The first public cloud service was a spin-off from Amazon’s main business. Mark Young thinks that cloud start-ups will have their own spin-offs, in Big Data
In 1995, Amazon.com was launched. Using the Internet as a store front and therefore not shackled in its product portfolio by what it could fit on physical shelves, the company was able to exploit the ‘long tail theory’ – that the majority of your customer base actually seek a greater spread of products rather than, or in addition to, the bestsellers.
Offering books to begin with, and music, clothing, electrical and other household products soon after, Amazon.com flourished by selling small quantities of a massive catalogue of individual items.
As the website grew, more and more compute power and storage had to be constantly added to handle the demand, with ample resource needed to cover the huge peaks of traffic that the website experienced in the lead up to festive periods. This resource took the form of huge data centres.
Amazon starts it all off…
But with much of this resource sitting unused for the majority of the time, Amazon looked for a way to make money from it. ‘Why not’, they thought, ‘rent it out to other companies that need compute and storage resource but can’t afford to buy their own?’ Amazon Web Services (AWS) – and the public cloud as many know it today – was born.
AWS is technically run as a separate business to Amazon but is included within the same reporting figures. It is included as part of Amazon’s ‘others’ list on its financial report and is widely assumed by analysts to be the majority contributor within this department. In Q1 2012 ‘others’ accounted for $500m in revenue.
… Google gets on board…
Similarly, Google began as a search engine and burst its bank account with advertising. But having created its Gmail email service to further engage with and track its audience, the company quickly saw an opportunity to market the service to businesses with cloud hosting.
Having added the instant messaging, document sharing services and video functions that it developed to communicate and collaborate effectively in-house, Google Apps for Business was born. There are many companies now that swear by it. Google is as coy in its reporting of Its Apps for Business figures as Amazon is with AWS. Recent estimates from the analyst firms put it at anywhere between $150m and $1bn annually, and this doesn’t take into account the value of its gains in the suites and hosting space with Microsoft.
This principle – taking the resources, skills and infrastructure you have built up to perform your primary business and making a new product line out of them – is nothing new, of course. Many of the world’s biggest conglomerates have been built upon it, steering natural courses through mining, steel, power, car making, logistics and so forth. Businesses even do it with their waste; Marmite exists because of the yeast left over after brewing beer, or so the legend goes.
Who is next for secondary services?
But the latest iteration of this vertical integration – the information-based resource repurposing that Amazon, Google and their kin are the chief proponents of – does have at least one new facet, and it’s one that has huge, positive ramifications for the rest of the business world. It has opened the door for any business of any size to follow suit with lucrative secondary services of their own. Although Amazon, Google and the traditional conglomerates needed huge amounts of capital to develop and release their secondary services, the cloud makers’ efforts means that other businesses that deal in information now don’t have to.
To be clear, this isn’t about improving existing services or adding new lines of the same ilk to existing portfolios, though there are plenty of examples where cloud helps with that – Barclays ‘Pingit’ text message transfer system is one; Autodesk’s cloud based CAD design rendering service is another. This is about doing something that might be related to your primary business in some way but essentially represents a totally different business; one that could eventually overtake the primary one in its value.
Take the start-ups who make Internet and mobile phone applications, many of whom now prosper in East London. Many of these companies begin by building a consumer application from an entertainment perspective (with most relying on public cloud and, it has to be said, primarily Amazon Web Services to launch in the first place, although that isn’t the point here).
They have an idea that they think their audience will enjoy and originally see their revenue model as charged usage and advertising. They use feedback from their audiences in terms of actual reviews and, perhaps more reliably, usage patterns of their applications to redevelop them and improve the product.
Big usage makes for big data
But pretty soon, the ones that get big usage numbers realise that the data that they don’t use for their own products but that they capture anyway by-the-by (such is the privilege of the digital world) could offer them something much more lucrative than their primary business models. Using the information their customers provide them they can map subcultures, commercial trends and consumer opinion which they can then package and sell to the highest bidder.
The music identification app Shazam, for instance, realised that their charts of which songs are being identified generally act as a precursor for the US billboard charts in six weeks time. The record labels, or anyone else with a vested interest can take this information and make interventions through their audience engagement according to their own objectives, whether that’s advertising more in a region where sales are low or marketing similar products in an area where a trend has arisen.
The cloud, offering the ability to spread huge data sets across a network of nodes using parallel computing with a few clicks of a button, is essential for this to happen. Without it, none of it would be possible.
It isn’t just the start ups either. BT – a seasoned performer in repurposing its internal tools – has recently announced its own foray into the data insight world. The company has released a new product, Assured Analytics, which is essentially the data amalgamation and reporting services that it uses itself to monitor the health and security threats of its copper cable networks.
With internal applications like these, the business users require the technological developers to present them with a polished, user-friendly interface with full functionality that can be exported across multiple platforms and with the ability to deliver it quickly and repeatedly to departments and sub-businesses all over the world.
Software-as-a-Service – only possible on a mass market scale because of cloud – has become the best way to do this and the norm. And now that there is no need for packaging, creating disks, physical retail outlets and such, what is the difference between delivering it within the organisation and selling it externally? A pay wall and marketing. There is also the added bonus that if it is hosted on your servers than you get access to the data the company generates, at an overall trends level which you can use for your own business intelligence or license back out to the market. The only cost is hosting and data storage and thanks to public cloud you only pay when you earn.
This ‘second service’ development probably falls under the bracket of agility if you were attempting to define it in the spectrum of cloud benefits. Essentially with information-based products and services the cloud has removed the barriers to market – whatever you create, whatever you use, there’s a good chance you can sell it, with minimum initial outlay and minimal risk. That’s why cloud is about so much more than process efficiencies.
Mark Young is editor at The Cloud Circle
Is cloud your main business or a side earner? Try our cloud quiz