百度 (BIDU) 2009 Q3 法說會逐字稿

Hello and thank you for standing by for Baidu's third quarter 2009 earnings conference call.

At this time, all participants are in a listen-only mode.

After management's prepared remarks, there will be a question and answer session.

Today's conference is being recorded.

If you have any objection, you may disconnect at this time.

I would now like to turn the meeting over to your host for today's conference, Victor Tseng of Baidu.

Hello everyone and welcome to Baidu's third quarter 2009 earnings conference call.

Baidu's earnings release was distributed earlier today and you can find a copy on our website, as well as on newswire services.

Today you will hear from Robin Li, Baidu's Chief Executive Officer and Jennifer Li, Baidu's Chief Financial Officer.

After their prepared remarks, Robin and Jennifer will be joined by Peng Ye, our Chief Operating Officer and Haoyu Shen, Baidu's Vice President of Business Operations, to answer your questions.

Before we continue, please note that the discussion today will contain forward-looking statements made under the Safe Harbor provisions of the US Private Securities Litigation Reform Act of 1995.

Baidu does not undertake any obligation to update any forward-looking statement except as required under applicable law.

As a reminder, this conference is being recorded.

In addition, a webcast of this conference call will be available on Baidu's corporate website at ir.baidu.com.

I will now turn the call over to Baidu's CEO, Robin Li.

Hello everyone and thank you for joining us today.

Throughout the economic challenges in the past year, Baidu has consistently kept our focus on execution and improving monetization.

And we are seeing results.

The third quarter enjoyed solid growth in both revenues and net income.

In addition, large customers are performing well and increasingly allocating more of their budgets to our platform.

China's Internet market is rapidly growing and search engine marketing is increasingly important.

And Baidu has been working hard to offer more advanced products and tools for users, customers and partners.

We delivered a solid result this past quarter.

We are in good shape to drive important initiatives in this quarter.

Phoenix Nest was launched six months ago and has been making good progress on migration.

Revenues from Phoenix Nest at the end of Q3 contributed to over 20% of our total revenue.

Among those customers who have adopted Phoenix Nest, the percentage is much higher.

In addition, we have seen key monetization metrics, such as coverage and cost-per-click trending upwards.

We are confident that customer acceptance for Phoenix Nest has reached the critical mass and we are ready to complete the transition in Q4.

Having a single platform will allow our customers to focus their resources and better realize the superior benefits offered by Phoenix Nest.

Likewise, by having a single bidding platform instead of two, we will be able to free up internal resources and more efficiently drive customer and revenue growth, as well as improve monetization.

While we are confident in our decision and our ability to execute the switch to Phoenix Nest, we do anticipate short-term negative revenue impact as some customers may lag behind in ramping up their spending on Phoenix Nest.

In preparation, we have implemented a customer service outreach program to help ensure a smooth transition.

For the duration of the quarter, our internal resources will be fully occupied by the Phoenix Nest migration.

Looking to other developments in the third quarter, in order to better monetize our union traffic, we introduced Baidu Union Marketing, an upgraded platform for placing contextual links on Baidu union sites.

This project was code-named [Baido], which you might have heard of.

The new Union marketing platform is independent of P4P, meaning a customer can run contextual campaigns separately from their paid search campaign.

We also launched a number of tools recently to assist our customers and union members to more effectively take advantage of the online marketing products we offer.

Baidu Statistics is an advanced tool which allows customers to track and analyze site traffic in order to optimize their online marketing campaigns and increase ROI.

And Site Manager is a tool for partners to manage their inventory on their sites.

Our intent is to create a more efficient online marketing ecosystem for our customers and partners, which Baidu will also benefit from over time.

Moving to the user front, Project Aladdin, an ongoing initiative to mine content from the hidden Web, is also moving along.

The Aladdin Open Platform has received thousands of applications from webmasters since it was launched.

And we have established information exchange interfaces with many of them.

Aladdin is part of a larger initiative, box computing, which will increasingly be the backbone of future Baidu initiatives.

The box computing concept is based on our belief that as technology becomes increasingly complex, the user interface will move toward simplification.

A person types a command into a box, similar to Baidu's search box, on any device screen and the computer will process the command and take an appropriate action.

The action could be to return a piece of information, run a software application or connect the person to online services directly.

In order to achieve the vision of box computing at Baidu, we have been working with third parties to make content and applications accessible through our search box.

Mobile search is also an area that is increasingly important to us.

And we are working to extend Baidu's leadership there.

You may have heard that we recently announced a partnership with China Unicom, which follows our partnership with China Telecom, formed in May.

More importantly, we are working hard to develop high-quality mobile search products.

Using our proprietary transcoder technology and Baidu Palm, which is Baidu's integrated version of mobile search services, mobile phone users can easily access and enjoy the user-friendly functions of our PC-based services.

We also launched a mobile search service in Japan where 80% of Internet users also log on wirelessly.

We will continue pushing forward innovations and strategic partnerships on this front.

Today, more than ever, Baidu is the number one search engine in China.

And we are seeing progress on all fronts, technology, user experience, customer offerings and branding.

But we believe there is still room for improvement.

Going forward, we will continue to enhance search user experience, customer ROI, as well as monetization.

We will continue to attract talented people.

We have tremendous opportunities in our market and we want to be well-positioned to seize them, through innovation and execution.

Lastly, in anticipation of our tenth year anniversary, we will be moving into our newly completed headquarters in Q4.

We welcome your visit.

And now let me turn the call over to Jennifer for financial highlights.

Thank you, Robin.

Hello, everyone.

During the third quarter, we remained focused on maintaining strong growth and improving profitability through efficiency management and economies-of-scale.

We'll continue with our balanced approach to controlling costs, while investing in important initiatives and ongoing improvements to our offerings.

Specifically, for the fourth quarter, we'll be allocating marketing budget for the celebration of our tenth anniversary, as well as moving into the new Baidu campus.

In addition, expenses associated with new hires in R&D and sales, which we acquired towards the end of Q3, will be reflected in Q4, along with expenses for new hires we plan to make during the fourth quarter.

Now let's take a look at some of the financial highlights for the quarter.

The amounts mentioned are in CNY, unless otherwise noted.

For the third quarter, total revenue were CNY1.279m -- CNY1.279b, within guidance and represents a 39% increase year-over-year.

Diluted EPS for the third quarter was CNY14.14 or $2.07.

This includes two non-recurring benefits in other income and income tax expenses, amounting to CNY29.6m.

Without these, the diluted EPS for the quarter would be CNY13.29, or $1.95.

During the quarter, Baidu had approximately 216,000 active online marketing customers, a 11% increase from the corresponding period in 208 and a 6% increase from the previous quarter.

Revenue per online marketing customer for the third quarter was approximately CNY5,900, a 26% increase from the corresponding period in 2008 and a 9% increase from the previous quarter.

Traffic acquisition cost as a component of cost of revenue, was CNY196m, or 15% of total revenue, as compared to 12% in the corresponding period in 2008 and 16% in the second quarter of 2009.

The slight sequential decrease over Q2 reflects normal market fluctuations.

Bandwidth cost and depreciation cost, as a percent of revenue, both continued to decrease in the third quarter, on a year-over-year basis.

This reflects efficiency improvements, as well as increased scalability of investment in capital expenditure.

We will continue to invest to ensure we have an efficient infrastructure and sufficient capacity to serve the growing and evolving Internet development.

Selling, general and administrative expenses in Q3 was CNY198m, a increase of 21% year-over-year.

With the ongoing rollout of Phoenix Nest and overall sales expansion, we expect to continue increasing sales headcount.

R&D expenses was CNY117m, a increase of 49% over the year-ago period, primarily due to increased headcount.

Operating profit for the third quarter was CNY521m, a increase of 42% over Q3 2008, primarily driven by good management of expenses and platform scalability.

Total headcount as of September 30, 2009 was about 7,000.

The increase in headcount mainly comes from sales and marketing, and R&D to support our many initiatives, including Phoenix Nest and Aladdin.

We expect to continue to increase sales and R&D headcount.

In other income, we received a government grant of CNY14.3m in this quarter.

Income tax expense was CNY49m for the third quarter.

The effective tax rate for the third quarter was 9.1%, the same with the corresponding period in 2008, and compared to 11.8% in the previous quarter.

The sequential decline in tax expense is due to a received tax benefit in this quarter, in the amount of CNY15.3m for the year 2008.

We were rewarded this refund because we obtained a preferential income tax status for one of our subsidiaries.

Net income was CNY493m, a 42% increase from the corresponding period in 2008.

Basic and diluted EPS for the third quarter of '09 amounted to CNY14.23 and CNY14.14, respectively.

Net income, excluding share-based compensation expenses, a non-GAAP measure, was CNY514m, a 41% increase year-over-year.

Basic and diluted EPS, including share-based -- excluding share-based compensation expenses, both non-GAAP measures, were CNY14.84 and CNY14.75, respectively.

As of September 30, 2009, the Company had cash, cash equivalent and short-term investment of CNY4b.

Net operating cash flow and capital expenditure for the third quarter of '09 were CNY734m and CNY130m, respectively.

Now let me provide you our top-line guidance for the fourth quarter 2009.

As Robin noted, we believe that our customer acceptance for Phoenix Nest has reached a critical mass and we are ready to complete the transition to Phoenix Nest in Q4.

Please note that our guidance already takes into consideration the anticipated temporary impact of the switch.

We currently expect total revenue for the fourth quarter of '09 to be between CNY1.19b to CNY1.23b, which would represent 32% to 36% year-over-year growth.

I do wish to emphasize that this forecast reflects Baidu's current and preliminary view, which is subject to change.

I would now open the call to questions.

Operator, please go ahead.

The question and answer session of this conference call will start in a moment.

(Operator Instructions).

Our first question will come from the line of James Mitchell with Goldman Sachs.

Please proceed.

Thank you.

Could you discuss why you decided to accelerate the full conversion from Baidu Classic to Baidu Professional, to complete it during the fourth quarter rather than complete it next year?

Hi James.

Yes that's -- I'll split down the answer into two parts.

One is the revenue part.

As Robin mentioned in his prepared remarks we have -- since about six months ago, we launched the Phoenix Nest [openly] to our customers.

We are seeing the key monetization metrics all going the right direction.

The coverage -- [extra] coverage provided by Phoenix Nest is definitely increasing and the CPC, cost-per-click on the clicks brought by Phoenix Nest has also been increasing steadily since six months ago.

So our monetization metrics, we're happy with where we are.

And on the operation metrics, which Robin also alluded to in his remarks, we're seeing -- on a monthly basis, we're seeing over 70% of our customers are already using Phoenix Nest.

And overall, at the end of Q3, on a revenue basis, over 20% of our revenue is already coming from Phoenix Nest.

So on those operation metrics and monetization metrics, we're happy with where we are.

We feel we're ready.

Then I guess, right now -- that's another part of the question.

Running two systems puts a lot of pressure and stress on management capacity and on our customers, and also on our service staff.

So that's one thing.

The other thing is it also puts some stress on our system.

Running two auctions is basically sub-optimal because these two systems are artificially separated and segmented.

So it will be hard for Phoenix Nest, over time, to reach its full potential, with running two systems.

So I think if we move now, we can free up a lot of resources on the management side and the customer service side, and also on the customer side, to focus on this new superior system.

And also, for our algorithm, it can improve faster so that we have the chance to reach the full potential of Phoenix Nest faster.

And we've been saying all along that this is a superior system.

And in the past six months, it has proved itself.

So given all the above reasons, I think we'll move -- we're ready to move now.

And if it's a superior system and the monetization is improving, where does the temporary hit to revenue come from?

Is it because CPCs are lower on Baidu Professional than Baidu Classic?

Or is it because you expect some customers to just stop spending?

The monetization improves on a few metrics.

So on coverage, it gives us more coverage, meaning we -- the system can match more relevant paid links than we can before, so coverage will go up.

So naturally, the click-through rates will go up if you look at the total impressions.

And also, eventually we think CPC will also go up just because the system -- the auction system encourages customers to tell the truth so that they can bid truthfully.

Eventually it will be better for them to measure their ROI and -- which will lead to more spending from them on Phoenix Nest.

Also because of the transition, we need to prepare for a lot of things, especially for the customer service team.

They basically started doing improvements in the Classic edition, which we expect would affect the revenue in a negative way.

Thank you.

Our next question comes from the line of Gene Munster with Piper Jaffray.

You may proceed.

Good morning.

If we could revisit, in terms of the impact from the shift over to Phoenix Nest, if we weren't going to do that transition all the way over this quarter, do you know what the impact would have been or what you would have guided to for December, if that was the case?

And I've got one follow-up question.

Roughly, we are seeing -- we are looking at roughly about 10% negative impact on a quarterly basis because of the switch, as Robin mentioned.

So there are two parts.

Leading into December 1, our customer service staff will be totally focused on working with the moving customers.

And also, at the time of the switch, we would expect that some of the customers may lag behind.

Either they are not ready to move immediately or even for some who move, because it's a new system, they might scale back their spending with us.

So in total, for this quarter, we think roughly the impact is about 10% -- 10 percentage points.

Okay.

When is the actual -- when is the switch [swept] over to Phoenix Nest?

Is it -- when will it be deployed?

Is it going to be gradual throughout the quarter or is it going to be kind of a one-day event?

It practically will be gradual over the quarter, leading to December 1.

We've noticed all of our customers today.

I think a lot of them will hear about this -- we noticed our staff yesterday and most of our customers will hear from our customer service staff today.

And we'll start the moving today.

So practically, if you move today, you will have Phoenix Nest spending tomorrow.

But December 1 is the clear cutoff date.

From December 1, the old system will not be available.

Okay, great.

So I guess in that case, if we have a full quarter impact on the new system in Q1, should we take that into consideration when we do our March estimates as well?

I think so.

Great, thank you.

Our next question will come from the line of Dick Wei with JP Morgan.

Please proceed.

Hi.

Thanks for taking my question.

My question is that for the Phoenix Nest (inaudible), especially the location, the display location is concerned, is it going to be different from the current -- the top panel and the right-hand side panel or it will also be on the left-hand main panel as well?

If you can discuss that, that would be great.

Right.

We have said all along Dick that Phoenix Nest is a new auction system.

It's a new mechanism and it's independent of where [bids] are displayed.

So this switch that we are talking about means Phoenix Nest as the auction system, as the bidding system, as the keyword management system, will take over all the inventory we have on our search result page, meaning the main panel will also be powered, if you will, by Phoenix Nest.

Great, thank you.

Can we have the next question, please?

Sure.

Our next question comes from the line of Catherine Leung with Citigroup.

You may proceed.

Hi.

Good morning.

My question is also on Phoenix Nest.

Given this has been launched for six months, I assume that the expected negative impact is at least in part based on how the existing customers' trend -- spends have trended when they switch over to this system.

Can you elaborate a little bit more on -- is the initial impact due to the fewer clicks or customers withholding their budget intentionally?

And is there much impact due to the lower click-through -- sorry, to the lower cost-per-click because there is a smaller group of keywords that are available for sale?

Or are the customers underbidding because this is kind of a [blind], kind of closed system?

And can you also elaborate on how these customers trend -- using Phoenix Nest have trended over the past few months since launch?

So in other words, have you seen the recovery in their spend toward the later part of these six months?

How fast has the spend recovered?

Thank you.

That's a lot of questions (inaudible) together.

But I guess it's just -- well, I'll give you an overview of what we're seeing in the past six months.

So we mentioned that on a revenue basis, at the end of Q2, 20% of revenue is coming from Phoenix Nest.

So that percentage has been steadily going up since day one.

And the other metric, which is 70% are on a monthly basis, already using Phoenix Nest.

That's definitely a metric that keeps going up.

And if you look at these two metrics together and then you ask yourself, okay, so what's the percentage of spending from Phoenix Nest accounts for those customers who are already using Phoenix Nest, that percentage, as you can imagine, is much higher.

As far as clicks and CPC, because not all the customers are bidding on Phoenix Nest yet, and right now the main panel takes priority, the CPC, in a sort of general sense, is still lower than the Classic system.

But that's for obvious reasons.

But again, we are seeing that metric, the CPC, going up pretty healthily in the past six months.

And also, the -- in terms of number of clicks, that's growing very fast as well.

The CPC as well -- because the (technical difficulty) gives us better capability in matching queries with relevant paid links, CPC is also going up at a very healthy rate.

Let me just add a few words.

Regarding to the number of keywords available for sale, right now they are separated from Classic Edition and the Professional Edition.

After the flip, there -- the keywords available for sale should be unified.

And the number of keywords available should not be less.

It should actually be more.

Thank you.

Can I ask a quick follow-up?

Given as you mentioned, the inherent monetization potential of the right-hand side of the page is less than the main -- left-hand side of the results page, how meaningful do you see -- just to read through on your current six months' experience with Phoenix Nest that was launched in a portion of your results page?

How meaningful is that when -- translating over the expected impact for the main results page portion?

Thank you.

Contribution from the right-hand side is still a small percentage of our total revenue.

We expect this will probably continue to be the case for a while, although we're working very hard on the right-hand side panel to make sure that we serve very relevant paid search [events] so that it can give us good user experience and good monetization.

And we're definitely making progress on that.

If you -- and as you noticed in the past six months, if you check out our right-hand side panel, it's definitely -- the appealing quality is definitely improving.

On the other hand, when we talk about Phoenix Nest, when we talk about a 20% revenue contribution, we're really talking about for the -- the majority of that is coming from the left-top panel of our search result page, which is managed right now by Phoenix Nest.

I see.

Thank you.

Our next question comes from the line of Jeffrey Lindsay with Sanford Bernstein.

Please proceed.

Hello.

Good morning.

Can I ask two questions?

First of all, will your move to greater automation using the Phoenix Nest system, give you an opportunity to reduce cost or headcount?

Or will you just be absorbing your additional staff by not hiring at the same rate?

And then my second question is I noticed that there was a significant increase in traffic acquisition costs for Baidu Union.

Does that mean partner deals are becoming more competitive to acquire or are there other factors such as a shift in mix?

The transition -- complete transition to the Phoenix Nest will not increase -- directly increase our sales force because we have noticed that with more customers using our platform, they also need to have better services.

As you have noticed, the last quarter, the ARPU has increased a lot and that means the more and more customers can benefit the value of our platform.

The question on whether the automation (inaudible) by the Phoenix Next has the potential of giving us the chance to reduce headcount, I think that's conceivable.

But that's definitely not the main objective of Phoenix Nest.

And Phoenix Nest is more -- I would say a more sophisticated system.

So on one hand, it is providing more tools and on the other hand, it probably needs customer service people to have -- to work more intensely -- intensively with customers to optimize their campaign to really get the benefit of Phoenix Nest.

And the question on TAC, I think you are referring to the year-over-year growth of TAC percentages.

Actually, if you look on a sequential basis, it's actually coming down a little bit.

There is no question that the [Union] business is quite -- especially on the search side it's quite a competitive business.

And the -- but if you look at year-over-year basis, the increase has been driven by and large by the faster growth of our Union revenue than our organic revenue.

Our payout ratio, although as I said it's a competitive business, our payout ratio has been quite steady and in the recent quarter, it actually came down a little bit.

And that explains the sequential decrease of TAC ratio from Q2 to Q3 because we implemented the new Union TAC payout policy at the end of Q2.

So a lot of [added] TAC is flowing through our P&L in Q3.

That's great.

Thank you.

Our next question comes from the line of Ming Zhao with SIG, you may proceed.

Okay, thank you for taking my question.

Good morning.

My question is still on the Phoenix Nest system.

So when we look at the impact to your revenue, temporarily right now, does that mean that a lot of the efforts will be focused on converting the usage on to the Phoenix Nest system and then what we'll likely see is basically ARPU going -- still going up steadily but you will lose some of the customers when you look at your two metrics there?

And if so, when do we see the customer addition becoming positive again?

It is -- Ming, I think you're probably right to think that the switch will impact both ARPU and the number of customers negatively.

On number of customers, I think that some will not be willing to move as fast.

That will have an impact on number of customers.

But if you look on a quarterly basis, probably not as much, because the switch date is December 1.

So I would imagine most of the customers will keep spending in the first two months.

And on ARPU, for the move -- on ARPU I think after December 1, for the customers who've already moved, some of them might scale back their spending just because they're not familiar with the system yet, either because of skill or because of will, the ARPU might come down a little bit.

So Q4 is definitely hit by Phoenix Nest.

We would probably expect Q1 will be as well.

So I think overall, it'll just probably take a couple of quarters from the date of the switch for the situation to settle down and for us to come back to the normal growth trajectory.

So from other hand, our staff will be fully occupied helping customers to migrate the existing system to Phoenix Nest.

That will certainly take some [benefits] from new customer acquisition.

And if you think about the beauty of the marketplace, the more people bidding for the same thing, you actually have better revenue opportunity for monetization capabilities.

Therefore, during the transition as customers are moving from the Classic edition to the Professional edition, we do expect to do some monetization power because less people are bidding certain keywords [in any] system.

Okay, just a follow up to that.

When you talk to your advertising customers, what their feedbacks are in terms of the return or the results after using the Phoenix Nest system?

In general, customers are happy with the return they're getting from the Phoenix Nest although we don't have scientific proof or -- no customers [brought their] comparisons -- comparisons between Phoenix Nest and Classic system to us.

On the other hand, because these two systems are running in parallel, there's no apples-to-apples comparison anyway.

But I would say in general, they're happy.

They're happy with the ROI.

They're happy with the tools and the data -- the actual tools and data we provide in the new system.

Okay, thank you.

Our next question comes from the line of Richard Ji with Morgan Stanley.

You may proceed with your question.

Hi, Robin, Jennifer, Peng and Haoyu, good morning.

Let me start off with the first question regarding the large customer.

And obviously, more large customers have adopted your search engine for their branding initiative.

And can you give us some color on the -- roughly revenue contribution from these large customer and what are the top advertising categories for these customers?

And more importantly, have they given you any [trends] what is their rough ROI for the search related initiative versus their so-called traditional brand advertising?

Hi, Richard, good morning.

Normally, we don't give the breakdown of the customers.

But as you have noticed, the tier one customers, our big customers grow faster than the small and medium enterprises last quarter.

Especially, in the following categories, they enjoyed much faster growth which are online gaming, e-Commerce, and auto industry as well.

So a couple of our customers allocated much more marketing budget in our platform.

For example, [Shanda], Perfect World, for eCommerce like Amazon China, [E-Trip], eLong and also Ctrip.

Also a couple of big customers joined us using our platform such as Audi, Mercedes Benz as well.

So those segments grow much more faster than other industries.

A couple of reasons for those tier one customers allocating more budget to Baidu, we noticed that under the tough economic situation, those few customers pay much more focus on the much efficient ROI marketing platform.

Also, last quarter, we introduced more marketing tools to them so that they can understand and appreciate much more the benefit of our platform.

That's helpful.

I have a follow up question.

And the ones who have adopted, fully adopted the Phoenix Next system, what kind of impact should we expect for the Baidu Union members and also for your large customers?

For large customers, I think there tends to be -- compared with SMEs anyway they tend to be more sophisticated keyword buyers.

So there's definitely, even until today, we've seen good adoption by these customers.

So I would think that the growth differential between large customers and SME will probably continue after we switch to Phoenix Nest.

As far as Union numbers, that's a good question.

So in -- as Robin mentioned, I think, in his remarks, we are launching this new product we call Baidu Union Marketing which is new -- which is a new contextual product which will gradually replace the old contextual product that we have.

But the old contextual product is very much coupled with old Classic paid search products.

So that will in that sense, because it is coupled with the old system, so in that sense, on the switch date, the revenue from that product will also get impacted materially.

Very helpful.

Thank you.

Our next question comes from the line of Steve Weinstein with Pacific Crest.

Please proceed with your question.

Great, thanks.

I just want to make sure -- actually, if I heard you correctly, I think you indicated that the move to Phoenix Nest was going to cost potentially about 10 points of growth for Q4.

If I do the math though, that would imply a much lower sequential growth this Q4 than you have historically.

So I want to make sure I understand that right and that [may be] correct.

Also, if you think about that transition, do you think the impact will be greater in Q4 or in Q1?

Thanks.

Yes, you heard it correctly.

We did say that the -- on a quarterly -- for the entire Q4, the impact of the switch is roughly 10 percentage points.

So if you do that math --

Is that a year-over-year or a sequential 10%?

It's sequential.

(Multiple speakers).

Okay, that explains it.

Thank you.

Yes, so you heard it correctly.

And as far as how big an impact it is on Q4 and Q1, I think when the switch happens, in December, we'll have a much better idea.

But we do expect in Q1, there will still going to be material impact from the switch.

Okay.

Thanks a lot.

Our next question comes from the line of Eric Wen with MainFirst.

Please proceed with your question.

Good morning, thanks very much for taking my question.

Jennifer, how are you and Robin?

I have -- my first question.

I have two questions.

My first question is if you can give us a quick comment on the advertising market outlook in the fourth quarter and 2009.

I guess somebody just mentioned that if we remove that 10% hit, we actually have a quite good acceleration on year-on-year growth on the advertising revenues or Baidu's revenues in the fourth quarter.

Of course, I understand the fourth quarter '08 is an abnormal quarter.

But just if you can comment on -- if we remove on these two temporary factors, how do you see the overall market picking up going forward in 2010.

Thanks and I have a follow up question.

Thanks.

This is Robin.

Let me give it a try.

I think we are seeing increasingly a separation of search engine marketing and online branded advertising.

What we have seen is that the search engine marketing has become increasingly important and more and more customers, especially the larger customers realize that the performance or the return on investment for a search is much better than any other means of advertising.

So we are very happy and we are very confident that the growth prospect for search engine marketing is very good and will be very good for a long time.

For branded advertising, it has been always been a very, very small percentage of our total revenue and we have not seen much movement or growth in the past twelve months.

Thanks very much for the clarification.

And my follow up question is to Jennifer, is that it looks like the scalability of the cost structure continues to be very good.

Even with the slowing down in revenue growth due to various factors, operating margin has held up and [bandwidth] and depreciation and TAC has all coming down in terms of revenues.

I just wonder -- this has been going on for a while.

I just wonder what is the endpoint that -- how low can we eventually, how scalable is this model eventually is?

I know this is an old question but I just wonder if you have any additional color to shed on that.

Thanks.

Yes.

The good performance on the P&L or margin side is a result of good expense management as well as scalability.

We take a very disciplined approach to our cost and we invest prudently for the future benefit.

As mentioned many times in the past, the model is inherently tremendously scalable.

And so at this point, it's not really our focus to target a certain margin percentage.

We are still at the very early stage of the Internet development.

So we want to invest and spend prudently to make sure we continue to remain at the leading position on the search engine space.

Having said that, you too have noticed that our infrastructure and equipment scalability has [exhibited] for many quarters as well as the bandwidth side.

The team is doing a good job to efficiently utilize our infrastructure and equipment and to optimize the infrastructure, the way we architect -- the overall architecture works.

Going forward, we will continue to make investment in CapEx.

We'll continue to make investment in important initiatives like Aladdin, like Phoenix Nest, and those require people investment.

So our focus is more long-term.

It's not margin management, it's focusing on driving the growth and maintaining a disciplined approach to run an efficient company.

Thanks.

Our next question comes from the line of Wallace Cheung with Credit Suisse.

You may proceed with your question.

Hi, good morning.

Thanks for taking my question.

Can you update us on the top five industry groups?

And also, can you give us some color on these top five groups regarding their [response] on the performance of Phoenix Nest?

Will there be some of them more favorable on Phoenix Nest where some may be less favorable?

Thank you.

The top five sectors continue to be the same.

They are healthcare, machinery, education, franchising and general travel, in that order.

As far as -- I think what we mentioned before was that in general, the B2C sectors growing faster than the B2B sectors and that continues to be true in Q3.

And I think actually, the launch or the switch to Phoenix Nest will continue to sort of push that trend even more just because the B2C --just by nature of the B2C sectors, they will stand to benefit from the benefits offered by Phoenix Nest because they will have more choices on keywords, they will have better tools to manage more keywords because they tend to have more keywords to purchase than the B2B sector.

Just one quick follow up.

So overall, these top five customer groups, are they spending more than 20% or less than 20% of the total revenue on Phoenix Nest?

I don't have the exact numbers but I think these are not -- some of these are B2B and some of these are B2C.

I would guess the B2C sectors -- customers in that sector -- in those sectors are spending more, meaning more than 20% of the revenue on Phoenix Nest in general.

But again, I don't have the exact numbers.

Thank you very much.

Our next question comes from the line of Wendy Huang with RBS.

Please proceed with your question.

Hi.

My first question is regarding the seasonality of Q1.

I remember that historically, Q1 exhibits weak seasonality.

And also, you just mentioned that the Phoenix Nest forecast date is December 1.

So actually, 4Q, we are not going to see the full quarter negative impact.

But in Q1, we will see the full quarter.

So in that case, given the two reasons I just mentioned, should we expect the sequential decline for Q1 2010?

We don't give -- we don't give guidance two quarters down the road.

But as we've said, the Q4 will be --

But what do you think with the two points I just mentioned?

The point being seasonality and Q1, we'll see full quarter impact.

Yes, those are true.

Okay.

And also, could you maybe give some color on the new Baidu Union Marketing [Baido], how that will affect the Union revenue and also that will affect the TAC ratio and how different that is from the historical P4P Union marketing tool?

Right, so the new product we offer is a [contextual] product.

Within Baidu Union, within our affiliate business, we have two types of business.

One is search related.

When they search, we basically -- the partner basically brings traffic and we share paid search revenue with them and the other is contextual.

So historically, contextual has been a smaller -- a much smaller percentage of revenue.

So this new product is an upgrade from the old contextual product that we have existing today which will be unavailable on December 1 together with the old paid search product because they're very much coupled together.

As far as payout, the payout on contextual is in general higher than the payout on search because it's not search.

The traffic, the content, is owned by the third party publishers.

And so as we grow [the business] going forward, because of the mix in our affiliate business might change and the TAC ratio on contextual is higher, you could potentially see a growth of TAC ratio.

But I don't think that's going to happen in the short term because the contextual part will be challenged by the switch and also at the end of the day, it's a very small percentage of the revenue anyway today.

Okay, great.

Thank you.

Our next question comes from the line of Eddie Leung with Merrill Lynch.

Please proceed with your question.

Good morning, guys.

Just a couple of questions.

The first one is actually on the cost side.

I'm wondering if Jennifer can share with us more about the costs related to the new headquarters.

For example, would there be any continuous depreciation cost and if so, what that amount could be.

Thanks.

Hi, Eddie.

The Baidu Campus, on a net-net basis, does not impact the P&L much.

Today, we do rent office spaces and so therefore we pay rent.

And going forward, we move into Baidu Campus and we'll recognize depreciation on the P&L.

So with the carrying cost and the depreciation net-net, it is about the same as we are seeing today.

So not much cost impact.

Got that.

My second question is just a quick question on click through rate.

I'm wondering if you guys can share with us the trend of click through rates in the past quarter and given the revenue changes in back end systems, shall we expect any material change in the click through rate in the upcoming, say, few quarters?

Thanks.

The click through rate on the top panel -- the top left panel and the right panel has been going up because more and more [dealings] in both spaces are actually managed by Phoenix Nest which is a system which is a better system -- [relevant] system.

So that process, that growth has been gradual.

So I think that will probably continue in the next few quarters but I don't think we will see a dramatic increase.

Understood, thank you very much.

Thanks.

Our next question comes from the line of Yu Jin with CICC.

Please proceed with your question.

Hi, good morning.

I have two questions and one question is on the margin side.

Given the impact of the system transition and I have heard that we will increase [the margin] on the marketing activities in the fourth quarter and going forward.

So (inaudible) that the gross margin, the operating margin could possibly turn down in the next couple of quarters, I believe before the revenue can go up benefiting from the new system?

And the second question is rather clear, that I heard that Robin said we teamed with Telecom and Unicom.

But I read from the local news that China Telecom choose another local search partner mInfo to be its strategic partner on the wireless side.

So can you please clarify on that?

Thank you.

On the first question, our margin is very sensitive to revenue.

So we guided for the Q4 reflecting the temporary impact of Phoenix Nest and we will maintain -- continue spending for the long-term benefit.

So margin will just naturally be the way it is.

Going into Q1, normally we do have seasonal quarters and Q1 is probably one of the lower quarters within the year and it has been the case.

So you know, you do see margin variations quarter-over-quarter.

But as I said, we don't focus on margin, we focus on long term, the revenue growth and running an efficient company.

So margin will just naturally be the fallout of that.

This is Robin.

Regarding to the mobile search partnership, I haven't heard about the deal you mentioned with a third company -- between the third company and China Telecom.

What we know is that we had a search partnership with China Telecom.

It was announced in May that we [serve] our search results to the China Telecom users -- mobile users.

As Robin mentioned before, establishing a partnership with operators is one thing.

But developing a very good -- excellent customer service and also the product is the other focus we need to make the effort on.

Okay.

Thank you.

Our final question comes from the line of Stephen Ju with RBC Capital Markets.

Please proceed with your question.

Jennifer, just to clarify in regards to the marketing spend in the fourth quarter you mentioned.

Should we expect this to be up on a CNY basis quarter-on-quarter?

Or are we thinking that the spend you have planned for the new building, anniversary, et cetera, it replaces what you spent on the World Forum during the third quarter?

And CapEx seems to have ticked up a little bit during the third quarter.

What is your outlook for the balance of the year?

Thank you.

We do expect a step up of SG&A expenses for Q4.

So when you look at it, we're basically -- two main things are happening.

One is we are moving to the Baidu Campus.

We take a lot of pride in moving into the new building.

We'll do a celebration.

And it also happens to be Baidu's tenth anniversary so we'll do a lot of funding activities surrounding these main events.

So marketing expenses will be higher than Q3.

And we're also planning to hire -- continue to hire sales and R&D talent going into Q4.

I did mention in the script that a lot of the headcount increases occurred towards the end of the quarter.

So you will see the full expense impact going into Q4 plus some new hires.

So SG&A expenses will have a step up from Q3 levels.

On CapEx, I have mentioned in the past that 2009 should be no less than 2008.

And you do see -- I have also said in the past that the CapEx is pretty chunky.

So Q3, there is a step up and I anticipate Q4 will be a sizable CapEx expenditure as well.

Okay.

Thank you.

We are now approaching the end of the conference call.

I will now turn the call over to Baidu's Chief Executive Officer, Robin Li, for his closing remarks.

Well, I understand many of you have concerns regarding to the switch or flip to Phoenix Nest.

I want to assure you that this is a strategic decision made by the management.

We did careful calculation and we think twelve months from now, when we look back, we will be happy with the switch sooner rather than later.

So we believe -- we truly believe that Phoenix Nest is a superior monetization system and time will tell that we are making right decisions and right moves.

Once again, thank you for joining us today and please do not hesitate to contact us if you have any further question.

Thank you for your participation in today's conference.

This concludes the presentation, you may now disconnect.

Good day.