About this Blog

The purpose of this blog is for my personal use. It serves as my personal diary as I investigate Chinese internet/gaming companies for investment purpose. If you have any comments or disagreement, please give me feedbacks.

Thursday, July 24, 2008

NTES – DT2 and FF2 go into Beta testing

My last post on DT2 can be found here:


My last post on FF2 can be found here:


My last post on TX2 can be found here:


My last post on XYW can be found here:


It is exciting time for NTES. All its major games are going to be out at about the same time.

TX2 went into Closed Beta on 6/6/2008. It is open to everyone and the game characters won't be deleted. Actually, it had already started commercialization already. The game is doing well and is growing steadily. It didn't grow like a block-buster. But that was planned. NTES was not looking for explosive start, but rather slow and steady growth until it reaches PCU of 1 million in 1 year (according to Netease's CEO).

It already started charging for in game items already. More excitingly, starting with a couple of weeks ago, players can now buy treasure boxes. This is similar to GA's ZhengTu game. The treasure boxes are really just online lottery system. It function like online gambling. It is like a slot machine. Every times it opens, items with different values rotates. The probability of getting really good item is very low, but also very tempting. Therefore, it is hard for players to only buy one. They buy one after another until they hit a good item. It may not have very good reputation. But players just can't stop spending money on those.

XYW started Closed Beta on 7/5/2008. In this game, NTES follows the traditional definition of Closed Beta by limiting the number of players. It will delete the game character after the Closed Beta. There is also no commercialization.

DT2 is a new free-to-play game. It went into Final Internal Beta on 7/21/2008. It is not clear whether NTES will kill off DT. It maybe possible that NTES will have both games operating at the same time, with DT2 the free-to-play game and DT the paying game.

Usually, the Internal Beta is limited to players working for NTES. But NTES uses outside players for this Beta. It is limited to about 1000 invited guests. The game characters will of course be deleted after the current beta testing.

Most likely, after the Internal Beta, DT2 will have Closed Beta and Open Beta.

FF2 is a pretty interesting game. It went into Closed Beta on 7/22/2008. It is open to everyone. But the game characters will be deleted. I don't believe it will be commercialized at this point.

I think FF2 will prove to be a very popular game. It will be cater to young players and female players. It might not be able to get a lot of money per players. But eventually, it could be a 5 to 10 million per quarter game.

Now, I am going to make a table on the characteristics of Beta testing on all four games:



Type of Beta

# of players

Delete game Character?




Closed Beta






Closed Beta






Internal Beta






Closed Beta




As can be seen from the above table, it is all over the map. There is no consistency in how NTES determines what is Closed Beta and what is Internal Beta. It is probably up to the chief game designers of each game to decide what he/she wants.

It may be confusing to the investors. But as long as NTES comes out with great games, that is all I care.

With all these major title coming out at about the same time, the next few months is going to determine the future of NTES.

Finally, one can download the official trailer of DT2 from the following address:


Wednesday, July 23, 2008

NTES – CEO predicted PCU of more than 1 million players in 1 year for TX2

My last story on TX2 can be found here:


First, the CEO of NTES thinks that TX2 is doing great. He predicted that it will be the next block-buster and will have a Peak Concurrent Users of more than 1 million one year from now. See the following story:


I always appreciate comments and emails. Some of the comments I had received recently had not been friendly. I was accused that I was way too harsh on TX2’s prospect.

I think there is a misconception. I think TX2 is a great game. In terms of technology and game design, it is probably the single best game in China right now. In terms of playability, it is still weak. But that is purely because of its young age. NTES is busying adding features to it right now. I have no doubt that it will have tremendous amount of playability features in one to two years.

My problem is in NTES’s aversion to marketing.

I think there is some kind of bunker mentality within NTES.

Both XY2 and XYQ had a rough start in their life. NTES had to slowly working on both games before both games caught on 6 to 12 months from the start of commercial operation. More recently, DT1 was heavily promoted, but it was a commercial failure.

XY2 and XYQ happened 4 to 6 years ago. The competition landscapes had completely changed. Whatever works at that time might not work now. Just because DT1 was a failure doesn’t mean NTES shouldn’t do marketing. (I think DT1 just got caught in a time when China’s game environment was changing from paying game to free-to-play game and DT1 is a paying game).

More importantly, all the recent successful games by all other game companies was achieved with the help of heavy marketing campaign.

For example, Sohu has the largest game portal in China, 17173.com. Sohu also has two major games (the existing TL and a new game). Sohu is not just marketing on its own game portal, it also heavily markets its games on other major game portals.

All other major game companies also do heavy marketing campaign for all their games.

Most of these heavily marketed games failed. But on the other hand, in recent years, ALL successful games were promoted heavily.

I think NTES is living in a cocoon and can see only what it wants to see.

As a long term NTES investor, I hope NTES can break out of it. NTES is going to have 4 or 5 major games coming up. Please maximize their chance to success. Don’t kill off their chance of success just because NTES wants to pinch pennies by saving marketing dollars.

Tuesday, July 22, 2008

Sohu 2Q2008 Prediction

My last prediction on 1Q2008 can be found here:

First, let’s see how I did in my last prediction. The following table compares my TLBB prediction on the previous quarter vs. the actual numbers:



My Prediction – ACU QoQ


My Prediction – PCU QoQ

My Prediction – APA QoQ


My Prediction – ARPU QoQ


My Prediction – Revenue QoQ


Actual – ACU QoQ

Actual – PCU QoQ


Actual – APA QoQ


Actual – ARPU QoQ


Actual – Revenue QoQ


The acronyms first -
ACU = Average Concurrent Users
PCU = Peak Concurrent Users
APA – numbers of Active Paying Account
ARPU – Average Revenue per Active Paying Account

From the above table, it is clear that I completely misjudged my prediction. I only have an estimate for the ACU, it is virtually impossible to give reasonable guesses for other parameters.

Now, let’s do the 2Q2008 estimate for TLBB.

The server counts for the TLBB can be found in the following table:

There are 3 months in a quarter. Each month has either 4 or 5 weeks. I only use 2 days in a week. I don’t do server count every weekend. I only put in the number where I have a server count in both quarter. I also pick 3 days of holidays in this calculation.

From the above table, I have a 2Q2008 ACU increase of 4.47% quarter over quarter (QoQ). But I need to take into consideration that the company suspended operation for 3 days. It counts as 2.56% of revenue in a quarter

By taking into consideration of the 3 days off due to Sichuan earthquake, the new ACU increase is only 1.8%

I used Zhengtu of Giant Online as a guide. It has maximum APA = 1.405 million and maximum ARPU = 309.

I am going to have the APA follows my calculated ACU estimate. I am going to predict the 2Q2008 APA = 2% QoQ increase.

Now, for the ARPU, I am going to give an increase of 20% QoQ. 20% is probably a pretty good estimate considering we had an increase of 35% in the 1st quarter.

Therefore, the revenue increase will be 1.02*1.2 = 1.224 or 22.4% increase quarter over quarter.

We can now put my new estimates into the following table:




My Prediction – ACU QoQ



My Prediction – PCU QoQ

My Prediction – APA QoQ



My Prediction – ARPU QoQ



My Prediction – Revenue QoQ



Actual – ACU QoQ

Actual – PCU QoQ


Actual – APA QoQ


Actual – ARPU QoQ


Actual – Revenue QoQ


Given the the 1Q TLBB revenue is 38.9M, 22.4% increase will give it $47.6M.

Now, let’s figure the revenue contribution from Blade Online. The following table is the revenue from Sohu’s other game, Blade Online





Blade Revenue





From the my server count, Blade Online seems to be increasing during the 2nd quarter. I didn’t bother to do a detailed calculation. I am going to give a rough estimate of 2.5M for the 2Q2008. Take off 2.56% (due to the Sichuan earthquake) gives me 2.4M.

Combine the two, we have the total game revenue of 47.6+2.4 = 50M for 1Q2008. This is considerable more than Sohu’s gaming revenue guidance of 43 to 45 millions.

For wireless, it was 9.1M for the 1Q2008. There were no adverse regulations from China during this quarter. China’s telecom companies are busying with merger and 3G technology. They don’t have time to mess with the wireless value added providers. Therefore, I predict 10M for the 2Q2008.

For brand ads, the 1Q2008 revenue is 33.2M (and 26.6M for 2Q2007). Sohu had a very stable year over year growth of 40 to 45% for the last 5 quarters. Most people think there will be a spike (or extreme acceleration) in brand ads as we get close to the Olympics. I never think so. I believe we will still see a YoY growth of 45% during the 2Q2008. Thus, I predict we will have 26.6*1.45=38.57M. This is going to be in the very low range of Sohu’s guidance.

Search, however, is going to be another story. Sohu had started to do some major push in monetizing search. But the push starts at the end of 2nd quarter, so we will not see it reflected too much during the 2Q. Still, I predict we will see the true turnaround in Sohu’s search business. I think 2Q will be the first quarter where it would experience both QoQ and YoY increase. I predicted 1.8M for search in 2Q2008.

If we add them up together, we would get
50+10+38.57+1.8 = 100.37M

This is better than both the Sohu’s estimate of 93M to 96M and analysts’ estimate of 96.12M.

Monday, July 21, 2008

Sohu – What was Sohu trying to tell us

My last post on the Olympics can be found here:

On 6/19/2008, Sohu’s Co-President Belinda Wang, said she expects core advertising to grow by 40% to 45% this year and 20% to 30% next year. See the following article:

The market interpreted as company’s admission of Sohu slowing down after the Olympics. It punished the stock on the same day and push the stock of Sohu down by 14% in a single day.

But what is Sohu actually trying to say to the wall street.

On 6/26/2008, Sohu published an interview with the CEO of Sohu. The article is as follows:

The main points of that article are as follows:
- Wall Street “mis-interpreted” the interview with Belinda Wang.
- The growth rate of 20% to 30% in 2009 is incorrect. The actual 2009 growth rate is unclear at this point.
- During the Olympics, Sohu will get a bigger percentage of internet users. This pattern will continue into 2009. There is absolutely no reason that the growth rate in 2009 is only 20% to 30%.
- Wall Street still don’t understand Sohu. For example, Wall Street didn’t taken into consideration of Sohu future killer product such as Sogou Pinyin Input Method.

What do I think of what exactly happened?

I think there is a basic misunderstanding between the Wall Street and Sohu management. The most likely case is that the Sohu management was trying to send the message to Wall Street that its estimate for 2009 (after the Olympics) is too pessimistic. On 6/18/2008, the average analysts’ estimate for 2009 is 22%.

Sohu’s management wants to up that estimate while still stays conservative. Therefore, it gave an estimate of 20% to 30%. Thus, its medium is 25%.

But somehow, the “whisper number” in Wall Street is probably higher and Sohu’s stock crashed 14% as a result.

It think it caught Sohu’s management by surprise. Thus the second interview on 6/26/2008. It is pretty clear that Sohu’s management is pretty confident that they can beat the 20% to 30% estimate they gave. But it is so far away that they don’t want to give out more precise estimates.

On a side note, it is pretty clear that Sohu regards the Sogou Pinyin Input Method as its ace, or its most important weapon to gain market shares. There are some recent news about it and I will talk about it in future posts.

Friday, July 18, 2008

iResearch’s 2008 2nd quarter China’s search market

My last post on the iResearch survey can be found here:


But that survey is not on China’s search market. Yesterday, it published a survey on the 2nd quarter 2008 China’s search market. It can be found here:


Some basic result:

- China’s search market in 2Q2008 is $183 million dollars. It represents a 35.8% quarter over quarter growth and 96.7% growth year over year.

- Baidu has 62.8% of market

- Google has 26.2% of market

- Yahoo has 7.9% of market

- Sogou has 0.9% of market

- Zhongsou has 0.8% of market

- Sina has 0.3% of market

- Netease has 0.3% of market

It is interesting that a survey would be published right before all the Chinese companies would announce its 2nd quarter earning. Either this survey is right or wrong. If it prove to be wrong from the upcoming earning announcements, iResearch is probably not a very good source of information. If it proves to be right, we would have a great source to know how the companies do just a few weeks before the earning announcements.

The following table gives the official 1st quarter result, the 2nd quarter estimate from iResearch, and the % increase quarter over quarter:

1Q2008 search rev.

2Q rev. (estimate)

% increase

















It predicted a search revenue for Baidu of 114.9 million for 2Q2008. Adds other revenue of about 0.2 million, we would have total revenue of $115.1 millions. It compares favorable to the average analysts estimate of $112.27 millions. If iResearch is right, we shall be a nice pop to Bidu’s stock price right after they announced their 2nd quarter earning.

On the other hand, if iResearch is right, then Sohu’s Sogou is rather disappointing. It is so close to the Olympics and you would think Sogou would have gain since Sohu is an official Olympics sponsor and the producer of the official Olympics web site.

For Sina, they gave up on search already. For Netease, they might as well.

For course, I have no idea how accurate is iResearch. I will do a follow-up after all the companies announced their 2nd quarter earning.

Thursday, July 17, 2008

Olympic battle between the portals

My last post on the Olympics battle between China’s internet portals can be found here:


First, Sohu got right to broadcast CCTV’s Olympics coverage. It is time for Sohu to crow about how much advantage it gives them:


Not to be out-done, the alliance members, Sina, Ntes, and Tencent also announced that they got the license to broadcast CCTV’s Olympics coverage also



From the above article, CCTV has 9 partners that have license to rebroadcast CCTV Olympics coverage.

Now, this is just re-broadcasting TV programs. I believe all four companies also have the right to do original programming on their own. For that part, Sohu has the advantage in that it has more reporters’ licenses because it is also the official web site provider.

Finally, the following article provides a pretty good insights to Sohu’s Olympics strategy


This is an article about an interview with Sohu’s CEO on 7/14/2008. Some major points on this article:

- Regards to Olympics coverage, he felt that Sohu has two major advantages, the breadth of coverage and the speed of coverage.

- There will be major investment for the Olympics.

- There will be marketing campaign such as ads on public buses. They will also do marketing on the Olympics relay cities. The video broadcasting on the Olympics relay run is pretty expensive also.

- The money made from the hit game TL is really helpful. It more than offsets the additional expenditure for the expense paid for the preparation of the Olympics coverages.

- The three major pillar of Sohu’s Olympics partnership is

1. Sohu is the official sponsor and official web site developer of Olympics

2. partnership with CCTV

3. partnership with www.sports.cn

- The Olympics Strategy for Sohu is basically to build the Sohu brand during the Olympics.

- Sohu had build all its products such as Sohu News, Sogou Pinyin Input Method, and Sohu community to keep those users after the Olympics.

Note that Sohu is the developer of the official Beijing Olympics web site, http://www.beijing2008.cn/index.shtml. On the other hand, www.sports.cn is the developer of the official web site of the Chinese Olympics Committee.

I think its partnership with CCTV gives Sohu a lot of TV programs. On the other hand, as the developer of the official web site and its partnership with www.sports.cn give Sohu a lot of reporter press pass to allow Sohu to do a lot of original TV production.

In the end, I believe Wall Street had a misconception about Sohu’s core strategy regards to the Olympics.

I think Wall Street think Sohu will benefit greatly during the Olympics and fades badly afterward.

On the other hand, Sohu’s core strategy is to use Olympics as a jumping point to dominate China’s Internet space. It will probably benefit from Olympics advertising. But it is also spending heavily to establish itself and grabbing market shares from the likes of Sina, Ntes, Bidu, and Tencent.

It is building capability so that it can keep the market share afterward.

If Sohu executed its strategy, it might not have make as much profit as the wall street expects during the Olympics; but it will trounce wall street’s expectation after the Olympics.

Wednesday, July 16, 2008

NTES may get the right to Starcraft 2

My last post on NTES can be found here:


There are quite a few stories on NTES. I will post them in the next few days.

There is a consistent rumor that NTES will become the operator for the game “Starcraft 2” in China. See the following story:


Starcraft 2 is a major game developed by Blizzard, the same company that made World of Warcraft. But Starcraft 2 is a single player game. It will probably have some limited online features such as battle field, etc.

The game itself is probably not that important for NTES. But its significance is that it gives NTES a leg up if Blizzard is going to develop “Starcraft Online”. Starcraft Online has the potential to take the mantel from the block-buster "World of Warcraft".

Interestingly, Blizzard vice CEO, Frank Pearce, hinted that the Starcraft 2 will be in China soon.


This is just a rumor for now. But if this is true, I am happy for NTES. But it will not be a good news for NCTY.

Tuesday, July 1, 2008

NTES – Two new games went into closed beta

My last update on the new game of XYW can be found here:


My last update on TX2 can be found here:


My last update on FF2 can be found here:


NTES had been very slow in releasing games in the last 3 to 4 years. It only release two major titles in the last 3 years, DT two years ago and XY3 last year. DT was a complete failure while XY3 is growing at a very slow pace. The verdict is still not out on the XY3 yet.

But for some reason, the second half of 2008 and first half of next year will promise to be somewhat of a watershed period for NTES. It will have four major titles all coming out at about the same time.

First, on 6/6/2008, TX2 went “unlimited and open closed beta”. It start to charge players money for items. It is about a month now and all indication had shown that the game itself is a great game. Players love it and it is very sticky. Unfortunately, NTES resorts to its old psychotic aversion to marketing its games so the growth had not been exponential. It had its sales team going to many internet café to push for TX2. But NTES’s ground sales team is no where close to the effectiveness of the ground sales team of Giant Online (for example). In addition, there are no advertising on buses, subways, LCDs, street signs, internet gaming sites, or TV programming, etc.

As a result, the growth of TX2 had been slow so far. But still, it is still on a solid growth path.

Another game, FF2, had been in closed beta since 2/1/2008. FF2’s development team had been very secretive about its progress. But I got the feeling that, if NTES wants, it could get it out in the next couple of months. I suspect the only reason it is being hold back is not to get every major title out at the same time.

From the following two articles:



Another major title, XYW, will go into closed beta (with several thousand players) on 7/5/2008. XYW is somewhat of a composite of XY2, XY3, and XYQ. It is build on the same game engine of XY3. It will have a graphic style of XY2 (cartoonish but not too cute) and many playable features of XYQ. Of course, it will be free to play and therefore will emphasize player combat more while the other three existing games emphasize community features more.

From the above two articles, another major title, DT2, will go into closed beta in July 2008. DT2 is another free-to-play title. For a while, I had thought that DT2 would be NTES’s first free-to-play title (either DT2 or FF2), but it turned out not to be the case. It took NTES two years for DT2 to come out. I don’t know why it took so long. I had thought that DT2 is just a minor re-write of DT and just a straight conversion from paying game to free-to-play game.

Obviously, that is not the case. This speak volumes about NTES the company. There is no cookie cutter for any of its games. Every games is a major undertaking. They strives for excellence for every projects.

It is just too bad the senior managers are not willing to spend the marketing dollars to push its new games.

TX2 had been under development for about 5 years now. It still have an open beta to go. FF2 had been under development since June of 2005. By the time it comes out, probably later this year, it would be 3.5 years of development.

DT2 had been under development for two years. If it comes out later this year, it would be 2.5 years of development.

The speed of XYW’s development surprised me a little. It had only been 6 to 9 months of development. I think the reason XYW can be out so fast is because NTES doesn’t have to re-develop the underlining game engine from the ground up.

TX2 had to go through two stages of engine development work with its Australian game engine developer to optimize its game engine. FF2’s game engine is complete re-wrote from the ground up. The original DT’s game engine is in-house developed. I am certain they did a lot of game engine upgrade for DT2 also.

But for XYW, it is different. NTES spent a lot of money and effort to develop a brand new game engine for XY3. But XY3, though a somewhat successful game, didn’t become a blockbuster. Therefore, NTES decided to use its brand new game engine to develop another game. That game is XYW. The analogy will be that the foundation is build and only the floor plan is changed.

Perfect World had been doing this for all its games. They use the same game engine and spitting out new games like automobiles from an assembly line.

Going back to NTES. If one purely based on development schedules, I am certain all 4 games could go into commercial this year. But for factors such as no competition with its own games, NTES might want to stagger the release date so that one or two of the major titles could come out early next year.

Anyway, it is exciting time for NTES. I only hope NTES’s senior managers can provide more marketing help to its games.


A test for now