Oct 25, 2017
Executives
Katya Zhukova - Investor Relations Arkady Volozh - Chief Executive Officer Alexander Shulgin - Chief Operating Officer Greg Abovsky - Chief Financial Officer Mikhail Parakhin - Chief Technology Officer
Analysts
Cesar Tiron - Bank of America Miriam Adisa - Morgan Stanley Vyacheslav Degtyarev - Goldman Sachs Lloyd Walmsley - Deutsche Bank Ulyana Lenvalskaya - UBS Sveta Sukhanova - Sberbank Vladimir Bespalov - VTB Capital Olga Bystrova - Credit Suisse
Operator
Good day and welcome to the Third Quarter 2017 Financial Results Conference Call. Today’s conference is being recorded.
At this time, I would like to turn the conference over to Katya Zhukova. Please go ahead.
Katya Zhukova
Hello, everyone and welcome to Yandex’s third quarter 2017 earnings call. We distributed our earnings release earlier today.
You can find the copy of the press release on the company’s IR website and on Newswire services. On the call today, we have Alexander Shulgin, our Chief Operating Officer; Greg Abovsky, our Chief Financial Officer; and Mikhail Parakhin, our Chief Technology Officer.
Arkady Volozh, our Chief Executive Officer will be available on the Q&A session. The call will be recorded and the recording will be available on our IR website in a few hours.
As usual, we have prepared a few slides supplementary to the story. Currently, they are available on the IR website.
Now, I will quickly cover the Safe Harbor statement. 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 those discussed in the Risk Factors section of our Annual Report on Form 20-F dated March 21, 2017, 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 this call, we will be referring to non-GAAP financial measures. These non-GAAP financial measures are not prepared in accordance with U.S.
GAAP. A reconciliation of the non-GAAP financial measures to the most directly comparable GAAP measures is provided in the earnings release we issued today.
And now, I am turning the call over to Alexander.
Alexander Shulgin
Thank you, Katya and hello everyone. Thank you for joining our third quarter 2017 earnings call.
In Q3, we delivered solid set of results across our segments, with consolidated revenues up 21% year-over-year. Revenue growth slowed down slightly from Q2 as we now had a full quarter’s impact of winding up our business in Ukraine due to local sanctions.
Strong performance on Yandex owned and operated properties, was driven by FX improvement, sale share gains in September and the stabilizing economic environment. Ad budgets across our key advertising categories continue to grow well.
Financial services and auto demonstrated solid growth at 48% and 33%, respectively. Growth in travel and consumer electronics ad budgets improved compared to Q2 and grew in line with overall revenue.
Now, turning to search share, after launch of Korolyov, our new intelligent search platform, our search share on desktop reached historical highs and in September was hovering slightly above 66%. Our search share on Android was 42.9% in September, primarily as a result of the choice screen, which was rolled out to the existing users of Chrome with the browser version update.
This is an increase of 530 basis points on a year-over-year basis. And importantly, we continue gaining share on Android.
We expect to see additional share gains from Android through year-end as a result of the choice screen and additional marketing activity. However, we think that the pace of gains will moderate as our internal data show that the majority of users have already updated their browser.
Further gains from Android are expected to come from new devices, which will be shipped into Russia between late 2017 and early 2018 and which will be equipped with a new search widget on the home screen. This new search widget will appear on the default home screen in the same manner as the current Google Search widget and will offer the choice screen to users of these devices.
The current impact from the choice screen on our search share on Android is in line with our expectations and we are optimistic about our further ability to gain share on the platform. Though we haven’t seen significant impact of these share gains on our financials in Q3 yet as they started in late August, as of today, each 4 percentage points of share gain on Android brings us approximately 1% of incremental revenue.
On iOS, our search averaged 40% in Q3, down approximately 50 basis points from Q2, as our ability to gain share on that platform is limited. Now let me update you on the performance of our segments.
Revenues of Search and Portal grew 21%, driven by strong growth in search and in our owned and operated properties including Zen. In Q3, we broadened presence of our Zen product to Yandex home page and Yandex Search app, in addition to Yandex Browser and Yandex Launcher, which weren’t available previously.
In September, we will launch an experiment with a standalone Zen app on Android. A few days ago, we launched a new publisher format called Narratives.
Narratives is a set of screens combining text, video, images and GIFs that can be swiped through. Publisher partners of Narratives include major Russian and international media, including GQ and Vogue as well as popular bloggers.
Number of daily users grew 33% compared with 3 months ago and reached 8 million users. Advance revenue almost tripled year-over-year in September and its annual revenue run rate reached RUB2.4 billion.
Now, turning to business units. Yandex.Market revenues decreased 12% year-over-year as we continue to invest into the quality of traffic to our merchants.
To remind you, in Q2 this year, we voluntarily started improving the quality of traffic coming to our merchants from partner websites. We believe this is a necessary step to get the platform well prepared for the future FX development, which we plan to manage in partnership with Sberbank.
In August 2017, we announced a nonbinding term sheet that we signed with Sberbank to form a joint venture based on the Yandex.Market platform, as part of the transaction Sberbank will invest approximately $0.5 billion into Yandex.Market. Sberbank and Yandex will hold 50% each before the allocation of employee equity incentive hold.
We are very excited about the opportunities. We aim to develop a leading e-commerce ecosystem where we will provide full scale and easy-to-use platform to our merchants to help them reach their customers.
We’ll give our users an access to a wide variety of local and international goods through a single entry point for online shopping on Yandex.Market platform which will provide simple and secure payment solutions, easy returns and consumer lending. And we’ll offer fulfillment and delivery services to create the best user experience.
Just recently, we started testing our first fulfillment center operated by a third-party partner. We expect the deal to close in early 2018.
Now turning to Classifieds. In Q3, this business unit demonstrated 58% year-over-year revenue growth.
The increase was primarily driven by 87% growth of revenues from licenses and value-added services. I’m very proud of all innovative work coming from the audit value team as they have an ambitious goal to transform how used cars are bought and sold in Russia.
Now, turning to Yandex.Taxi. In Q3, we continue to expand them across Russia and neighboring countries.
As of today, we are available in 150 cities with population of 100,000 people. Expansion into new regions as well as development of the service in the existing markets allowed us to grow our total number of rides at 360% year-over-year in Q3.
Furthermore, rides in Moscow and St. Petersburg grew faster in Q3 compared with Q2.
On a GAAP basis, Yandex.Taxi revenues grew 96% year-over-year in Q3. This is a substantial acceleration compared with 46% year-over-year growth in Q2.
In Q3, our spend was focused on solidifying our position in the expanded regions leading to negative RUB3.2 billion of EBITDA. We believe that now is the right time to invest aggressively in increasing the geographic coverage and the presence of value proposition to riders and drivers.
With this, I am handing the microphone over to our CTO, Mikhail Parakhin. Mikhail, please go ahead.
Mikhail Parakhin
Thank you, Sasha and hello everyone. I am very excited with the company performance in terms of technological innovations and product launches in Q3.
In late August, we released our new version of the search platform called Korolyov. Korolyov is our deep new natural-based search algorithm which matches the meaning of the creating with the content of pages.
It works with the semantics of queries and learns from users’ search behavior. Every page is converted into semantic vectors during indexing and thus allows the algo to work quickly with every new query.
Like all of our machine learning-based technologies, Korolyov improves with every single query. It also incorporates feedback from Yandex.Toloka, our crowdsourcing platform, which helps properly train our machine learning algorithm.
As a result of this launch, our desktop search share reached an all-time high of slightly north of 66%. On the ad tech front, we continued delivering various advertiser tools to increase the efficiency of ads.
Let me briefly cover a few of them. Polygon is our new hyper local targeting tool in Yandex.Audience.
It allows creating geo segment by selecting a radius on Yandex.Maps. The new format makes it easier to target residents of a certain block, visitors of certain shopping centers or students of a particular university.
Moreover, polygon allows advertisers to set up an automatic upward bid adjustment, typed to searches in a particular area. Advertising on Yandex.Maps got more attractive and informative by automatically combining Yandex.Direct ads with portals, reviews, ratings of the businesses from Yandex business directory.
In addition, Yandex business directory now shows a carousel of photos from a local business, along with price ranges right in the business listings. Burger King was one of the first to test the feature and demonstrated very encouraging results.
They got 35% more interactions with their ad on Yandex.Maps. Extended second app title is a new feature that allows advertisers to better explain their proposition to a potential customer with CGAR up to 5% higher on desktop and up to 10% higher on mobile compared to our regular ads.
On the past earnings calls I spoke about Turbo pages for publishers. In Q3, we started experimenting with Turbo pages in search.
To remind you, Turbo pages are in many ways similar to instant articles. However, we believe that they are even easier to implement and offer publishers monetization from Yandex out of the box.
Consumers love Turbo pages and they tend to load many times faster than the regular web pages and optimized for smaller screens. In Q3, our mobile traffic constituted 38% of total search queries.
Mobile revenues constituted 30% of search revenues. In Q3, we also released Alice, the first conversational assistant for the Russian market.
Alice is built using our expertise in machine learning, deep neural networks, trained for massive data sets, our speech recognition and natural language processing. It has a few distinct features.
Unlike major voice assistants it’s not limited to predefined scenarios but also have a chitchat mode. Alice has its own personality with a sense of humor and an ability to understand incomplete phrases and questions.
Alice benefits from the most accurate Russian language recognition technology. Based on the word error rate measurement Alice demonstrates near-human level of speech recognition accuracy.
Currently, Alice integrates Yandex services such as search, news, weather, music and maps. We have been working on expanding this list to other Yandex services like Taxi and to third-party products and services.
We believe Alice will become a major distinguishing feature of our Yandex Search app. During the first days after the launch, we saw a great increase in the number of Yandex Search app downloads on Android and iOS.
I believe that intelligent voice assistants will increase user search intensity, help user attention and become a single entry point to major internet services going forward. With this, I turn the microphone over to Greg.
Greg Abovsky
Thank you, Mikhail and thank you all for joining our call today. In Q3, we delivered another solid set of results.
Our consolidated revenue grew 21% year-on-year and reached RUB23.4 billion. Online advertising revenues accounted for 93% of total revenues in Q3 and increased 19% year-on-year.
Yandex properties’ revenue grew 22% year-on-year in Q3 and accounted for 70% of total revenues, driven by ad tech improvements, search share gains in September and growth of Yandex Zen revenues, which Sasha and Mikhail highlighted a few minutes ago. Revenues from our ad networks grew 11% year-on-year and constitute 24% percent of total revenues in Q3.
Share of ad network as a percentage of total revenues decreased 80 bps compared to 2Q of 2017. Other revenues grew 83% year-on-year in Q3, driven by Yandex.Taxi revenue, which constitutes the bulk of the other revenues line in consolidated revenues.
Traffic acquisition costs related to the partner advertising network grew 19% year-on-year and increased to 59.4% as a percent of partner ad network revenues in Q3, up 110 basis points compared with the previous quarter. In Q3, we ran a number of experiments with our existing partners aim to test new monetization opportunities on new inventory.
These experiments carry higher revenue-sharing terms. Another reason for TAC growth was the change of our product mix, primarily driven by the growth of in-app advertising and introduction of mobile Turbo pages which carry higher-than-average TAC.
We expect that our partner TAC will increase slightly as a percent of our partner revenue going forward and will be hovering in between 59.5% to 61% in the midterm. We plan to continue developing our partner advertising network to bring incremental revenues to our core business.
Traffic acquisition costs related to the distribution partners increased 16% year-on-year and constituted 6.8% of advertising revenues from Yandex properties. This is 30 basis points lower compared with Q3 of last year and 10 basis points lower compared with the previous quarter.
Total TAC grew 18% year-on-year and constituted 18.8% of total revenues, 55 bps lower than Q3 of 2016 and 30 bps lower compared with Q2. Paid clicks grew 6% year-on-year, while cost per click increased 12% year-on-year.
Turning to our cost structure, total OpEx, excluding TAC and D&A, grew 50% year-on-year. Excluding stock-based comp, expenses increased 53%.
The growth was primarily driven by our investments in advertising and marketing related to Yandex.Taxi and Search. In Q3, our headcount was up 17% compared with September 30 of last year and up 6% from June 30 of this year.
Our new hires contributed to our core business initiatives as well as to the business units and Experiments. In Q3, our personnel costs constituted 21% of total revenues.
Stock-based comp increased 12% year-on-year in Q3 and constituted 3.7% of revenues, down from 4.1% a year ago. D&A expense for the quarter increased 18% year-on-year.
As in Q2 of this year, the growth was driven by our investments in servers and data centers in 2016 and 2017. Our consolidated adjusted EBITDA decreased 17% year-on-year and our consolidated adjusted EBITDA margin was 24.3%, down 11.4 percentage points compared to a year ago and down 8.3 percentage points compared to Q2 of 2017.
The decrease in adjusted EBITDA margin was mainly resulted from investments in advertising and marketing in Taxi and Search. Yandex.Taxi remained one of the main areas of investments among our business units.
Excluding both revenues and losses of Yandex.Taxi from our consolidated results, the consolidated adjusted EBITDA margin would have been 1,500 basis points higher. This quarter, the impact from ForEx was a loss of RUB0.6 billion related to the appreciation of the Russian ruble during Q3 2017 from RUB59 to the dollar on June 30 to RUB58 to the dollar on September 30.
Net income was down 65% primarily due to increases in SG&A as well as a foreign exchange loss in Q3 of 2017. Net income margin was 3.6%.
Adjusted net income margin was 10.1% compared with 19.7% in the previous year. Our CapEx was RUB2.1 billion or 9% of our total Q3 revenues.
9 month CapEx-to-sales ratio was 14.8%. We continue to expect our full year CapEx to be in the mid-teens as a percentage of revenues.
Turning to the performance of our business units. Search and Portal revenues grew 21% driven by strong growth in search as well as growth in our owned and operated websites.
Revenue growth, rates of Search and Portal slowed down slightly from Q2 as we now had a full quarter impact of the loss of the Ukrainian business due to the sanctions in Ukraine. Adjusted EBITDA of Search Portal grew 19% in Q3 and its adjusted EBITDA margin reached 42.3%.
This is 60 basis points lower compared with Q3 of last year and 290 basis points down sequentially. This decrease is driven by our advertising and marketing campaign aimed at growing our market share on Android.
Revenues of Yandex.Market were down 12% year-over-year. Adjusted EBITDA margin of Yandex.Market was 21% in Q3.
Turning to Yandex.Taxi, revenues of Yandex.Taxi were up 96% year-on-year in Q3. On a gross basis, before subtracting marketing costs and minimum fare guarantees from our commission revenues, Yandex.Taxi’s commissions revenue grew 212% in Q3 compared to a year ago.
We continued the rapid geographical expansion, rides grew 360% year-over-year in Q3. As a result of our investments, adjusted EBITDA at Yandex.Taxi in Q3 was negative RUB3.2 billion.
Figure also includes accruals for professional services related to the Yandex.Taxi and Uber transaction. Revenues of Classifieds business grew 58% year-on-year in Q3 primarily driven by revenues from listing fees and IVAS, which increased 87% year-on-year.
Adjusted EBITDA margin of Classifieds was 9%. Revenues of Experiments represented by Media Services, Yandex.Türkiye, YGF and Discovery grew 96% year-on-year.
The growth was mainly driven by Media Services, primarily by revenue growth of Yandex.Music. The magnitude of investments in this segment continued to rise across all of the constituents of our Experiments line.
Getting back to corporate matters, during Q3, there was no repurchases of convertible debt notes. As of today we have approximately RUB321 million of the bonds that remain outstanding.
We ended the quarter with approximately RUB66.6 billion in cash and equivalents, which is approximately $1.15 billion of the exchange rate as of September 30. Now turning to guidance.
Based on the solid 9 months of 2017, we are raising our revenue growth outlook for the calendar year. We currently expect our ruble-based revenue to grow in the range of 22% to 23% for the full year 2017.
Our guidance includes negative impact from the loss of our Ukrainian revenue due to sanctions imposed on us in Ukraine and the potential positive impact from share gains in Android. And with this, let me turn the call over to the operator for the Q&A session.
Operator
Thank you. [Operator Instructions] We will now take our next question from Cesar Tiron from Bank of America.
Please go ahead.
Cesar Tiron
Yes, hi, everyone. Thanks for the opportunity.
I have a couple of questions. The first one, can you please explain why the revenue growth on Yandex properties is about 2x higher than on the advertising network?
Then moving on to Android, you gave us the revenue implications for the market share gains. But can you please also discuss the implications for TAC going forward?
And then finally, on the taxi losses, you mentioned that this was mainly due to expansion. Can you please say if most of the incremental spend between 2Q and 3Q relates to advertising or if it’s mainly right subsidies?
Thank you very much.
Mikhail Parakhin
Hello. This is Mikhail Parakhin.
I guess I will take the first one on growth of the network versus our own properties. That is true our properties actually grew much faster this – in this last quarter than external network.
It’s mainly driven by very successful uptake of Zen, as you know, as well as improvements in our image search monetization and other things that we were describing previously, the services that we grew. Overall, the time spent and our share of our properties on the web increased, and correspondingly, the growth, money wise, was also higher on our properties vis-à-vis the rest of the web.
On distribution TAC, yes, the – I think – so we have partner TAC and distribution TAC. On distribution TAC, we...
Alexander Shulgin
Let me take one. This is Alexander speaking.
So on distribution TAC, our revenue share and rates with desktop partners are sorted and there are no substantial changes there. On mobile, we have multiple search entry points on devices.
And this search entry points are driven by our advertising activities where we promote our Yandex Browser and Yandex Search application and a number of other search entry points that we have. We also do deals with OEMs, so we preinstall our apps on the manufacturing lines when platforms get produced.
And as part of the settlement with FAS, Yandex Search is part of the choice screen which is displayed on the Android devices. So to a big extent, impact on distribution TAC depends on the combination of those search entry points.
So for some of the search entry points, we do only advertising. They do not impact distribution TAC completely, so zero TAC.
And for some other search entry points we share our revenues with the partners. Given that there are many moving parts, I’m pretty sure that we will be able to manage the distribution TAC as a percentage of revenues on Android so that it remains fairly stable and similar to the rate that we used to have in the previous quarters.
On iOS, there should be no distribution TAC because there is no one to pay for us.
Greg Abovsky
And Cesar, on the taxi subsidies front in Q3, most of the subsidies were spent on aggressive regional expansion where, as you know, we are covering many, many more cities now than we have before, whereas the magnitude of subsidies in larger and more established cities has been more controlled.
Cesar Tiron
Okay, thank you so much.
Operator
We will now take our next question from Miriam Adisa from Morgan Stanley. Please go ahead.
Miriam Adisa
Hi, everyone. A few questions for me.
Firstly, on Taxi, just wondering what drove the revenue growth acceleration in the third quarter? Were there any changes to commissions or ride frequency and given the fact that you are still heavily subsidizing.
And then just looking into 2018 provided the deal goes through, just wondering how quickly we should expect to see the losses reduced on the Taxi business next year? And then secondly on the core search EBITDA given the resilience of the EBITDA despite the marketing investments in the quarter, do you still expect core search EBITDA to be flat on incentive margin?
Thank you.
Greg Abovsky
Hey, Miriam, it’s Greg. On revenue growth rates within Taxi, as you recall, the magnitude of GAAP revenues that we book within Taxi is a function of subsidies that we have in particular cities.
So if you have cities where you are – have both commission revenues coming in and you decrease the amount of subsidies, then the GAAP revenue for that city ends up increasing. And so given that most of the investment was in the expansion cities, where the commission revenue tends to be lower that tends to create more EBITDA losses, but doesn’t impact your GAAP revenues as much.
In terms of the sort of the phasing of Taxi investments, look, we continue to believe that this is an incredibly large market and obviously to the extent that we combined with Uber, once the transaction is approved. The way to think about this is you are really trying to take a very, very large offline business and try to bring it online.
And so I think we will continue to expand aggressively in terms of Taxi even after the acquisition closes. However, we also recognize that at some point that this should become a standalone publicly traded company.
And as such, it will transition to – from loss-making position to a profit-making position and so on. Hopefully, that answers your question on Taxi losses.
And then finally on margins for the core business, yes, you are quite right, it does imply that the margins in 2017 in the core search business will be up year-on-year, probably, maybe 100 basis points maybe slightly more as we continue to invest aggressively to grow Android search here, both ahead of the choice window on Chrome as well as to support the shipment of new devices, which, as you know, will feature a choice screen for the search widget on the home screen.
Miriam Adisa
Great. Thank you very much.
Operator
We will now take our next question from Vyacheslav Degtyarev from Goldman Sachs. Please go ahead.
Vyacheslav Degtyarev
Yes, thanks for the presentation. Couple of questions on Taxi.
And so first is, can you share some data on usage following the eliminations of the discounts from September. Is there some data on demand sensitivity that you may share probably some loyal customers that are driving to work daily so have their behavior changed?
And so secondly, also on Taxi, do you think that you already reached a sufficient level in terms of the number of cities or you think that you might, let’s say, double or triple number of cities, let’s say in 2 or 3 years from now? And also probably on profitability, if you are close to profitability in some of the cities except for Moscow and St.
Petersburg if you can share that? Thanks very much.
Greg Abovsky
Sure. So obviously, as you cut back any number of subsidies, you do have slight deceleration in growth rates and we kind of observed that as we experimented with the right levels of subsidies.
However, bottom line is that this is a business that is still growing significantly. You saw the number of rides grew 360% year-over-year in Q3.
And furthermore, in many cities, we actually saw acceleration in the rates of growth from Q2 to Q3. If you look at GMV for example, within Taxi, that rate of growth accelerated from Q2 ‘17 to Q3 ‘17, which is pretty remarkable, even though we kind of changed a little bit how we are doing subsidies.
And somebody who takes Yandex.Taxi to work and from work everyday, I could tell you that my usage hasn’t really changed as a result of changing subsidies, but that’s kind of on the side. I do think that we are building a habit, and I think people realize every single day that the value proposition of ridesharing is infinitely superior to owning your own vehicle.
And so we would like to build that habit more and more. In terms of cities expansion, we cover roughly, I would say, probably 40% to 50% of the population within our footprint.
So, I do think there is still plenty of opportunities left to expand our footprint further.
Vyacheslav Degtyarev
Okay. Thanks very much.
Operator
We will now take our next question from Lloyd Walmsley from Deutsche Bank. Please go ahead.
Lloyd Walmsley
Thank you. Two, if I can.
First, the high end of your full year guidance implies about 23% growth in 4Q, which is a nice acceleration. And I guess looking at last year, you beat your October full year guidance in the end by about 3 points.
So wondering what are some of the key drivers of top line acceleration for 4Q? Maybe you can help us think through how much of that might be Android share gains versus core search monetization improvements versus maybe moderation in contra revenue of Taxi?
Any color you can share there would be helpful. And then second one, looking at Android, you’ve obviously seen some nice market share gains already and you quantified a bit the impact for us.
But wondering if you can just talk to how quickly you see ad budgets reacting to the incremental share. Are mostly your clients on open budgets and migrating pretty quickly to buy that incremental volume, or is that something that happens with a lag such that it could have a bit of momentum as it continues?
Any color you could share there would be great. Thanks.
Greg Abovsky
Hey, Lloyd, it’s Greg I will take the first half. So with respect to acceleration in Q4, I think you actually listed correctly all of the reasons that should lead to slight acceleration growth in Q4.
We do have market share improvements both – obviously our overall market share is up and Android market share is up significantly. So that should drive core search revenues to be up.
We do have a bunch of monetization improvements which Mikhail can address in a little bit more detail. And obviously, there should be an acceleration in terms of GAAP revenues on Taxi.
So I think you hit on all the right points. I will hand it over to Mikhail to talk a little bit about ad budgets as well as talk about monetization improvements a little bit.
Mikhail Parakhin
Yes, it’s Mikhail here. So, that’s true.
We continued this quarter as usual improving our monetization technology. We shipped a few improvements in the targeting.
We shipped a few of our – first version of the templates, a new way to show ad that we will continue running out throughout this year and maybe even early in the next year. So there will be few more launches there.
Overall, of course, share gains started helping maybe in September. And it is true that it does have kind of like positive feedback loop is that overall improvements and share gains helps a little bit to redistributing budgets towards Yandex advertisers budget, towards Yandex we see a little bit of an uplift there and that potentially should percolate into network and to other – our other monetization properties.
Lloyd Walmsley
Thank you, guys.
Operator
We will now take our next question from Ulyana Lenvalskaya from UBS. Please go ahead.
Ulyana Lenvalskaya
Hi, everyone. Thanks for the call.
The first question will be a clarification on Taxi please, if I may. First, could you please provide us with your current estimates of the timing of Uber transaction closure?
And then in terms of EBITDA loss, shall we think about the fourth quarter being kind of similar or higher in terms of EBITDA loss on Taxi?
Greg Abovsky
Hey, Ulyana. First of all, our expectation is that the transaction should be approved in November.
And secondly, on a stand-alone basis, we sort of expect that Q4 losses should be lower than Q3 losses.
Ulyana Lenvalskaya
Okay, great. Thank you.
And secondly on the extensive marketing campaign in Search, can you please quantify the amount of the ad spend on search in the third quarter?
Greg Abovsky
I mean, no. I guess, we can’t really quantify it, but you should assume that outside of Taxi, most of the investments in the Search Portal were for Android share gains.
Ulyana Lenvalskaya
Yes, but I mean, was it the one-off? Is the campaign over by now?
Greg Abovsky
Well, remember, there is two parts of note. First of all is the choice window on Chrome on Android, right, where users have an ability to choose which search engine they would like to use and that’s something that’s we have seen probably 60% or 70% of the consumers actually make that choice at this point and we are kind of past the majority of the adoption for that particular choice window.
If you remember back to our initial call on the Android settlement, there is another choice window which will come for all devices shipping into Russia that are Android which will have a search widget on the home screen, which will ask consumers to make a choice as to which search engine they would like to use, Yandex or Google’s. And so we will support that as well.
Although that, just logically speaking, that campaign is less intense than the current one given that the Chrome choice window covered all devices in the market, while the home search widget will only cover new devices. Right, so there is a replacement cycle to think about.
Ulyana Lenvalskaya
Okay. Search and Portal margin in the fourth quarter, therefore, should be somewhat stronger as compared to third quarter, right?
Greg Abovsky
Well, so what I have said before is that we expect that Search Portal margins on a full year 2017 basis to be up 100 to 150 basis points year-over-year.
Ulyana Lenvalskaya
Yes, thank you.
Operator
We will now take our next question from Sveta Sukhanova from Sberbank. Please go ahead.
Sveta Sukhanova
Hello, everyone. Can you please talk a little bit about Yandex.Market?
Honestly, we were very surprised to see minus 12% in the revenues, especially after Alexander comment on the second quarter conference call, which was end of July that you expect Q3 growth rates in Yandex.Market to be higher than Q3 growth rates to be higher than Q2. So what went through during August and September?
That would be my first question on Yandex.Market. And my second question on Yandex.Market would be your guidance about the deal closure.
Now, you are talking about Q1, I quickly checked your earlier disclosure and you were talking about closing the deal by the end of 2017. Is this decline in the rates, in the revenues, do they somehow affect the potential closure of the deal with Sberbank and both probability and both terms, timing?
Thanks.
Alexander Shulgin
Hi, Sveta, this is Alexander speaking. So in Q3, we continued our initiatives to improve course of traffic that we change relative to our partners, especially from the Yandex.Market network.
And this resulted in a temporary decline, let’s say, of the revenue growth rates. So these activities I think will have an impact on the growth rate of market for the next several months.
On the deal with Sberbank, we are moving on. There are no specific deadlines and no specific time line which we can guarantee that the deal will be close by this particular moment in time.
But I can assure you that it’s moving on at the right pace and we are positive about our closing deal with Sberbank.
Sveta Sukhanova
Okay, very clear about Sberbank, but if I may move back to my first question on, again, I recall your guidance which was given in the end of July, so something really went wrong in August and September. And because in July, you were expecting traffic grow – revenues growth to accelerate.
So what happened in August and September? And do you expect Q4 Yandex.Market revenues to be down, or you expect them to be up or what should we expect from the Yandex.Market in Q4?
So, what happens and what should we expect?
Alexander Shulgin
I wouldn’t say that anything particular happened in August or September. Yandex.Market strategically is a very strong asset and we are simply focusing on improving the quality of the values that we generate all versions as simple as this.
Sveta Sukhanova
Clear, but what should we expect from Q4?
Alexander Shulgin
I am not sure we are in position to give revenue guidance on a particular business unit.
Sveta Sukhanova
Okay, thanks.
Operator
We will now take our next question from Vladimir Bespalov from VTB Capital. Please go ahead.
Vladimir Bespalov
Hello. Thank you for taking my question.
My first question will be probably on the Classifieds business and we see a good improvement both in terms of growth and in terms of profitability for this business. Do you think it’s a kind of sustainable improvement and we are going to see any further profitability gains and like growth acceleration going forward or it’s still in a kind of heavy investment phase and the situation might change in the future?
And the second question that I have is again on using your cash probably were some reports that Yandex is going to build a new headquarters for itself by 2021. So if this is the case, are you going to use your cash on this, or any other projects, again, the issue of dividends?
So what’s the outlook here? Thank you.
Greg Abovsky
Hey, Vladimir, it’s Greg. On Classifieds, look, we feel very good – we feel really well about that business right now.
We do expect that the revenue growth rate should sustain for some time. Just basically at this point, it’s in a position where it has become in many regions the de facto place to sell or buy used cars, which is obviously a very enviable place to be.
We have a bunch of tools that we are providing to buyers and sellers, which allow them to carry out a transaction in a way that is much more secure and they can feel much more confident about. From the point of view of margins, look, the goals that we give to the team are not focused on margins they are focused on market share gains and on revenue growth.
And the EBITDA that then just falling through is essentially when you have revenue growth that kind of exceeds your marketing investment budgets. And we want to give the guys running Classifieds more capital to invest and grow the enterprise value of the Classifieds business.
On the question of headquarters, obviously, I wouldn’t comment on rumors, but it’s an NPV type of investment. And from what I’ve seen, this would be an extremely attractive NPV investment for the company, if we were to make that.
So that sort of leave it up to you as to whether or not we will do it, but obviously it’s many years out currently, it runs out in 2021. And so, well a new headquarter represents a substantial investment, it’s made over many years, and therefore, it’s not something that ends up draining your cash pile, but I do think that it’s a highly accretive investment opportunity for our shareholders, if we were to do it.
Vladimir Bespalov
Also any other plans to use your cash pile?
Greg Abovsky
Nothing to comment on at this point.
Vladimir Bespalov
Okay, thank you very much.
Operator
[Operator Instructions] We will now take our next question from Olga Bystrova from Credit Suisse. Please go ahead.
Olga Bystrova
Hello.
Greg Abovsky
Go ahead, Olga.
Olga Bystrova
Yes, sorry. Thank you very much for the opportunity to ask the question.
I have two questions. One on the mobile, your sort of the traffic share gains in Android are obvious, but how do you view your trends in iOS traffic shares and how does it impact your ability to improve mobile revenue share?
And maybe if you can, as part of that question, suggest how well is iOS monetized relative to Android? You mentioned that for every, sort of, 1 percentage point increase in traffic share in Android you gain 25 basis points in revenues.
So that’s the first question. And the second question on e-commerce, you have elaborated a sort of what happened in the third quarter and what can happen in the next couple of months.
When do you think you will be in a position to talk to the investment community in more detail about the strategy of e-commerce, more specifically about whether you – how you will view capital investments in the business? Whether you will continue to be predominant in the marketplace or will go in more in the logistics, etcetera?
And if you could clarify maybe how will 50-50 JV impact your reporting of that segment once the deal is closed? Thank you very much.
Mikhail Parakhin
Hey, Mikhail, here. I will take the first one.
So obviously, we did not get as much – or rather initial list on iOS, because it was not affected by our settlement with Google. It – of course, it does somewhat adversely affect the monetization opportunity there and it’s true that iOS devices tend to monetize better than Android devices.
Correspondingly, we are focusing there on the culture of the product and in particular in my opening remarks, I mentioned Alice, our assistant that we launched – artificial conversational assistant that we launched that actually led to some increase in adoption of our apps on iOS as well. So, it still remains a headwind and we expect it to remain a headwind for foreseeable future.
For the second one, Alexander will take that.
Alexander Shulgin
Olga, this is Alexander speaking. So on Yandex.Market strategy, the big goals haven’t changed.
The plan is to combine the strong sites of Yandex, it’s an excellent position in online, lots of traffic, lots of users, lots of technologies and expertise in different areas, combined with a big number of customers of Sberbank and thus create a marketplace, a leading e-commerce marketplace in Russia where transactions with merchants happen, get fully paid and delivered, not necessarily by Yandex.Market logistic services but by partners which are deeply integrated with our e-commerce marketplace. So this is the big picture of what we are planning to create.
I think in more detail to view we’ll be ready to present that once we closed the transaction with Sberbank. On reporting, so Yandex and Sberbank will own 50% of the entity before the employee equity incentive comp, which means that each part will have less than 50% stake and will not be controlling the entity.
So this is yet to be decided by the accounting experts, but most likely, Yandex will not be considered in Yandex.Market once the deal with Sberbank is closed.
Olga Bystrova
Okay, thank you very much. Can I just maybe a clarifying question on the more – on the first question on mobile.
So you are suggesting that there is a headwind in terms of market share gain, I don’t want to emphasize that specifically, but how much of a headwind do you think it could be? Maybe if you could give some indications on the breakdown between within the mobile how much iOS contributes to revenues and how much Android currently?
That would be great. And also, can you maybe talk a little bit about how advertisers look at the mobile budgets in general?
So when you present the inventory, do they reallocate the existing desktop inventory or it’s an additional budget that they allocate into your mobile product? Thank you.
Greg Abovsky
Hey, Olga, a quick follow-up on traffic, Android is roughly 70% of the traffic for us with the balance coming from iOS and other legacy operating systems on mobile.
Olga Bystrova
And revenues?
Mikhail Parakhin
Yes, and on revenues, it’s Mikhail again. So, it’s – well, it depends on the situation.
It’s never black and white. On average, it tends to be new money.
So new inventory usually brings new budgets, and that’s what I alluded in one of the previous questions I answered that we do see this positive feedback loop that the more inventory you get, the more money is distributed towards you. And then this way you can buy more inventory, so there is little a bit of this positive impact that we are seeing now.
Olga Bystrova
Okay, thank you very much. And Greg, can I ask just maybe to give me numbers – number breakdown on revenues, not just traffic between Android and iOS, if possible, of course?
Greg Abovsky
It’s roughly two-thirds, one-third in the favor of Android.
Olga Bystrova
Okay, thank you.
Operator
[Operator Instructions] There are no further questions in the queue at this time. So I’d like to turn the call back for any additional or closing remarks.
Katya Zhukova
Thank you very much to all of you to join our call today. You know how to reach out to us, please feel free to e-mail or call our direct IR line.
We are happy to answer your follow-up questions. Thank you and see you in a quarter.
Bye-bye.
Operator
Thank you. That will conclude today’s conference call.
Thank you for your participation. Ladies and gentlemen, you may now disconnect.