Cloud // Infrastructure as a Service
Commentary
12/6/2012
08:37 PM
Charles Babcock
Charles Babcock
Commentary
Connect Directly
Twitter
RSS
E-Mail
50%
50%
Repost This

Why Amazon's Cloud Business Is Like Kindle Sales

Amazon Web Services is cheap upfront, and Amazon only gets paid if people keep using it, CEO Bezos says.

10 Cloud Computing Pioneers
10 Cloud Computing Pioneers
(click image for larger view and for slideshow)
Amazon Web Services (AWS) had a coming out party of sorts when it staged its first partner and customer event, dubbed Re: Invent, and held in Las Vegas Nov. 27-29. The most revealing moments came when Amazon CTO Werner Vogels took the stage Nov. 29 to pitch questions at CEO Jeff Bezos.

In creating its infrastructure-as-a-service (IaaS) unit in 2002, Amazon did something both risky and bold. It concluded it had a new type of infrastructure behind its retail operation and further services could be created on it. Better yet, it could create a new service on an infrastructure optimized to work with end users. There were no cloud users back then. There were hosted service providers and managed service providers and colocation facilities. But letting people commission and control infrastructure on an hourly basis, paying as they go with a credit card, was unique.

Vogels started his conversation with Bezos by pointing out that the last time they shared a stage was at the announcement of Amazon's second-generation tablet, the Kindle Fire. This was a soft pitch, of course. Bezos picked up the cue and drew an analogy between the $159 Kindle and IaaS.

"One of the unusual things we do in the Kindle device business is we sell our hardware at near breakeven. We make money when people use the device, not when they buy the device ... If I buy the device and put it in my desk drawer and never use it, then Amazon doesn't deserve to make any money," said Bezos.

[ Want to know how Amazon drives down computing service prices? See Amazon Web Services Slashes Storage Prices. ]

Amazon Web Services uses the same principles, Bezos continued, closing the loop. "It's a pay-as-you-go service. We are not incented to get people to overbuy hardware and operate at low utilization rates," he said, taking a shot at traditional hardware businesses.

Furthermore, such an approach is a good discipline for any business because it will stay tuned to what customers actually need, not what they can produce. "Our point of view is if we can arrange things in such a way that our interests are aligned with our customers', then in the long term that will work out really well for customers and it will work out for Amazon," Bezos added.

In the same vein, he threw in a plug for how Amazon's tablet business is different from Apple's: "Likewise, it causes us to have the right kinds of behavior. If we're not making money when people buy the device, then we don't need people to be on the upgrade treadmill. We have people using 5-year old Kindle Ones and we're perfectly happy with that.

I find this analogy strained but still revealing of the core of AWS. Selling a hardware device is different from providing a cloud service, although the analogy works somewhat when the hardware device becomes a platform for the same vendor to sell services. That part's OK. The main point of these comments is that Amazon is not making money until its customers decide what has value to them in the form of an Amazon service. That's a major differentiator between Amazon and HP, Dell or IBM. And the fact that it sells hardware at breakeven is a major differentiator between Amazon and Apple. Point established.

Next, Vogels solicited Bezos to address the notion of staying focused on essential business services. "I'm always amazed that you talk about the notion of flywheels," he prompted.

To Bezos, a business must be built around building out long-term services, not chasing short-term profits with short-lived devices. "I frequently get the question, 'What's going to change in the next 10 years?' I almost never get the question, 'What's not going to change in the next 10 years?' I submit that second question is the more important of the two -- because you can build a business strategy around the things that are stable over time. We know the customers want low prices. We know that's going to be true 10 years from now. I can't imagine a customer coming up to me and saying, "I just love Amazon. I just wish the prices were a little higher. I just wish delivery would take a little longer." Invest in those things today (that will) pay dividends 20 years from now."

Bezos again closed the loop to show how this core Amazon ideal is part of AWS. "At AWS, the big ideas are also pretty straightforward. It's impossible for me to imagine 10 years from now someone would tell me, 'I love AWS but I just wish you were a little less reliable. I love AWS but I wish you would raise prices ... I wish you would innovate and improve APIs at a slower rate.'

"The big ideas in business are often very obvious but it's very hard to maintain a firm grasp of the obvious. But if you can do that, if you can continue to spin up those flywheels, and put energy into those things, like we do with AWS, over time, you build a better and better service for your customers," he said.

Concentrating on core competencies is not a new idea, but what Bezos said next sets up what must be the tension between Amazon.com and its AWS business unit. If you are setting up a new business, you can't be sure where you'll find the next "flywheel." Having gotten a new style of retail going online, Amazon.com had a chance to generalize upon it and move beyond retail into specific services based on the lessons learned.

Previous
1 of 2
Next
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
melgross
50%
50%
melgross,
User Rank: Ninja
12/10/2012 | 7:16:05 PM
re: Why Amazon's Cloud Business Is Like Kindle Sales
Bezos is one of the world's greatest salesmen. As an innovator, he's not as great as is being stated here. One fact about innovation is its success in changing the way things are done. They've has some success in that, but not uniquely.

But what he prefers to gloss over, and those who interview him gloss over as well, is that he badly underestimated the expenses his business would need. Lets be honest here, his original concept was to have a small staff handling the web site, and sales, while product shipment would be handled by manufacturers and wish fulfillment companies. They quickly discovered that it wasn't to be. Their expenses opening and running product warehouses and shipment centers has cost them dearly. These expenses will never go away. The larger they become, the more centers they will need to open. Their expenses will always remain high.

As for them selling Kindle readers and tablets, remember his words carefully. In sales parlance, "near breakeven" means loss. It doesn't mean breakeven. So they are taking a loss on all their Kindle sales. If they are duplicating that performance for their cloud services, that's breaking rule number one in class 101 economics, which is to never take a loss on one product in order to make a profit on another, or on the same, later on. This doesn't equal the razor and blades for obvious reasons, or what should be obvious reasons.

I have little confidence that Amazon will ever make real profits, and that gives me a nervous feeling about those profitless products and services. There is only so long before you must raise prices substantially, or discontinue the product or service. Which will it be?
2014 Private Cloud Survey
2014 Private Cloud Survey
Respondents are on a roll: 53% brought their private clouds from concept to production in less than one year, and 60% ­extend their clouds across multiple datacenters. But expertise is scarce, with 51% saying acquiring skilled employees is a roadblock.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Government, May 2014
Protecting Critical Infrastructure: A New Approach NIST's cyber-security framework gives critical-infrastructure operators a new tool to assess readiness. But will operators put this voluntary framework to work?
Video
Slideshows
Twitter Feed
Audio Interviews
Archived Audio Interviews
GE is a leader in combining connected devices and advanced analytics in pursuit of practical goals like less downtime, lower operating costs, and higher throughput. At GIO Power & Water, CIO Jim Fowler is part of the team exploring how to apply these techniques to some of the world's essential infrastructure, from power plants to water treatment systems. Join us, and bring your questions, as we talk about what's ahead.