Jump to content

GOOGL - Google


Liberty

Recommended Posts

Merci Liberty! Very interesting stuff showing one of the main reason why Google has such a moat! And showing nice engineering and a great vision too!

 

Note that anything you see here is probably a few years behind Google's state of the art (this is according to an engineer who helped build these data centers on another discussion forum).

Link to comment
Share on other sites

  • Replies 2.1k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

Google has some very impressive cloud infrastructure technology.  They:

- Built their own file system and storage (bypassing EMC at its ilk)

- Build their own networking (bypassing Cisco)

- Build their own servers (bypassing Dell, HP, Supermicro)

- Have a very low PUE

etc. etc.

 

But I'm not sure that it gives them a huge advantage.  Look at Google Video versus Youtube, or Google+ versus Facebook.  Sometimes the operating costs of the servers aren't necessarily what matters.  For many situations it's the quality of the software that matters the most.

 

*Granted, In the long run, low costs may be what matters for online distribution of video.  That is possible.

Though I think that what will matter in online video is the ability to monetize effectively.  In ad-supported online video, Google has an advantage because it can use information about the user (e.g. from Google searches) to target ads.  If online video moves towards subscriptions (e.g. Netflix), then Google may be left out of the game because it does not sell its infrastructure to third parties.

Link to comment
Share on other sites

Google has some very impressive cloud infrastructure technology.  They:

- Built their own file system and storage (bypassing EMC at its ilk)

- Build their own networking (bypassing Cisco)

- Build their own servers (bypassing Dell, HP, Supermicro)

- Have a very low PUE

etc. etc.

 

But I'm not sure that it gives them a huge advantage.  Look at Google Video versus Youtube

 

You do know they own Youtube right?

Link to comment
Share on other sites

It's definitely a cost/efficiency advantage.  It's also a reliability and flexibility advantage.  Google's infrastructure allows them to switch computing tasks at huge scale.  YouTube independent of Google could never reach the scale that they have.  Youtube's founders were great marketers first and decent engineers second.

 

I agree with a couple more of your points here..  Google has excellent insight into individual users' preferences, which gives them an advantage in ad display efficiency.  The other point I agree with is that Google's "go to market" strategies are really hit and miss.  Search, Gmail, Maps, and News were all but  flawless, but Buzz, Google+, Wave, and Video were pretty terrible.  Early on in this discussion thread I stated that Google makes very public and risky product decisions and that we're just witnesses to the resulting creative destruction.  I have to constantly remind myself that this is part of the process.  It's a bit like when value investors take a loss on a position and remind themselves it's OK because the process was right :)

 

Link to comment
Share on other sites

Nah you guys aren't hitting the point. Their advantage is not their server infrastructure - other companies can build those things if they need to. See Facebook and MSFT or AWS.

 

Their advantage is the quality of their search function. EVERYTHING feeds into that, and search gets stronger the more people use it, and the more people use it, the better it gets, which also attracts advertisers. Their business model is their competitive advantage. It is very difficult to replicate. The server infra mentioned earlier simply enables this.

 

Once you get used to google, there is no reason to switch to another search engine. It is like a "consumer staple".

Link to comment
Share on other sites

Nah you guys aren't hitting the point. Their advantage is not their server infrastructure - other companies can build those things if they need to. See Facebook and MSFT or AWS.

 

Their advantage is the quality of their search function. EVERYTHING feeds into that, and search gets stronger the more people use it, and the more people use it, the better it gets, which also attracts advertisers. Their business model is their competitive advantage. It is very difficult to replicate. The server infra mentioned earlier simply enables this.

 

Once you get used to google, there is no reason to switch to another search engine. It is like a "consumer staple".

 

Google has more than one area of competitive advantage.  The search piece is obviously Google's greatest advantage and everything you said is completely true.  However, the infrastructure piece is a completely different area that's interesting on its own and still worth a discussion.  The point I'm making is that it's incrementally harder to displace Google in any of its leading categories because of their approach to infrastructure.

 

Taking YouTube as an example, about 16 years of video are uploaded every day.  Assuming it's 480p video and they have decent compression, that's about 50 terabytes of new data every day.  And it's available on tap immediately.  Just think through what that requires in terms of both hardware and software.  That Google handles this problem seamlessly is an engineering marvel.  That ability stems directly from their infrastructure, and it's mostly a software thing, not a hardware thing.

 

Link to comment
Share on other sites

The hardware often permits them to do software that others can't. Instant Search, for example, requires a lot more processing power and bandwidth than just a regular search. Even the adsense network is HUGELY hardware-intensive and not easy to do for a smaller/less-efficient competitor. The quality of searches is also dependent on hardware bottlenecks; more processing power and bandwidth means you can crawl more often and take more signals into account to determine rankings, and refresh cached results more often.

 

Software and hardware aren't as independent as some people think.. One usually limits the other.

Link to comment
Share on other sites

That Google handles this problem seamlessly is an engineering marvel.  That ability stems directly from their infrastructure, and it's mostly a software thing, not a hardware thing.

 

Here is where I see their hardware advantage working:

-Gmail used to offer a huge amount of storage compared to other free email.  Now I don't think anybody cares about this.

-Google search indexes far more sites than Bing.  Sometimes people use Google search to find a specific page on a website... you can't do that with Bing.

 

On the other hand, the availability of venture capital and capital from companies like Microsoft means that many companies are willing to run money-losing operations against Google.  Losing money is not stopping the competition... Google's lower costs does not protect them against competition.  It improves their profits but does not protect their market share.  It's the quality of software IMO that determines market share.

Link to comment
Share on other sites

On the other hand, the availability of venture capital and capital from companies like Microsoft means that many companies are willing to run money-losing operations against Google.  Losing money is not stopping the competition... Google's lower costs does not protect them against competition.  It improves their profits but does not protect their market share.  It's the quality of software IMO that determines market share.

 

I think you're overlooking the meaning of what an advantage is.  A cost advantage is an advantage.  If Google can operate at half the cost of a competitor, then they can invest those savings into extending their lead against that competitor.  The competitor must spend twice as much to get the same thing.  The existence of competition hurts Google, but suggesting that you can just throw money at a problem and that negates the advantage simply ignores the economic reasons for developing an advantage in the first place.  So yes, a cost advantage absolutely protects their market share and discourages competition.

 

Link to comment
Share on other sites

Guest valueInv

That Google handles this problem seamlessly is an engineering marvel.  That ability stems directly from their infrastructure, and it's mostly a software thing, not a hardware thing.

 

H  It improves their profits but does not protect their market share.  It's the quality of software IMO that determines market share.

Take a look at today's earnings release.

Link to comment
Share on other sites

I think you're overlooking the meaning of what an advantage is.  A cost advantage is an advantage.  If Google can operate at half the cost of a competitor, then they can invest those savings into extending their lead against that competitor.

 

I guess what I am trying to get at is that I make a distinction between an advantage and a moat.  To me, a moat is something that helps you retain market share.  These companies are safer in the long run and deserve a higher multiple in my opinion. 

 

Google's cloud infrastructure advantage didn't help in competing with Youtube (which they ended up buying) and didn't seem to help it in competing with Facebook.

 

2- In technology, there are limits to R&D spending.  Doubling the number of programmers will NOT double productivity.  Hiring more good programmers is hard and doesn't scale well.  Getting large groups of programmers to work productively doesn't scale well (e.g. see the Wikipedia entry on the mythical man-month).

 

In the history of tech, it's often the case that some small company develops some type of disruptive technology that causes titans (with bigger R&D budgets and more cumulative R&D spending) to fall.  This is what Google did to its predecessors like Yahoo.  Having a big R&D budget is not that strong of a moat.

 

I would agree with the argument that low costs --> more R&D --> slightly better product --> somewhat of a moat.  But... if a company develops better search than Google with higher costs, I will want to bet on better search.  The company with the better search will be the market share winner, not the company with the lower costs. 

*Right now I think Google has the best search.

Link to comment
Share on other sites

Guest valueInv

Take a look at today's earnings release.

 

Could you provide a little more analysis on this?  Take a look at what exactly?

 

I think we would all prefer a specific point instead of a vague suggestion.

 

http://online.wsj.com/article/SB10000872396390443684104578064671358259436.html?mod=pls_whats_news_us_business_f&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+wsj%2Fxml%2Frss%2F3_7014+%28WSJ.com%3A+US+Business%29&utm_content=Google+Feedfetcher

 

Revenue growth was not a problem, profitability was. Mobile CPCs are lower, so as people transition to mobile CPCs decrease. Further, on mobile they have revenue sharing agreements with operators (one of the reasons why Android has grown so quickly). So while on the desktop, they keep almost all of their revenue, on mobile they don't. The technology cost to serve a search or an ad is only part of their cost.

 

There is a new cost coming up - device subsidies. They are subsidizing tablets and now laptops. And they are locked in a price war with Amazon. Amazon introduced the $199 Kindle, Google follows with the $199 Nexus 7. Amazon drops the price to $159, now Google is rumored to introduce a $99 Nexus tablet. They just introduced a $259 Samsung Chromebook. Acer and Samsung are not going to be selling their devices at cost, guess who is the sugar daddy?

Link to comment
Share on other sites

I think you're overlooking the meaning of what an advantage is.  A cost advantage is an advantage.  If Google can operate at half the cost of a competitor, then they can invest those savings into extending their lead against that competitor.

 

I guess what I am trying to get at is that I make a distinction between an advantage and a moat.  To me, a moat is something that helps you retain market share.  These companies are safer in the long run and deserve a higher multiple in my opinion. 

 

Google's cloud infrastructure advantage didn't help in competing with Youtube (which they ended up buying) and didn't seem to help it in competing with Facebook.

Well it probably did help them in both cases, because they were able to spin up an application that could reach scale with very low incremental cost.  In this scenario you're simplifying the complexity of bringing a service to market as simply "win" vs. "lose".

 

I will gladly agree that infrastructure alone does not create a moat for the businesses that Google is in, but that's not really what we were talking about.

 

 

2- In technology, there are limits to R&D spending.  Doubling the number of programmers will NOT double productivity.  Hiring more good programmers is hard and doesn't scale well.  Getting large groups of programmers to work productively doesn't scale well (e.g. see the Wikipedia entry on the mythical man-month).

 

This part is getting a little bit tough for me to follow.  Initially you had argued that competitors can just dump money into beating Google and therefore infrastructure doesn't count.  Now you've reversed your position by saying that a competitor can't dump money into solving a problem because there is an upper bound to what a large team can achieve.  Which do you actually believe?

 

In the history of tech, it's often the case that some small company develops some type of disruptive technology that causes titans (with bigger R&D budgets and more cumulative R&D spending) to fall.  This is what Google did to its predecessors like Yahoo.  Having a big R&D budget is not that strong of a moat.

Really? I've never heard of that scenario before...

 

I would agree with the argument that low costs --> more R&D --> slightly better product --> somewhat of a moat.  But... if a company develops better search than Google with higher costs, I will want to bet on better search.  The company with the better search will be the market share winner, not the company with the lower costs. 

*Right now I think Google has the best search.

 

You've somehow latched onto the idea that reinvestment only comes in the form of R&D.  I told you that saving money allows Google to use those savings to extend their lead over competitors.  The form of that extension doesn't necessarily have to come from R&D activities.  It could be from marketing, sales, investing in complementary products, buying other companies (like YouTube maybe).  I really don't think you're going to get very far with the argument that costs don't count in tech.  They absolutely matter, and reducing them leaves more money in the war chest for, like, war stuff, yo.

 

 

Link to comment
Share on other sites

http://online.wsj.com/article/SB10000872396390443684104578064671358259436.html?mod=pls_whats_news_us_business_f&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+wsj%2Fxml%2Frss%2F3_7014+%28WSJ.com%3A+US+Business%29&utm_content=Google+Feedfetcher

 

Revenue growth was not a problem, profitability was. Mobile CPCs are lower, so as people transition to mobile CPCs decrease. Further, on mobile they have revenue sharing agreements with operators (one of the reasons why Android has grown so quickly). So while on the desktop, they keep almost all of their revenue, on mobile they don't. The technology cost to serve a search or an ad is only part of their cost.

 

Well the CPC story isn't told entirely by mobile.  There were also significant currency fluctuations that caused the CPC to drop.  But yes, mobile is causing CPC to drop because fewer advertisers are bidding on mobile traffic.  Last I checked, everybody in the tech world has a big hard-on for mobile technology and they tend to lead the pack with this kind of thing.  Care to wager how long it's going to take before advertisers decide that they should attract more mobile traffic?  Or do you think that advertising on mobile devices will never take off?

 

My bet is that certain keywords on mobile will rise above their desktop counterparts, where others drop below.  For example "french bistro" will probably be highly coveted in mobile form.  Whereas "enterprise resource planning software" can likely get picked up on the cheap.

 

There is a new cost coming up - device subsidies. They are subsidizing tablets and now laptops. And they are locked in a price war with Amazon. Amazon introduced the $199 Kindle, Google follows with the $199 Nexus 7. Amazon drops the price to $159, now Google is rumored to introduce a $99 Nexus tablet. They just introduced a $259 Samsung Chromebook. Acer and Samsung are not going to be selling their devices at cost, guess who is the sugar daddy?

 

Where is the device subsidy cost in the earnings report?  What is the figure?

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...