A look at the customers of the Amazon Web Services business
"It is a riddle, wrapped in a mystery, inside an enigma," was Winston Churchill's timeless quote about predicting Stalin's Russia. The same might be said by investors about Amazon Web Services; ostensibly the big growth opportunity in the Amazon (AMZN) jungle under Jeff Bezos, who has a knack for pleasing investors, but doesn't always seem so keen on doing so.
Seeking Alpha contributor Dana Blankenhorn observes that "No company is causing as much consternation here at Seeking Alpha these days as Amazon.com," while Michael Fu states that "Amazon is one of those stocks that either you love or hate," preceding his excellent stock analysis. Love Amazon or hate them, there is no doubt that a majority of the hot startups start up in the Amazon cloud - see my blog for a partial list - and that is where the opportunity purportedly lies.
AWS can be frustratingly opaque to the investor. As noted by Shrideep Murthy, the AWS service is a small percentage of Amazon's revenue, placed in a segment called 'Other' on Amazon's annual report. This reticence tends to obscure a key component of the future of the company; I thought it would be interesting to reverse engineer the question and consider what makes AWS tick.
Ruby on Rails is one of several technology frameworks that have skewed new Internet development toward the Amazon cloud. Ruby on Rails applications are known for being easy to build - new product concepts can be prototyped, pointed, clicked and put on the Internet in a day. This trend toward easier to use tools triggered a shift toward the cloud reflected in most modern frameworks. Consequentially, over the past few years we see startups with highly developed products and business models.
Amazon's AWS platform is dominant in the area of web application hosting and has a competitive product offering for virtually every technology startups might consider. AWS is so pervasive in fact that two most notable alternatives, Heroku and Engine Yard, use AWS to power themselves. The AWS metered model essentially allows the user to log on, point, click and build a server - this being a process that could have taken weeks and tens of thousands of dollars in the past.
Some of the most dominant startups are Amazon customers. For example, part or all of Reddit, Foursquare, Pinterest, Instagram, Dropbox and even that Internet juggernaut known as Seeking Alpha find a home in the Amazon cloud - and that is just the tip of the AWS iceberg. As these companies continue to gather steam it is reasonable to anticipate Amazon doing well in concert with them, but to what extent, has been the subject of much debate.
What is clear is that Amazon has been showing the startups a whole lot of love for some time, for instance giving hard cash (not to mention publicity, technology and intangibles) every year since 2007 in the AWS Startup Challenge. That strategy of growing one's own customers has worked well for AWS.
Note that the utility of AWS is not limited to Internet businesses and startups; a quant friend related that he keeps the Microsoft (MSFT) SQL Server database for his trading analytics in the AWS cloud. Established companies such as CareerBuilder.com are deploying new product features with AWS as an alternative to expanding their current data centers. Much has been said, of course about Netflix (NFLX) finding coal in their stocking when the service went down the chimney on them over Christmas.
It could be argued that Amazon is doing for Internet businesses what Apple (AAPL) did for consumer electronics - it is shifting paradigms and in so doing, seeping into the popular bloodstream of the startup culture. The other service providers are service providers, but Amazon has begun to corner the market in mind share as to how a software startup should be built, such inertia could prove difficult to disrupt or to compete with.
Even so, it seems clear that this upside is currently priced into what is currently a very expensive stock. It seems likely that the stock will be available at a lower price later. Certainly the premium currently carried with the stock includes the prospect of some volatility. Smart money will wait to buy AMZN once that premium is priced out.