The biggest tech challenge I faced and how I overcame it

Stable, cost-effective Web presence



The Web is the virtual face of many businesses today and, indeed, many businesses exist almost entirely on the Web. 

Our business, contests2win, falls in the latter category. This translates into a need for us to have a stable and reliable Web presence. It is a basic requirement for us. We need to be able to handle millions of transactions everyday–from millions of consumers from any part of the world to a

media-rich interface. And, of course, all this needs to be done at a reasonable cost.

Gopala Krishnan  Gopala Krishnan is COO of contests2win and CTO of mobile2win. He spends his time equally between Mumbai and Shanghai. You can reach him at gopal@c2w.com

In early 2000, consumers and, therefore, traffic started shooting up at contests2win.com. For me, to meet these demands was a challenge. We needed to architect, design and implement a robust, hugely scalable, transaction-handling system. We also had to ensure that the consumer experience was predictable and pleasant on a network we did not own (the Internet) and a consumer-computing environment we did not control! The plans of contests2win to expand to other regions, starting with China, and the prospect of our services being used by consumers on other media devices (like wireless devices), meant that our technology had to be multilingual and device agnostic.

Faced with these problems, I realized very quickly that the standard and widely accepted technology architectures or computing paradigms would not be suitable to achieve my objectives. 

The three-layered architecture with a client-server or distributed or centralized computing model would either fail or impose huge per-transaction costs that would threaten the viability of our business.

My first problem was to decide whether we should hire our own team to architect and implement the solution or to outsource the entire project to a third party. Outsourcing to a third party would have meant one of the large IT services companies in the country (one of the top six) or a focused Web technology company (of the many that existed at that time, most do not exist anymore).

Outsourcing seemed to be the way to go. Especially, as ours is a media and marketing services business. We were dependent on technology, but then, so are most other businesses! We did not want to lose focus from our core business by building and managing a technology division within the company. But, I was soon to be proved wrong. 

After extensive conversations with potential vendors (partners), presentations from many of them, and much analysis of the approach being taken by these various software services vendors, I was quite dejected. None of them seemed to have the experience of designing low-cost, high-transaction, systems for the Web. If they did, they had no ‘consumer perspective’ that made sense to us. They understood the corporate user but not the Web consumer. One or two of them, who seemed to have a somewhat reasonable understanding of both issues, proposed solutions that would have taken cost-per-transaction to stratospheric levels.

After losing a lot of time and some money, we decided that it was time to switch to an in-house model. I put together a team of technology people by mid-2000.

We devised a radically new architecture–something that still remains proprietary to contests2win (and now mobile2win China). Our architecture is a five-layered one as opposed to the prevalent three-layered model. It employs a mix of centralized and distributed architectures at the backend with a lot of the computing (in fact, most of it) happening at the client.

I am yet to see any major online business use the same methods we employed, and we are now in 2003!

The net result is that the contests2win technology platform today, provides one of the lowest costs per transaction in the world. It retains high

reliability, stability and is hugely scalable. We have gone a step further and made the core technology device agnostic and use the same engines to power our wireless brand marketing business in China, mobile2win. It is multilingual and operates in China in Mandarin.

To give you an idea of the kind of work that has been accomplished on the platform, I would take the example of the recent Coke wireless promotion we conducted for Coca Cola in China. One of the most successful wireless marketing campaigns in the world, it handled over 4 million consumer interactions, translating into a multiple of that number in terms of transactions and a much larger multiple of that number in database queries. The technology infrastructure, which handled that campaign, was capable of managing a 300% increase in transaction volume and could be easily scaled at short notice to handle several hundred times the volume. The cost advantage that our technology provides, enabled us to provide the solution to Coca Cola for a very reasonable price.

Technology, therefore, has become a major competitive advantage for contests2win. It acts as an entry barrier to competition, as it would cost them much more to do what we do. 

It is not possible for me to provide further detail of this because of space constraints. However, be sure to expect a white paper in the public domain on our learning and experiences, soon.