Lightweight Models and Cost Effective Scalability

Overview

The stock.xchng Logo

The stock.xchng Logo

One of the most important principles behind successful Web 2.0 startups are without doubt the adoption of Lightweight Models and Cost Effective Scalability patterns. In todays market environment, starting small and then being able to scale with demand is absolutely crucial for a successful Internet-based business. With the advent of numerous open-source operating systems, databases, web servers, content management systems and programming frameworks, businesses are now able to create application servers for virtually free—not counting the money spent on hardware and the services of a technical expert, of course. But with the recent proliferation of cloud-based hosting services, such as Amazon EC2, even the hardware part can be outsourced. Of course, the scalability does not only apply to the technical level but to the business model too. Adopting a scalable and cost-effective market strategy and business model, along with scalable technology, ensures that the final product will hit the market in the shortest possible timeframe while at the same time allowing for future growth.

Comparisons

stock.xchng (also abbreviated as SXC) was launched in 2001 by a Hungarian web designer and entrepreneur Peter Hamza. The unique aspect of the service was that it allowed users to share and download high-resolutions, royalty-free digital photographs and illustrations for free, in contrast to other popular commercial stock exchange sites of the time which often required paid membership. The service has social networking aspects as well: users may set up personal profiles, write personal blog entries, participate in discussions in the forums and comment on each other’s work.

Clearly, the service adopted a very effective scaleable business model right from the start: the technology stack itself was based on open-source technologies (Linux, Apache, MySQL, PHP, also known as the famous LAMP stack) and the website was operated in the beginning solely by its creators (three people in total), thus minimising the startup costs and the initial investment. The choice of flexible, robust open-source technologies allowed for the site to grow rapidly as the number of users steadily increased. In a relatively short timeframe, stock.xchng gained a massive popularity and evolved into one the most visited stock images providers on the whole Internet. In 2009, the site had over 2,500,000 million registered users and hosted around 400,000 photos, which eventually resulted in the acquisition of the company by Getty Images in 2009.

Future directions

stock.xchng is a prime example of applying the principles of Lightweight Models and Cost Effective Scalability into practice. It is indeed possible to build a highly successful online business by putting in minimum effort at the start in order to achieve maximum gain in a very short amount of time—just an excellent, unique idea is needed. As more and more free tools, hosting services and cloud-based platforms are becoming available, the barrier of entry gets even lower: one just needs to wire together existing components, services and frameworks, which can often be achieved even without writing a single line of new code. The result of this will be that we will see more and more interesting services emerging and gaining popularity, which is a benefit for the all of us.

References

stock.xchng – the leading free stock photography site
stock.xchng – site info
Wikipedia – stock.xchng
LAMP (Software Bundle)
Stock Photography, Royalty-Free Photos, Video Footage & Music | Getty Images
Stockxpert / SXC Press Release – HAAP Media Ltd Announces Full Acquisition by Getty Images

Leveraging the Long Tail

Overview

The Last.fm Logo

The term “Long Tail” was popularised by Chris Anderson in his famous Wired magazine article in October 2004, where he demonstrated the concept by citing Amazon, Apple and Netflix as examples. The theory of the Long Tail in a nutshell is that we are experiencing a shift in our content consumption patterns from focusing on a relatively small number of mainstream products (also know as “hits”) towards the much larger market of niche items. In the physical world, shelf place has been historically physically constrained and expensive, therefore stocking only the best-selling products was the only economically viable alternative for a given store. However, it turns out that the Internet is an ideal marketplace for catering for the niche interest groups. Since there are no physical shelf-space limitations to speak of, it is just as easy to provide access to a huge number of niche products as to only the most popular, selected “hit” items. Online retailers, such as Amazon, can stock virtually everything, and these millions of narrow niches seem to better suit the individual interests than of the customers than the generalised best-sellers.

But not only online retailers can capitalise on the long tail paradigm. There are a growing number of services who are leveraging the power of the long trail in a number of unique ways, to cater not only for the mainstream audience in the head, but also to address the many millions in the tail. Last.fm is such a service that helps its users discover new music based on their listening habits, which is especially useful for individuals interested in obscure, hard-to-find niche genres.

Comparisons

Last.fm is a music service that helps its user discover new music based on their music listening habits. To quote a large recent study done by Yahoo on consumer preference data, “the vast majority of users are a little bit eccentric, consuming niche products at least some of the time”. Last.fm capitalises on this exact aspect; it offers individuals with niche musical interests the possibility to discover similar bands they might like and connect with people having the same musical tastes, while also catering for people interested in more popular forms of music.

The heart of the service is the so-called “Audioscrobbler” recommendation system, which collects data on the songs the user listens to, builds a detailed profile on each user’s musical profile based on the collected information, then uses sophisticated algorithmic methods to recommend similar music to the user. Audioscrobbler can collect data from a computer via plugins installed to the user’s music player, from Internet radio stations the user listens to, or directly from a portable music playing device. Apart from the extremely popular recommendation system, which uses collaborative filtering algorithms to discover new content suited to a particular user’s tastes, the service also provides social networking features, such as formation of groups based on a common interests, friends lists, manual music recommendation by users or groups and journals written about artists. What’s even more interesting, the recommendations system also sports a widget reminiscent of the long tail diagram itself, in which the user can drag a slider towards the head or the tail part of the chart to explore more popular or more obscure recommendations.

Potential issues

The Long Fail

The Long Fail

Not everybody, however, agrees on Chris Anderson’s theory of demand distribution in the Internet age. Some recent statistical data seem to support the views of the opponents of the Long Tail theory: out of the total 13 million of tracks available on the internet, 10 millions have never been purchased even once. Scientific analysis has shown that the sales distribution of online music retailer do not follow the power curve of the long tail curve, but rather exhibit normal logarithmic characteristics. Furthermore, social aspects and basic human behaviour have to be taken into consideration as well. According to Jonathan Karp—the founder of the highly successful book publishing company Twelve, whose objective is to focus on quality rather than quantity and publish no more than twelve books each year—”It’s been a truism among my colleagues that generally people want to be reading what other people are reading”.

Future directions

While the universal applicability of the Long Tail theory is being challenged currently, it goes without saying that online retailers such as Amazon and Apple have made quite sizeable profits due to their huge size of online catalogue which were inherently niche-buyer friendly. What me might see in the near future is a refinement of the theory and a more detailed understanding of the exact scenarios where it could be applied with great success.

References

About Last.fm
Wikipedia – Last.fm
Anatomy of the Long Tail: Ordinary People with Extraordinary Tastes
Last.fm user distribution, long tail of mainstream-ness
Last.fm: The Long Tail music picker
Music Recommendation and the Long Tail
Last.fm the Robotic DJ
Twelve Publishers – About Us – Mission Statement
Long Tails and Big Heads
The Long Tail Theory Gets Challenged, Just Not in Search Query Demand