Google’s Economic Engine

To communicate the credibility of this challenge, it is salutary to remind ourselves that when it comes to servicing the consumer segment, Google’s economics are superior to Windows’, and they can afford to think big in their plans to corner this segment.

At present, Windows generates approx $9 per user per year for consumers in the US; Google generates an ARPU of around $30 in the US.  The Windows ARPU is projected to be flat (absent gains in piracy), while Google’s ARPU is expected to exceed $50+ per user by 2009.

Google economics are supported by three pillars:

1. The vast size of the global advertising market ($500B globally, $300B in the US), and its structural drive to find virgin-territory to distribute its messages. It colonises new territory by overpowering the economics of conventional business models and making them uncompetitive (cf. TV content, public transport infrastructure, certain print media.) At present, online ads are worth $16B, by 2008 will be worth $30B, and by 2015 will be worth $100B+. Search advertising alone is estimated at $15B in 2008. Search therefore has the capacity to consume the consumer software market.

2. The high margins of the search business. Search has gross margins of 85%+ per search (this includes amortization of Google’s back-end costs). The margin on each incremental search is probably 95%+. These margins allow Google huge capacity to fund marketing/distribution or development efforts that will increase their query share. While margins will probably deflate for a period when competition increases, the effectiveness of Search as an advertising vehicle, combined with low marginal cost and high barriers to entry, means that the long-term margins should stay very high.

3. The economies of scale in search.  Search shares many characteristics with the OS business. Two in particular are relevant here:
   1. Although operating a search engine has a significant service cost, this cost decreases substantially with scale, meaning that (given architectural parity) the highest volume provider has higher margins than lower volume providers.
   2. Search has ‘winner takes all’ attributes that we expand on in the next section.

The third driver mandates that Google aim for market share dominance, and the first two drivers give them the firepower to achieve that aim.  We believe they will do this via organic growth (driven by software innovation), increasing marketing spend to bolster their brand, and accessing an untapped distribution opportunity in their value chain: the OEM.

The fact that scale drives Google’s (and Microsoft’s) bottom line is obviously not news, and if Google was sincerely eyeing the OEM distribution opportunity simply to amortise cost, then Microsoft would have no choice but to respond. It would certainly dent margins, but it would not unravel our business model.

However, there is a new sting in the tail, and that is the ad-funded business model and what drives it. 

The business model of ad-funded services has made the position of the second-mover considerably more challenging, and it is this principle that we explore in the next section.

The Search Story, and the ‘Winner takes All’ effect

Over the past two years, Google has gained significant distribution of its search engine across the web.  Fully 50% of Google’s gross revenues today are generated outside of Google’s own site.  Much of this revenue is generated from large sites such as AOL.com, where Google pays at or near 100% revenue share in exchange for the traffic (search providers call this revenue share percentage Traffic Acquisition Cost, or TAC.)  Other smaller sites get less TAC, but none lower than 50%.  On average, Google shares about 80% of the revenue generated by a third party site across its syndicated network.

The reason Google is willing to pay up to 100% TAC through syndication is two-fold: 1) increasing end user exposure to Google’s search technology increases the likelihood those users will visit Google.com directly in the future, and 2) more search traffic increases the price-per-click (PPC) Google can get across its entire network.  

That second reason is particularly important, because PPC is a virtuous cycle.  It works like this: as total search traffic increases, more advertisers are willing to spend money on the platform.  As more advertisers come on board chasing the larger audience, coverage of search terms increases.  And, as more advertisers are attracted to bid on the same search terms, the price per click also increases.  So in a nutshell, the way you earn a superior return and gain a competitive advantage as a search platform is to drive more queries than your competitors.

As PPC increases on the platform, the total amount of money the third party distributor makes increases as well.  This effect enables the top platform provider to pay its third party distributors less on a percentage basis, but more on a total revenue basis versus the competition.  For example, if Google has a PPC of $0.50 and MSN has a PPC of $0.40, MSN would have to pay 125% TAC to equal Google’s 100% TAC (assuming equivalent query volume through the third party.) 

Google has taught MSN an expensive lesson: if you plan to make your money from advertisers, volume is everything. Compromising this volume for short-term margins is a losing game. This is not just about capturing market share for market power’s sake, or driving down cost with scale (as detailed above). There is a powerful positive correlation between volume and unit price:

-          For ad buys placed in a bid model (e.g. keywords, as discussed above), volume sets