact our sales. This is a little bit more indirect.
But over the last couple of years, we have put a lot of effort into our measurement technologies. We started with ZOZOSUIT, ZOZOMAT, and ZOZOGLASS. You may remember our technologies. But over the last six months, we basically had a new technology we released is ZOZOMAT for kids. These are kids' shoes that we recommend. This can recommend kids' shoes for children. What's most different from the traditional ZOZOMAT is that we actually took into account a child's growth trajectory. We can say these shoes might be slightly bigger now, but the child will be able to wear it for a year, for example. This is very helpful, convenient to mothers with small children. We hope that you will give this a try. Of course, we will utilize the data to see how that child has grown. We actually statistically analyze that child's data to make more accurate recommendations.
We focus on the usage of technologies that contribute to sales or retention. But from here onwards, we want to talk about Gen AI. We have a lot of data. We are an internet company, so we want to organize our data. We're utilizing Gen AI behind the scenes as well. It's not really anything new. It's quite primitive, but we're utilizing Gen AI in all areas. One example is the item review. We implemented reviews last year, and we need to monitor these reviews as well because sometimes it goes against the rules or our manners. Before, we used to check the content of the reviews manually, but we can use Gen AI, and we can automatically patrol the content now. Thankfully, we were able to significantly reduce the time spent on these kinds of tasks.
Moving on, this is the automatic categorization. You might not know what this is about, but on Fashion EC, products, it's important to tag products and make sure that they go into certain categories because people search for items by selecting categories. Sometimes categories are in the gray, and the sales will differ considerably depending on which category you categorize them in. We can utilize Gen AI to automatically categorize these products and choose the optimal category. We implemented an automatic system that will enable the brands to do so as well. Next, this is more about customer support.
We use a BI tool that will actually... We are using LLM to basically... A lot of other companies may do so as well, but we actually look at the comments or the content of the inquiries and feed that into LLM. The system is a dream-like tool because we used to do it manually, but now the system will automatically categorize the type of inquiries. We're using AI in that way. That was about making our operations more efficient. From here onwards, we can go to the next page. Over the mid to long term, we want to use LLM in these areas. This is just our direction. First, we want to utilize it for niaulab. This is just one physical store that we currently have. This is a prototype to actually understand what looks best on people, but we want to use Gen AI as well.
We introduce on wear as well, but styling photos as well. We want to make sure that we can take fuzzy images and virtually recommend what the best looks are for individuals. This might be difficult to understand if I just explain. We have made this publicly available, but there is a video that explains what we mean by this. Please take a look. Choosing clothes can be fun, but it can also be hard. What should I wear? What should I wear? Something trendy, my personal color. What's the right look? Everyone is, of course, worried. This is a place where everyone can find their best looks. That's a niaulab. This is the niaulab. Let's check it out. Based on basic information such as height and weight and your favorite style.
If you upload your usual styling, your current fashion analysis will be displayed. To try on clothes and get styling suggestions, please visit the niaulab. Welcome. Good afternoon. niaulab AI is your partner to find your best looks. Please feel free to consult with us about your fashion concerns and styling challenges. Consultation with AI, pre-questionnaires and measurement results, also purchase history, wear information. The AI will analyze various information and match the best styles, and you can choose what you like, and you can try on these items as well.
At the niaulab, we have prepared a large number of items to suit a variety of styles. Haruka, who is tall, is wearing a pair of wide cargo denim to lower her center of gravity for a more balanced look. In addition, an oversized trench coat will give her a seasonal silhouette. The contrast between the large collar and the oversized trench coat makes her shoulders less noticeable, which she had felt self-conscious about. We will also share tips on how to wear hair and makeup to suit your unique style. Next, a layered style in spring colors. It looks great. Let's try a mode style. Match a long gilet with a bottom in the same color to make it look like a maxi length dress. It also emphasizes the I silhouette, making the most of her height.
We are glad she liked it. I know she tried on on the niaulab. I linked to her list of liked items on ZOZOTOWN. Please take your time before purchasing, and thank you for your purchase. Further consultation with AI is also available on ZOZOTOWN and WEAR. We can, for example, suggest jackets that go with the skirt that you uploaded on WEAR in the past. If you want to find the best look, please feel free to contact us at any time. Now we get it. You were looking for an outfit for the party. You guys look great. Finding your best looks can be fun and exciting. With love for fashion and the power of technology, we hope to make the world a better place where people can find their best styles. [Foreign language]
Thank you for watching. [Foreign language] that's what we have planned for the future. We haven't realized everything yet, but AI agents are going to be right around the corner. I also believe that as well as ZOZO. We want to create a fashion AI agent or stylist. We're trying to do that, and the technology and data of course we believe that we have the right technologies and data to do so, please look forward to what we have planned for the future. Lastly, this is called REVINAL. It's a virtual clothing shop. We're working on this as well. In the chat, Metaverse, ZOZO NEXT, we are going to sell these items on ZOZO NEXT. It's not just real clothes, but we are going to sell virtual clothes.
That's something that we're trying to do as well. Gen AI, with the power of Gen AI in the virtual world, we can actually design clothing with great freedom. This is something that anyone can do now. But as a technology, we don't have a lot in the world. There's no technology that to create the virtual clothing and make it into a real clothing. You need to create 3D models, make sure the sizing is right, the drapes on the skirt are right as well.
There's a lot that we, I mean, if you had the time, you can do that, but it's not automated. But in the future, we will be able to connect virtual with physical. We are actually studying the technology that will enable us to do so. That's it for me.
Thank you for watching. That concludes ZOZO's FY 2024 second quarter earnings briefing. Thank you for watching today.