Review: Designing Systems for Internet Commerce

Small companies can purchase off-the-shelf store-building software and rather rapidly open up a retail shop. But if your company is larger, or conducting business-to-business sales, designing your Internet sales platform and integrating it into the rest of your company's systems requires a great deal more thought and planning. This book is designed for a company's "Internet commerce team ... both those involved in the business end of planning as well as those responsible for technology," say the authors. They discuss the business strategies commonly overlooked by techies, and the technical issues commonly bypassed by business planners.

Treese and Stewart, both software developers at Open Market, Inc., define Internet commerce as "the use of the global Internet for purchase and sale of goods and services, including service and support after the sale," with a focus on enabling the complete business transaction online. The style is what you'd expect from system engineers -- logical, meticulous, plodding, but it's not a book you can skim. I found myself reading carefully, challenged by the authors' new ways of looking at the process of Internet sales.

Strategy Development

The first third of the book is devoted to the business end of Internet commerce, drawing upon the ideas from Open Market CEO Shikhar Ghosh. The Internet, the authors contend, offers two key sources of value: the ability (1) to transform customer relationships and (2) to displace traditional sources of business value. On the Internet, customers can now choose the hours of business they desire, receive service at their location, serve themselves, and receive focus on their particular needs. The source of value moves from physical products (atoms) to digital products (bits). The Internet is able to deliver information in digital form, mass customized for each user, easily assimilated, economically distributed, and global in its reach. Now all this may seem a bit cerebral, but stay with it for a moment. Based on these two sources of value, the authors outline four key competitive Internet strategies that a company might adopt:

1. Channel Master strategy opens efficient new Internet channels to customers, providing the best possible delivery of products and related services (example: Cisco).
2. Customer Magnet strategy attracts customers over the Internet through meeting knowledge needs, and then providing ready access to online sales (example: Tripod).
3. Value Chain Pirate strategy uses the Internet to bypass existing distribution chains, offering a more efficient buyer/supplier relationship (example: ONSALE).
4. Digital Distributor strategy unbundles traditional groupings of products and then rebundles some of them for efficient delivery by means of the Internet (example: Classifieds2000).

An online business that offers a combination of these strategies becomes an even more potent competitor.

Much of ensuing discussion is couched in terms of the Commerce Value Chain: (1) attract (advertising and marketing), (2) interact (catalog sales), (3) act (order capture, payment, and fulfillment), and (4) react (customer service and order tracking). Using this framework, the authors examine in some detail customer retail, business-to-business cataloging, and information commerce. Developers and consultants will find this especially helpful in working with clients to adapt off-the-shelf or custom-programmed store-building software to meet precise needs.

Integrating Technologies

The second two-thirds of the book focuses on the technology of Internet commerce. The authors come with very strong credentials in the technical area, though they patiently explain the basic technologies of the Internet in terms business people can grasp.

In a chapter on System Design, the authors outline their own philosophy of design: (1) understand the customer's requirements, (2) plan for evolution to incorporate technology changes, (3) start small and build the system incrementally to take advantage of improved technologies as well as a better understanding of business needs, (4) keep options open, so that new modules may added in the future without requiring a complete system redesign, and, finally, (5) develop an architecture, a framework for solving individual design problems. Then they outline typical design issues: scaling up system performance by an order of magnitude; ensuring reliability; furnishing atomicity, consistency, isolation, and durability of transactions; managing "state;" and providing security.

I found the authors' outline of overall security design especially helpful, as well as their explanation of various kinds of attacks to which commerce systems are vulnerable, ways to thwart these attacks, and a brief discussion of how firewalls operate. The chapter on payment systems is also quite thorough. Not only are US retail payment systems covered, but also requirements for international retail as well as business-to-business transactions. The book concludes with the Open Market platform as an example of how to integrate all of the factors into a single system.

This volume provides the most comprehensive overview of Internet commerce I've seen to date. It will be especially helpful to larger companies preparing to enter the Internet marketplace, along with consultants and development firms.