Nebius Group NV (NBIS) 2020 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to the second quarter 2020 financial results call. (Operator Instructions) I must advise you this conference is being recorded today, Tuesday, the 28th of July 2020.

We'd now like to hand the call over to your first speaker today. Yulia Gerasimova, Investor Relations Director. Please go ahead.

Hello, everyone, and welcome to Yandex's Second Quarter 2020 Earnings Call. We distributed our earnings release earlier today. You can find its copy on our IR website as well as our news flyer services.

On the call today, we have Tigran Khudaverdyan, our Deputy Chief Executive Officer; Daniil Shuleyko, our Chief Executive Officer of Yandex. Taxi; and Greg Abovsky, our Chief Operating and Chief Financial Officer. Arkady Volozh, our Founder and Chief Executive Officer; Vadim Marchuk, our VP of Corporate Development; and Yevgeny Senderov, Chief Financial Officer of Yandex.taxi will be available on the Q&A session.

The call will be recorded. The recording will be available on the IR website in a few hours. As usual, we've prepared a few supplementary slides, which are currently available on the website.

Now I will quickly walk you through the Safe Harbor statement. The various remarks that we make during this call about our future expectations, plans and prospects constitute forward-looking statements. Our actual results may differ materially from those indicated or suggested by the forward-looking statements as a result of various important factors, including the impact of the ongoing COVID-19 pandemic as well as those discussed in the Risk Factors section of our annual report on Form 20-F dated April 2, 2020, which is on file with the SEC and is available online.

In addition, any forward-looking statements represent our views only as of today and should not be relied upon as representing our views as of any subsequent date. Although we may elect to update these forward-looking statements at some point in the future, we specifically disclaim any obligation to do so even if our views change. Therefore, you should not rely on these forward-looking statements as representing our views as of any date subsequent to today.

During the call, we will be referring to certain non-GAAP financial measures. These non-GAAP financial measures are not prepared in accordance with U.S. GAAP. A reconciliation of non-GAAP financial measures to the most directly comparable GAAP measures is provided in the earnings release we issued today.

And now I'm turning the call over to Tigran.

Thank you, Yulia, and thanks to everyone for joining our call today. Despite the difficulties of the second quarter, we managed to mobilize the company's resources in order to overcome the challenges while supporting our users, partners and employees. In total, we allocated over RUB 1.5 billion to various initiatives, including our Helping Hand program, our support fund for taxi drivers and couriers, advertising credits to small and medium businesses as well as our education efforts. The hard work of all our teams allowed us to minimize the negative impact on our financial performance, and Greg will share more details on this a bit later.

Today, we are focusing on supporting our businesses in the recovery phase while also capitalizing on new opportunities. One of these opportunities was the launch of logistics services inside our Yandex. Taxi segment, which I believe will further expand the total addressable market for our mobility businesses.

Another important milestone for us in Q2 was the reorganization of our joint ventures with Sberbank. The transaction was closed last week. As a result, Yandex became the controlling shareholder in Yandex. Market while exiting the Yandex. Money joint venture. We believe the full integration of e-commerce services into the Yandex ecosystem will unlock significant synergies in multiple areas. We're also excited about the termination of noncompete obligations in financial services, which will allow us to pursue new opportunities in the fintech space.

Let me give you a brief overview of some of the key trends in Q2. Search and Portal segment performance in Q2 was clearly impacted by COVID-related self-isolation restrictions, particularly in April. However, our platform remains a destination of choice for our users, which is reflected in the strong growth of our search share and in the number of the queries. In June, we reached a record 58.5% share on Android, growing 260 basis points from March and 600 bps from June 2019. Our overall share grew to 59.6% in June, up [270]basis points from previous year. Over half of our Search and Portal revenue comes from mobile.

Strong search query growth continued throughout the quarter, up 29% year-on-year, driven by both mobile and desktop traffic. After the peak in April and solid growth in May, we began to see a normalization of total queries growth in June and July on the back of easing lockdown measures. At the same time, we saw a recovery in commercial queries, which is important for monetization.

Yandex is constantly improving its advertising technologies. And in Q2, we launched a fixed CPA model, which is a pay-per-action model, helping our ad clients to increase the effectiveness of their advertising placements. We continue to be excited about that, and the team keeps introducing new features and demonstrating strong audience growth and user engagement. In June, Yandex audience reached 16.8 million daily active users. An important milestone in Q2 was the full integration of Zen as the main feed in Yandex search app on Android. As a result of the time spent on Zen within our search app improved by 10%.

Another important milestone this quarter was the progress on integration of video into the feed. Currently, video accounts for 15% of total time spent compared to a low single digit a year ago. Based on the trends we see month-to-date, we expect Zen's annual revenue run rate to exceed RUB 8 billion in July compared to RUB 7.9 billion in March. In terms of year-over-year growth, we have not come back to pre-COVID dynamics, although we are seeing continued improvement here.

Turning to Media Services. As we previously mentioned, the stay-at-home restrictions led to a solid inflow of new users onto our video platform. The total number of Media Services subscribers reached 4.5 million in June. The retention rate of [new trial] subscribers who joined KinoPoisk (inaudible) in April has been better than we anticipated. We are particularly pleased to see that the share of paid subscribers increased significantly compared to March.

Now let me give you an update on Yandex. Market. GMV Marketplace increased by 3.5x in Q2, supported by wider assortment, higher number of orders and strong (inaudible). Similar to other businesses, which have benefited during the pandemic, the GMV dynamic for the marketplace has been normalizing in July, and month-to-date growth is 2.6x, now about 6,000 partners, so very good on our marketplace. And the share of (inaudible) sales in GMV reached 56% in Q2. The assortment expanded to over 1 million SKUs, and we expect that it will reach at least 2 million goods by the end of the year. In our price comparison business, the daily audience reached 4.5 million unique daily users in Q2, and the platform continues to demonstrate solid revenue performance in July.

Turning to self-driving. As of the end of June, we drove over 4 million miles in the autonomous mode, which puts us firmly among top 3 companies globally by total distance driven. Our fleet reached 130 self-driving cars, and we expect to add 70 more by the end of the year within the framework of our cooperation with on Hyundai Mobis. In Q2, we began commercial usage of Yandex. Rover, our delivery robot, with [coc coc] being our first client. We believe there is a great potential for rovers commercial deployment in FoodTech and e-commerce businesses.

Last but not the least, education. In April and May of this year, over 2 million children were studying from home with our Yandex. School platform. The users watch our video classes more than 4 million times. Practicum by Yandex, our online educational platform offering multiple IT career advancement programs, launched 6 new courses. And the platform total bookings in Q2 increased by 1.6x compared to Q1 and by 6.6x versus Q2 over the last year. Education remains an important area of focus for us, and it's also central to a number of our corporate social responsibility initiatives.

Although we see some continued uncertainty in the short term, we are confident that we remain very well positioned to deliver sustainable long-term growth. With continuous improvement in our core services, the record development of new businesses in FoodTech and logistics, the return of the e-commerce business to Yandex and the possibility to seek new opportunities in financial services, our ecosystem is stronger than ever. We believe that after the recent capital raise, with ample financial flexibility to continue investing into all these projects while maintaining our usual prudent approach to capital allocation.

And with this, I'm turning the mic over to Daniil.

Thank you, Tigran, and hello, everyone. Q2 was indeed very challenging, but I'm very pleased with what we were able to achieve during this quarter. The quarter began with a significant decline in ride-hailing, and we had to rapidly adjust to the new reality. We focused on creating additional demand for our driver platform by connecting them to food delivery last-mile logistics services. This means that our driver supply allowed us to quickly expand the coverage of our food and grocery delivery businesses and to provide restaurants with a wider delivery parameter while our users were able to order meals from their private restaurant even if this restaurant are located far away from their homes.

Because pre-COVID, our restaurants and grocery delivery businesses utilized the [best trend] of bike couriers. Grocery delivery is now available to more than 20 million people in Moscow, Moscow region and Saint Petersburg and (inaudible). We started to provide last-mile delivery options for large retail chain, e-commerce and classified companies. Now we offer thousands of businesses of all sizes a last-mile delivery solution. Our ultimate goal is for companies to see that there is no operational or financial rationale for them to support their own logistics service given that the combination of our logistics technology and driver and courier supply can deliver a low-cost, fast and reliable delivery service.

The pandemic became an opportunity to further strengthen our business. Within a short period of time, we were able to make our platform more flexible, thus, allowing us to adjust quickly. We expanded taximeter, our driving based software to couriers, which allow them to take delivery orders. This was previously available for drivers only. This led to an improvement in delivery service coverage and quality.

Within our Helping Hand program, while working on delivering COVID test, we significantly fine-tuned our route optimization efficiency, which we recently rolled out for multideliveries and (inaudible)

We rolled out our logistics services in 350 cities across 12 countries. And we launched our food delivery in 87 new cities where we [have food service] pre-COVID. As a result, it is now available in 130-plus cities, and 86 of them will provide our own delivery. All these initiatives allowed us to finish the quarter in excellent shape.

Total revenues were up 42% year-over-year, driven by FoodTech, B2B Taxi and B2B Logistics. Ride-hailing was up 3% in revenues with B2B Taxi and B2B Logistics together, up 49% year-over-year. Ride-hailing, excluding B2B, however, was down 7% year-over-year.

Rides were down 6% year-over-year in total. The rough week was the week of March 30, right after the official lockdown started. During this week, rides were down 39% year-over-year. Less than 2 months later, during the week of May 25, rides returned to the positive year-over-year growth. While in June, they grew in the mid-teens year-over-year.

As you see, it really was a V-shape recovery, GMV didn't keep up with growth of rides due to a decline in (inaudible). But in July, we have been seeing a continued strength in GMV trend. July month-to-date GMV is single digit above pre-COVID levels, up 20-plus percent on a year-over-year basis.

We see room for further recovery. Airports rides are still down approximately 75% from pre-COVID level. Many of our B2B customers are still working from home, while our most active rider cohorts has not come back fully yet. As far as I've seen in market share trend, in April, we see that some portion of share to our competitor in Moscow. This was because of the tariff mix effect. Our share in payment tariffs, which were restricted during the pandemic, has historically been high. So with premium tariff temporarily gone, economic tariffs grew as a percentage of the total and resulted in short-term redistribution of shares. But we strengthened our market position in May and even further in June.

Our competitors' share returned to the pre-COVID levels in the low teens in terms of the number of ride according to our estimates. Obviously, in terms of bookings, their share is lower still.

So that was a clear beneficiary of COVID. Yandex orders doubled year-over-year. GMV grew 2.3x for our GAAP revenues tripled. The number of restaurants exceeded 26,000 in June, up 2.5x versus last year. We lowered our cost per order, thanks to an increase in order density as well as the part that we implemented a new dispatch algorithm, optimized incentives, introduced free floating couriers, and we saw the shares of self-employed couriers significantly increase. July trends continue to be encouraging. Delivery has been growing in very high double digits year-over-year, and GMV has been doubled.

Lavka is growing rapidly. GAAP revenues net of incentives reached RUB 2.3 billion. We have 175 dark stores opened as of the end of June. However, our approach to grocery delivery is beyond just Lavka. We partnered with (inaudible) across a number of our businesses. It's now delivering groceries from over 850 stores of (inaudible) with approximately 7.5000 SKUs in aggregate. Our last-mile delivery services provide white label delivery services for many other retailer chain, including Globus, Metra, Ashan, Dixi and others. All in all, we are entering Q2 with a much more diversified portfolio of exciting businesses. And what is also important, we are in a strong position from a cost-efficiency perspective. In Q2, we cut approximately RUB 1.7 billion in fixed cost versus our recorded budget. As a result, ride-hailing continued to be profitable and funded our new initiatives despite a significant decrease in demand on the ride-hailing front.

With this, I'm turning the call over to Greg.

Thank you, Daniil, and hello, everyone. We've disclosed our key financials in our press release. It's been a very challenging quarter, but we're very pleased that we've managed to finish with a positive adjusted net income and higher-than-expected adjusted EBITDA despite the fact that we had a number of one-off expenses. We had over RUB 1.5 billion spent in a variety of targeted COVID-related initiatives, including RUB 408 million of expenses for personal protective equipment, our Helping Hand project and the driver and courier support fund. Our adjusted EBITDA also includes RUB 177 million of advisory fees related to the recently announced transactions. Overall, the results exceeded our internal expectations. The key drivers behind this better-than-expected profitability were rigorous cost control, in particular, optimization of personnel expenses, advertising and marketing costs as well as other overheads, improving new economics in certain businesses, particularly in Yandex. Eats and faster-than-expected revenue recovery in our ride-hailing, advertising and classifieds businesses.

Now let me focus on the performance and current trends across our business units. Search and Portal. Search and Portal revenue declined by 12.5% year-over-year on a reported basis and by 8.8% on an ex-TAC basis. As we said previously, we've seen a sequential recovery in ex-TAC ad revenue trends from a high-teens decline in April to a low double-digit decline in May to flat growth in June, all on a year-over-year basis. The positive trend is continuing in July. Our ex-TAC revenue is growing by mid-single digits year-over-year on a month-to-date basis. The performance was primarily driven by SMEs, which are recovering faster than larger enterprises. There's still a high level of uncertainty over the pace for the recovery given the limited visibility on the economic consequences of pandemic as well as the oil price shock. We see faster recovery in search and Yandex websites compared to those of our partner (inaudible), which remains under pressure. We expect this trend to continue, and growth of Yandex properties is likely to outperform as that of our advertising network.

Sector-wise, we see recovery in some of the areas that suffered the most during the pandemic, for example, auto, apparel, real estate and even to a limited extent, travel. All these sectors are still in negative territory in terms of year-over-year growth. However, the second derivative is positive in all of them. Among the sectors demonstrating the highest recovery are finance and insurance as well as B2B. We continue to see strong growth in home and garden, consumer electronics and home appliances categories.

Two sectors which benefited most during the pandemic, IT and telecoms as well as FMCG, continued to perform well, although the growth rates are normalizing as the lockdown measures are easing. Overall, slightly more than half of all sector categories we track are demonstrating positive growth in July. While in April, this was only a quarter.

Moving to Taxi. Taxi revenues increased by 42% year-on-year, primarily driven by the strong growth of our FoodTech services as well as our B2B business. The team has achieved impressive results in terms of adjusted EBITDA. And here, I particularly like to highlight the following. We finished the quarter with an adjusted EBITDA of RUB 253 million and a 2% margin. This number includes a RUB 740 million loss in our self-driving business as well as segment-related COVID expenses, which amounted to RUB 273 million in Q2. And it also compares to an adjusted EBITDA of RUB 115 million and a 1% margin in Q1 of this year. The combined adjusted EBITDA of our ride-hailing and FoodTech businesses, i.e., excluding investments in self-driving, was RUB 993 million in Q2 and was driven by solid profitability of our ride-hailing service as we were able to optimize a portion of our costs almost immediately after the start of the lockdown, significantly improved losses in Eats, which helped to offset increased investments in our grocery delivery business, Yandex. Lavka, on the back of strong demand for this service. The adjusted EBITDA margin, excluding self-driving, amounted to 8%, up from 7.6% in Q1.

Turning to other businesses. Media Services again demonstrated very strong revenue growth of 94% year-on-year despite a much higher base from Q2 of last year and close to 0 revenues from Afisha on the back of the COVID. Subscription-based revenues of Media Services grew by 164% year-on-year in Q2, an acceleration from the already strong 152% achieved in Q1. We continue to invest in improving content quality and growing our base of paid subscribers. Trends in July to date are comparable to the Q2 dynamic.

Classifieds revenues decreased 32% in Q2, primarily as a result of the outer dealerships being shut down from late March to early June and the dealer support mechanisms, which we launched immediately the after shutdown. We're encouraged by the rebound in revenue growth that we saw in June after the dealerships were allowed to reopen, which shows that we were able to monetize the deferred demand from April and May. The improvement has continued in July month-to-date with the revenue growth in the high teens.

Finally, onto other bets and experiments. Our revenues have declined by 18% year-over-year, driven primarily by the decline in Yandex. Drive as car-sharing services were completely suspended in Moscow and Saint Petersburg for nearly 2 months between mid-April and mid-June. Demand for car-sharing is now recovering, which drives an improvement in cars utilization. As of now, we have fully recovered to pre-COVID levels in terms of car utilization. The trends in revenue are even better, supported primarily by the change of user's behavior post pandemic. Consumers now rent the car for a longer period, and an average order duration in Moscow is now 30% higher than in Q1. Revenue growth in other businesses and the other bets and experiments line was positive. In absolute terms, GEO and Zen were the key contributors to Q2 revenue.

While the growth rates in both businesses have slowed down, we're now seeing an improvement on the back of increasing advertisers' activity. Despite the much of the revenue hit taken Yandex. Drive, we've managed to contain the losses for this business, which in absolute terms were actually better than in Q1 due to the optimization of our lease agreements and some other costs optimization. Total losses of other bets and experiments amounted to RUB 2.3 billion, with GEO and cloud contributing the most after Yandex. Drive.

Now let me give you an update on the performance of Yandex. Market. In our Q2 results, we still accounted for Yandex. Market using the equity method. But we will be consolidating the business of Yandex. Market from July 23 when the deal actually was closed. Total revenues of Yandex. Market Group amounted to RUB 7.2 billion, which implies year-over-year growth rates of 89%. Revenue from the price comparison segment accelerated to 33% in Q2 compared to 27% in the previous quarter. The adjusted EBITDA loss in Q2 of Yandex. Market was RUB 1.8 billion, and adjusted net income was negative RUB 1.8 billion as well. If we had included Yandex. Market into our Q2 financial results, our consolidated adjusted EBITDA corrected for intercompany eliminations would be around RUB 6.6 billion, and our adjusted net income would have been around RUB 1.2 billion. Yandex. Market had RUB 17.8 billion of cash and term deposits on its balance sheet as of the end of June and no outstanding debt.

Our balance sheet remains strong. We had RUB 3.5 billion in cash as of the end of Q2, excluding RUB 250 million related to Yandex. Market, and this includes the proceeds of the recent capital raise. Around 2/3 of the cash is in U.S. dollars. Our cash position allows us to maintain strategic flexibility in our capital allocation decisions, including our decisions to invest into existing businesses or look for potential new opportunities.

With this, I'm turning the mic to the operator for the Q&A session.

(Operator Instructions) The first question we have today comes from the line of Slava Degtyarev from Goldman Sachs.

My question is on Taxi. You executed better versus your end June guidance on taxi line, both on revenues and EBITDA. What was driving that, please? And was there any material incremental acceleration of growth?

Slava, it's Yevgeny Senderov. So look, if you decompose our 42% revenue growth rate, right, the main contributors, as Daniil already said, were Lavka, Eats and B2B, including B2B logistics, right, in a descending order. Ride-hailing revenues, excluding B2B, were slightly down. But include b2B itself, grew 49% year-over-year, gave us about 8 percentage points out of the MOU revenue growth rate. And B2B logistics, which partly accounted in B2B, was roughly half of that growth. So FoodTech drove the rest of the growth. Eats' revenues tripled. Lavka revenue reached RUB 2.3 billion. Lavka revenues booked on gross basis less incentives, and actually about just under 20% of MOU revenue.

If we're looking -- getting back to ride-hailing, looking at July, our GMV month-to-date is growing 20% year-over-year, and rides are roughly in line with that number as well, over 20%.

Okay. And my follow-up would be also on Taxi. There were press comments about DiDi entering the Russian taxi markets in the coming months. Do you have any thoughts about the ambitions in the local market and the potential strategy to define the market share on your side?

Thanks, Slava. Well, look, DiDi obviously is a strong player with access to capital and sophisticated technology. Obviously, their launch, when it happens, will increase the competition level, both on user side and driver front. But we believe that, first and foremost, DiDi will be -- the market position of smaller share players. As far as competition for driver supply is concerned, we think that multi-apping capacity for drivers is limited. They really cannot use, efficiently, more than 2 driver apps at the same time. And we believe that our Yandex driver app will remain one of the mainstays for the drivers. It provides the highest utilization rate on the market. And the higher utilization rate, the higher driver earnings are, and this has always been our primary focus. As far as users are concerned, I mean, DiDi and Citymobil will both, and [PL], for that matter, already have sophisticated user apps. And they're backed up by good technology. So they provide similar user experience. Both companies are well capitalized. And City and DiDi and both compete for audience by subsidizing. Meaning, initially, their target audience are going to be very similar, and we expect them to primarily compete for the same riders. So looking ahead, we feel confident about our competitive position and strategy on the market. I think we have a clear plan on how to protect our leadership by focusing on superior efficiency of service, higher driver utilization, quicker ETA rates as well as driver and partner earnings, as I said, it's ready for us, and by utilizing our multi-brand approach. So we feel pretty confident about our market position long term.

The next question today comes from the line of Ulyana Lenvalskaya from UBS.

I just wanted to elaborate on taxi a bit more. The self-driving car revenue we see for the first time ever is just from the delivery robot you've mentioned? And do you have any longer-term targets for the self-driving car unit in terms of potential revenue contribution?

Hey, Ulyana, it's Greg. Let me jump in on that. The revenue represents both the small contribution from Skolkovo as well as some revenues generated from the robotaxi service that we will be launching in Michigan later this year. In terms of longer-term plans, obviously, the goal is to develop a fully-fledged robotaxi service where we're developing just the driver. And that's why we're investing as much as we are and putting driverless miles and testing in real-world conditions on city streets in Moscow and elsewhere. And hoping that over time, this will be something that will really differentiate both the Yandex. Taxi business model and as well as provide further revenue streams beyond the applications outside of the footprint of Yandex. Taxi.

That sounds very good. And I also wanted to ask about the services. You reported that the subscribers of 4.5 million, which is basically more or less unchanged, first quarter. I was really surprised why it wasn't stronger during the lockdown. Maybe you can comment on that. And if possible to KinoPoisk versus Music? And on that, just speak for a while about Yandex. Plus, how -- the strategy there?

Sure. So a couple of things. The number of subscribers did grow. And in terms of total subscribers, they're growing strongly on a year-over-year basis. In July, we're already at 4.6 million subscribers. And Q1 level, I think, was probably in the low 4s, 4.2 million, 4.3 million. And I think we're generally very happy with what we're seeing on the subscription front. If you take a step back and think about kind of what is the goal of Yandex. Plus, Plus is this, sort of, umbrella subscription service, which includes access to music, to videos, to movies, to TV shows, to discounts on Yandex. Taxi, to other benefits from -- within the Yandex ecosystem, whether that's in drive or whether that's e-commerce or anything else. And so over time, what we'd like to do is we'd like to get as many subscribers as we can, paying a certain fixed monthly subscription fee and benefiting from the entire variety of Yandex services that we have to offer. So I think the differentiation, if you will, between a stand-alone streaming music service and us is that for the same price, you essentially get a whole lot more. You get original content in the form of TV shows and movies. You get discounts on Yandex. Taxi. You get discounts on shipping on our marketplace, e-commerce business and so on and so forth. So that's -- we believe this is going to be one of the key differentiating factors of the Yandex ecosystem going forward.

And what's the number of Yandex. Plus users now? Can you disclose?

It's 4.6 million as of July.

(Operator Instructions) The next question comes from the line of Lloyd Walmsley from Deutsche Bank.

Two questions. First, on marketplace. Can you talk a bit more about how full ownership frees up, kind of, strategic considerations for e-commerce? And is payments a strategic focus for you all now that Yandex. Money is fully separated? And then the second one also related to the marketplace. Can you help us understand Beru and kind of some of the KPIs and how much of the business is 1P versus 3P? And how to think about unit economics, maybe kind of take rates on the 3P side, gross margins on the 1P side and kind of how you'll be accounting for revenue kind of gross versus net? Any kind of high-level color you can share on that would be helpful.

Lloyd, so yes, the transaction closed last Thursday, and we've obviously been very busy working on how best to integrate the various assets together. Those include, kind of, utilizing the full breadth of services that we offer to merchants on Yandex. Market and then driving them to advertise on other places in Yandex, in Yandex. Direct and vice versa, taking advertisers e-commerce and moving them over to the Yandex. Market side. I'd say that's one area of focus. The other one is just driving more traffic to our marketplace business from various Yandex properties. I'd say the third one where the focus is on increasing the number of transactions in terms of improving conversion rates and more traffic. And then finally, it's the integration of logistics piece that is offered by the Yandex. Taxi platform.

In terms of accounting, it's -- currently, the marketplace generates approximately 56%., i.e., 3p generates approximately 56% of the business. And the other 44% is 1P. 1P is obviously accounted on a growth basis. 3P business is accounted as a commission-based business. Obviously, our take rates are currently low, and we expect that they will evolve and grow over time. The other thing to keep in mind is within that revenue number that I gave in my prepared remarks, it also includes the price comparison business, which is the, kind of, the original Yandex. Market, where people go and compare prices on different goods and where merchants can bid for traffic with CPC-like advertising. That business is alive and well. It's growing rapidly. It's up 33% in Q2 compared with a 27% growth in Q1, and it has extremely high EBITDA margins. So in Q2, it had 44% EBITDA margins. So taking a step back, you obviously have the marketplace business, which you're developing, which consists of 1P and 3P, where 1P is 44% of the GMV and 3P is 56% of the GMV. That business is loss-making. So we're still working on improving the unit economics there. And attached to it is the price comparison business, which is healthy and growing, and in some ways, it helps the 2 develop. And we think by integrating it together with Yandex, we can even accelerate even more the rate of growth of our e-commerce business.

Your second part of your question was on payments. And I would say that it's -- I would say, our ambitions are probably bigger than just payments. I think we are very interested in financial services in general. We're spending a lot more time on it. And I think importantly, post the sale of our minority stake in Yandex. Money, we are free and clear to go and pursue those opportunities. We'll probably do those prudently and slowly, but we think that this is a very important business for us to have together with e-commerce, together with taxi, together with all the other businesses that we have.

The next question today comes from the line of Cesar Tiron from Bank of America.

Yes. I have 2 actually. They're linked to your Search and Portal business. The first one, I wanted to understand, how do you think the Search and Portal revenue or the contacts revenue in general will evolve versus social networking in Russia in the next couple of quarters? Of course, it underperfomed in Q2, but do you think that will normalize in the next couple of quarters?

And then the second question that I have was really on the cost side of the Search and Portal. It looks like the TAC was very low in the quarter. Is there anything delayed? Or is there any savings that you've made? And what do they relate to?

Cesar, it's Greg. So I think we are very optimistic about the outlook for the Search and Portal business. We're seeing extremely strong trends, I believe, in that segment. And I could just walk you through sort of the pacings, if you will, over the last couple of months, right? March, which was sort of the last month here before COVID shutdown really took place, our Search and Portal ex-TAC revenues were up about 7%. A month later in April, this was kind of the trough. And revenues in Search and Portal on ex-TAC basis were down high teens, about 17% down. May, down about 10%. June, flat. And now in July, we're actually seeing 5% to 6% growth in Search and Portal business on an ex-TAC basis, which I think is pretty remarkable, and I think compares us extremely favorably versus other digital advertising platforms. And as I've said on the previous calls, I think the main reason for that is just the extremely high ROI that we are driving for our customers, especially on search. So inside that Search and Portal business, obviously, you have different parts of advertising, right? You have our ad network, our Yandex website, you have search. But I'd say most importantly is that search is growing the most.

And then on your question on TAC, I think that what we've also done is we have gone in and looked to optimize our traffic acquisition costs, both on the partner side as well as on the distribution side. And you see the benefits of that flowing through, and we expect that those will continue into Q3 and later.

The next question comes from Vladimir Bespalov from VTB Capital.

My first question would be on the competitive environment. We recently heard Spotify entered the Russian market. DiDi started to provide services in (inaudible). So maybe you could comment what you expect from these new players on the market, how you would compete with those? And in general, maybe not only with them, but with the players on the market?

And the second question would be on CapEx. I see on your slide that there was a big spike in CapEx, and (inaudible) revenues in the quarter. Maybe you could provide more color what is behind this and what you expect going forward?

Vladimir, I'll take the CapEx question first just because it's a really simple one. That's a timing question. So we're sticking to, sort of, our initial guidance that we kind of set out at the beginning of the year, which is we expect low to mid-double digits, so call it, 12% to 15% in terms of CapEx to sales. And we are not expecting any changes from that. We've made some optimizations inside of that and kind of realign where we're spending that CapEx during the year. But overall, same as it was, and so you should expect it to come down as a percentage in the second half of the year.

On the competitive front, look, I'd say competition is not new. Russia is a very competitive market. It's one of -- as far as I know, one of the only 2 really open markets where companies like Google have real competition, right? Google is not limited in any way, shape or form from participating in this market. And despite of that, our market share on Android, which is Google’s own operating system, is almost 60% compared to Google's 40%. So whether it's DiDi or Spotify, I think we've always operated an environment of competition. We take it very seriously. And our focus is on just develop the best products we can and provide value to consumers. And if you take somebody like Spotify, I think that's not -- it's not BS, it's real value that we drive to consumers. So if the consumer is buying the same subscription to Spotify or to Yandex. Plus, the value that he's going to get out of subscription for Yandex. Plus is considerably higher. We have incredible content assembled in KinoPoisk. We have the best ride-hailing service in the country on which you get significant discounts, and the list goes on and on. So you're sort of comparing apples and oranges. And beyond that, I think I'm pretty sure we're the leading player in streaming music in Russia. We have great recommendations, great play lists, lots of AI that's used there, and people really do like the service.

So on the question of DiDi, I think Yevgeny kind of addressed that really well as well. So look, I think it's great that Russia is a very open and very competitive market. I think it drives us to build better products and deliver more value to our consumers.

The next question comes from the line of Ildar Davletshin from Wood & Company.

So I'd like to follow up on a couple of points. One on this Search and Portal business and your margin, which seem to be pretty good despite challenging macro environment. So I'm wondering how much of it was due to your redistribution of expenses, maybe some kind of one-off cost savings such as reduced compensation for top management, which you introduced earlier this year? How much of it did you benefit in the second quarter? Or how much it's really due to resilience of the business? And going forward into the year-end, would you think about your margin in that segment? Also like video content, I think you invested previously, which was part of your costs as part of your Search and Portal.

And then separately, on food and tech -- FoodTech, sorry. Like profitability, I understand these are still loss-making businesses, but maybe any outlook when this could be -- particularly Lavka or eats achieving profitability? Any milestones?

Ildar, it's Greg. There's quite a lot of questions packed in there. Let me try if I can piece them apart. In terms of what helped Search and Portal margins in Q2, the optimization of TAC that we've undertaken was certainly a big help, and that's going to be something that's going to be with us going forward as well. We also did take a number of targeted measures at optimized cost structure, at slowing down the pace of hiring, at being more aggressive with performance management, at reducing top management salaries and eliminating top management cash bonuses and many other things. I'd say bottom line is we've had quite a number of cost savings already in Q2. Some of them will stick around. Some of them will obviously return over time. For example, we cut back on some advertising and marketing expenses, or office and utilities were obviously lower as offices mostly sat empty and we essentially were able to save on utilities. That's temporary. I think it shows you kind of the steps that we took early on and how we try to manage the business proactively and manage the profitability there.

As far as things like investments in media and content, of course, we're going to be doing more of those. We're really, really happy with how Media Services is doing. We're also investing in content and in Search and Portal as well. And look, those are decisions that we're making because we think that they have positive NPV for the business.

Now let me address your question on FoodTech. So we saw really strong improvements in -- on the restaurant delivery side of the business. And despite that, we actually supported our restaurant partners to a very large extent during the quarter. We spent just under RUB 200 million in Q2 on support measures for our restaurant partners, which we think is very important. It's a 3-sided marketplace, and it's very important to make sure that your restaurant partners are happy and that they're able to survive and that you're supporting them through the cycle. And so I'd say we're much more positive on the food delivery business because of the unit economics that we're seeing there, because of the optimization measures that we were able to implement. For example, we decreased our cost per order by about 10% in June versus March, which I think is pretty significant. We also had a number of implementations which improved the utilization rates for our couriers. And we introduced batching, which also lower our delivery costs. I'd say Lavka, our grocery business, is probably earlier stage. We're quite excited about it, but it's just -- given that it's less than 12 months old, it's too early to judge. We're happy with what we're seeing in terms of, kind of, same-store sales improvements and so on. But that's kind of a longer-term investment. Hopefully, that kind of clears things up a bit.

The next question today comes from the line of Anna Kupriyanova for Gazprombank.

My question will be regarding your other bets and experiments. Maybe you could provide some more color regarding how much is -- of that is online education and cloud path in terms of revenues?

Anna, it's not something that we provide yet. We'll probably provide over the course of the year. but basically, in terms of size, the biggest contributor to that other bets and new initiatives line is drive, Geo, Zen, and after that, it's cloud, and after that, it's education. So that's kind of the rough order, which hopefully gives you some color. And in terms of losses in that segment, they were up slightly on a sequential basis. But based on what we're seeing in drive, for example, we're probably more optimistic about it going to Q3. Especially because most of Q2, Yandex. Drive was essentially shut down, so we had a lot of the rental cost without any revenues to offset them. So...

And maybe you can add some details regarding year-on-year growth rates for Geo, cloud, online education, I mean, if it's like double, triple digit, at least?

Yes. I think we said that education was up like 6x year-over-year, and cloud is up massively as well. So it sounds like -- yes. It's still a small line.

And just a follow-up. If you could provide some details regarding your new building, if you can give any estimates for the cost of building and timing when it starts.

Yes. We're obviously still waiting to get all the necessary permits into place. The expectation is to start construction sometime this fall. I think rather than give you sort of estimates on how much it's going to cost us to build, I'd rather give you the expected NPV of this project, which is massive. We think that by building it, we're going to generate hundreds of millions of dollars of NPV to our shareholders, potentially as much as $0.5 billion versus comparable market rents in Moscow. So we're quite excited about it.

The next question today comes from the line of Anna Kurbatova from Alfa-Bank.

I have also 2 questions. First of all, about your outlook, or better to say that you abstained from giving the outlook for the remainder of the year. So generally, you sound quite confident in terms of, let's say, advertising market or demand recovery and across other business lines. So would be great understand what is counterbalancing your positive observations? What do you see on the negative side? What -- because you point in the press release, macroeconomic developments. So would be great to understand if business is generally going well, where are the risks that do not allow you to give some guidance for the next 2 quarters?

And my second question would be, again on the tax maneuver in the Russian IT industry. So some industry players already say that this is an opportunity for them to save on certain cost items like social tax contribution on the software developers, et cetera. So it would be great hear once again your estimate expectations. Would you be able to leverage positively on this tax regime changes or not?

Anna, so look, I think we're withholding guidance at this point just because it's unclear how the situation will develop. And primarily, I think what's not clear is what will happen as the economy starts to reopen. Will there be a second wave? How significant will it be? Will there be further shutdowns? I would say if the -- if there are no further shutdowns of the economy from the coronavirus pandemic, I'd say our outlook is quite robust. I think the trends that we're seeing in the Search and Portal business, again, we're seeing ex-TAC revenue growth of 5% to 6%, which is I think extremely healthy under the current environment. Obviously, things will turn out quite well. I would also say that if you look at the trends that we're seeing in the Taxi business in July, those are extremely strong. We are growing GMVs, as Yevgeny said, or rides around 25% year-on-year, right? Which is, I think, compares us extremely favorably. And on top of that, obviously, you have all of the revenue coming from Eats and groceries and so on. So assuming that there's no shutdown measures, we should do quite well.

And then on the question of tax, yes, we will have some savings next year. We expect positive effect from the social tax impact. We expect some minor negative effects from the cancellation of 0% VAT for certain activities. Net of it, net-net is we probably think that this should be a tailwind in the order of kind of low single-digit percentage of EBITDA in 2021, which I think is in line as what we said on the previous call.

The next question today comes from the line of Sebastian Patulea from Jefferies.

Sorry, I did not hear before. The line dropped for a few minutes. The first one is about Taxi. Can you please talk a bit about the Moscow market? In particular, Citymobil looks to be quite strong there lately. Is there something that they're doing better than you? Or is it just a result that your corporate Taxi business is not operating fully at the moment? And secondly, you mentioned there are benefits when operating a multi-brand approach in taxi, such as the Yandex and Uber brands. Would you say there are benefits in operating the same brand across verticals, such as Taxi, Yandex. Taxi, delivery and exceed car-sharing with Yandex. Drive versus City Mobile, Delivery Club, DiDi so on and so forth? Or it doesn't matter in this case?

Sebastian, it's Yevgeny. Look, if we look at our overall competitive position in Russia, ride-hailing, we've been pretty stable over the past several month. You asked in Moscow, in particular. So in Moscow, in April, we did have a slight decrease in our share as a result of the tariff mix effect. We had premium classes were restricted. And that's for us, historically, pre-COVID. That's a large share of our total GMV in Moscow. Economy-tariff proportion as a percent of total increased. And actually, after the premium tariff restrictions were withdrawn, it took us a couple of weeks, but we did get back to pre-COVID levels in terms of market position very quickly. And in June, we strengthened our market position even further. So based on our competitive intelligence, our competition, market share returns from high-teens in April to low teens in June, in terms of rides and in terms of GMV, they are even lower.

And Sebastian, on the question of the multi-brand approach, recall that we operate both the Yandex. Taxi brand as well as the Uber brand, and we position them slightly different for our consumers. And that gives us, I think, additional flexibility in terms of how we run that business. But I mean, look, ultimately, the proof is in the pudding. And so if you look at our competitive position, if you look at the ability to sustain strong margins in this business, I think we feel very good about where we are. And I think we're very happy with kind of how this business has been executing. The other thing that we've recently rolled out is we've rolled out a second driver app. One we have -- so we actually have 2 consumer apps and 2 driver apps. There is Yandex. Taxi and there's Uber for consumers, and there is Taximeter and Uber Driver for drivers. And that is a strategy that has also served us well.

The last question today comes from the line of Vladimir Bespalov from VTB Capital.

I have actually two. The first would be on [ex-TAC here]. So far, Yandex has been developing, at least this is my expression, this area is more a social kind of thing. But this factor is definitely ready for disruption. And online education, the penetration is very low. Maybe you could talk a little bit about strategic and business opportunities that you see in this sector and how maybe you're going to take advantage of those opportunities? And the second question is on the cost-per-click trend. There was a pretty big decrease in the second quarter, which implies that the prices were under pressure during the quarter. When you are speaking about the recovery in the third quarter, do you -- is it coming from, like, more paid clicks? Or are the prices increase? Maybe you could provide some color on that.

Sure. Vladimir, on the education initiative, there, we kind of see our mission is twofold. On the one hand, we're focused on providing free education services to students in middle schools, in high schools and so on. And there, it's more about corporate social responsibility, if you will. And I think we take that matter very seriously. And we are extremely grateful to the, sort of, the Russian education system in terms of the students that it's able to graduate is the people that we're hiring, the very strong foundation in data analytics and data science. Beyond that, we also have a focus on paid education initiatives. There, the program called Yandex. Practicum, which has been growing, as I said, in terms of revenue by something like 6x year-over-year, and we are very excited about that service. That one is focused on paid education. it's a combination of online learning with follow-ups with kind of professionals in the field and one-on-one tutoring. And so these are, I would say, real focus areas for us and something that we will continue to invest in over time.

On the CPC question, obviously, this is a result of the pandemic. So what happens is as you obviously have a huge increase in the total volume of queries, which you could sort of see in terms of clicks and in terms of the queries that we report, in Q2, you kind of -- if there's less demand from advertisers for this traffic, and therefore, they end up pulling back and the auction cools down. And that's why, in fact, ROI start going up massively for our advertisers, and that's why they turn to search, and they flock to search to the advantage of other forms of digital advertising. So for example, it sounds like Russian social networks are actually struggling a bit with advertising revenue growth in July, right, with sort of flat revenue growth on social network side, where search is up much more than that. And our Search and Portal business is up 5% to 6%, so you see advertisers redistributing their ad dollars to higher ROI platforms. And so over time, you will see those CPCs start to come back up, and you will start to see clicks slow down a bit. And that's, I think, just kind of normal tendency of the auction.

Thank you very much. No further questions. Please continue.

Thank you very much, everyone, for all your questions today. If you have still any other follow-up questions, please contact the IR team. We wish you a good day and see you next time. Thank you. Bye.

Thank you very much. That does conclude the conference for today. Thank you for participating. You may all disconnect.