Sharing structured data

XML Magazine

Subscribe to XML Magazine: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get XML Magazine: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


XML Authors: Jayaram Krishnaswamy, Chris Pollach, Jason Bloomberg, Peter Silva, Mehdi Daoudi

Related Topics: Yacht Charter Blog, XML Magazine, ERP Journal on Ulitzer

Yacht Charters: Article

Business Process Reengineering Through Web Services

Business Process Reengineering Through Web Services

Promise:Web services will be revolutionary!
Reality: What's a Web service for?
What's the investment case?
Problem: Most discussions of new solutions based on Web-services architectures focus on them from an IT-centric perspective
Solution: An alternative view that helps to envision new solutions and prioritize investment in them starts with understanding business processes end-to-end from the perspective of user experiences. Specifically:

  • Are users' choices usefully expanded?
  • Is the aggregation of information automated sufficiently to save valuable time?
  • Do integration savings get channeled into features, content enhancement, and design improvements that raise user adoption and retention?
A Better Noise Filter
Web services, XML, UDDI, SOAP, WSDL, .NET. Today it seems that anyone even thinking of investing in technology to solve a business problem is immediately overwhelmed by a cacophony of alphabet soup-like acronyms with the common promise of... well, just what isn't entirely clear. But "this is going to be really big, so you have no choice but to pay attention."

Many clients have asked us for help in making sense of these new technologies. They ask, "What possibilities - both opportunities and threats - do these technologies imply for us?", and "What should we do about them?".

We think about questions such as these "outside-in." To what business process will a new technology be applied? How will it change the experiences of the various users who are involved in the business process? What economic impact could those changes have?

Web Services for Business People
Just what the world needs: another Web services primer! But unfortunately, to imagine what a new technology makes possible in any original way requires at least a basic understanding of how it works.

Consider this analogy:

The Human Experience: You are a tax accountant who specializes in the Andorran tax code. You need a job. Your friend helps you write a summary of your experience in English using a conventional resumé format and posts it on Monster.com. With luck, an employer browses the site, you get a phone call or email, you're hired, and you have a long and lucrative career.

The Machine Experience: You are an Andorran tax accounting program. Your authors write what you do in a programming language, following a protocol or format called SOAP, using a conventional structure called "WSDL (Web Services Description Language)." They post you on a "UDDI (Universal Description, Discovery, and Integration) registry." With luck, a financial accounting package that doesn't calculate Andorran taxes itself browses this registry, identifies you as a relevant "Web service," and you get a "Remote Procedure Call" or "RPC" (itself expressed in SOAP). You return a valuable answer, and earn your subscription fee or "micropayment."

Continuing this analogy, if your authors are Microsoft customers they will use tools such as Visual Studio .NET to write in Visual Basic just as Microsoft Word is used to write English. SOAP is a protocol - an agreed-upon format for transmitting data between a Web service and the application that would like to use it (in this case, the financial accounting package). Microsoft BizTalk servers are the traffic cops that log incoming RPCs/requests and route them to appropriate Web services.

The bottom line: With the proliferation of valuable content and applications made possible by the Web, there is a huge opportunity associated with automating the discovery and exchange of those assets. Just as people use tools and conventions such as search engines and "About Us" pages, machines need conventions to find each other and communicate effectively. A Web service is thus a computer program exposed to and used by other computer programs across the Web using pre-agreed conventions as its lingua franca.

Great! Now What?
Okay, we now understand there is a commonly agreed way to get programs talking to other programs over the Internet. Which programs and why?

People and programs work together toward various objectives through "business processes." For example:

  • A customer interacts with salespeople and/or a Web site to identify and purchase a product to meet a particular need.
  • A manufacturer interacts with various suppliers to line up various resources.
  • A channel partner checks on inventory to make sure it can fulfill orders it has taken on behalf of the manufacturer.

    Each of these processes typically encompasses several major activities. For example, selling something involves:

  • For a successful sale, a customer first has to perceive a need, get educated about product options, order and receive the chosen product, and use it successfully so it won't be returned.
  • To acquire the inputs it needs to produce a product, a manufacturer first must identify, screen, engage, and transact with its suppliers.
  • Similarly, a channel partner must qualify itself into the manufacturer's distribution program (or vice versa), and then get "hooked up" appropriately before it can view inventories. The partner may even need to attain certain performance levels before it will receive allocations of popular but scarce products.

    Each of these activities is both enabled and constrained in certain ways by the technologies applied to it. Consider these three examples:

  • Our customer may have received a CD from a financial planning software company with a "configurator" application to help him think through his needs and determine which of the company's offerings might best fit those needs. Let's assume that unfortunately the customer left the CD at home and he's on vacation. There happens to be a new Web-based version of the program that he could use available at a nearby Internet café, but the program requires up-to-date information from various financial services providers, and is not yet able to reach across the Web to get the information directly from them.
  • Our manufacturer had joined a consortium to sponsor and exchange information and resources, and spent many millions implementing the software and recruiting suppliers. Unfortunately, adverse selection lowered the quality of the supplier pool, and the exchange never aggregated enough demand to attract the best suppliers in the market. Now our manufacturer needs to create closer ties with his preferred vendors, but must find an inexpensive and nonproprietary way to connect with them.
  • Our value-added reseller combines products from multiple plants in our manufacturer's network to create solutions for her customers. Tracking inventories online allows her to do some "critical path planning" so she doesn't take assets onto her loading docks and balance sheets in advance of the full solution being ready to roll. Unfortunately, the manufacturer's plants, brought together through many acquisitions over the years, have implemented EDI links to their distributors inconsistently and at great cost to all parties. The result is that the process of managing a "bill of materials" for the final shipment is manually intensive and error-prone.

    A "Web services" approach to the supporting technologies for each of these examples could change the user experience both within each activity and across the entire process. For example:

  • If the financial services vendors' applications were to be made available as Web services, the Web application that our customer visits from the mythical Internet café could determine the impact that day-to-day changes in his investment portfolio would have on the length and amenities of the yacht he can afford to buy.
  • By exposing elements of their respective supply-chain planning platforms as Web services based on neutral protocols, our manufacturer and its suppliers can exchange demand forecasts and inventory information automatically, without committing to expensive and proprietary formats (assuming their industry already has an agreed-upon XML "DTD" or dictionary that defines which tags stand for what data elements).
  • Just as our manufacturer could work with its suppliers, our reseller can work with plants from which it sources parts to "wrap" and "expose" inventory applications as XML-based Web services that the reseller's tracking applications can "consume." If the reseller's own supply network comes up short, it can "browse" UDDI registries for other vendors that meet its needs.

    What is the value of these solutions? First-order effects include:

  • Lower integration costs: A good rule of thumb from the ERP implementation world of the early '90s was that integration costs would run 3x original license costs. In the more significantly networked Internet world of the early 21st century, some top analysts estimate that this multiple has increased to 10x. Since not everyone can afford 10x, a lot less integration will take place unless an alternative, less-expensive approach can be found. Enter Web services, based on a lingua franca without proprietary technology requirements.
  • Expanded choice: Automated search increases the likelihood of finding an alternative source of supply, distribution, or information.
  • Increased service performance throughout a given business process: Because supply, distribution, and information options have increased, waiting time for any item in a business process "queue" will decline.

    Second-order effects are equally powerful:

  • Employees are freed from manual and mechanical search and aggregation tasks and can focus on service and innovation.
  • Customers experience greater choice, better and faster service, and lower costs when they conduct business with vendors whose infrastructure is enabled through Web services.

    From Here to Where?
    How do we go about realizing this glorious future? There are three questions to be answered:

    1.   What business process offers the greatest return on investment from improvements made to it?

    2.   What specific change in a given business process drives the most improvement in performance?

    3.   Is a fix based on Web services the most appropriate solution?

    Answering the first question requires some strategic research, such as:

  • Modeling the value of improvements in performance to customers (or other consumers of the outputs of a given business process)
  • Benchmarking how such improvements would position your organization relative to competitors

    Good answers for the second question can be found through basic business process analysis and redesign:

  • Mapping steps in a process, and associated performance metrics (time, cost)
  • Understanding the resources (material, human, technical) employed at each step

    As for the third question - "Are Web services the answer?" - it makes sense to look beyond what you do see to what you don't:

  • What valuable information that might help answer customer questions or issues is not provided due to cost or availability limitations?
  • What segments of suppliers or distributors are not included in current networks due to cost or technology?

    Practical Magic
    Web services are subject to network effects. The more of them that exist, the more valuable they are, and the challenge is that we are just at the dawn of the Web services era. Since the use of Web services is a fairly new concept, it is also still relatively expensive to implement. And since the pool of available Web services and users is still nascent, the return on the initial investment is limited.

    Fortunately, major information technology infrastructure vendors understand these issues and are lowering the cost and complexity barriers to getting started. Major telecom vendors are getting into the UDDI registry business - essentially Yellow Pages for computers. Early pricing is quite favorable. Vendors such as Microsoft and Sun Microsystems have developed frameworks for building and deploying applications so that the last step - exposing them as Web services that can be consumed across a network - occurs automatically. Microsoft's .NET family of products is probably the best example of this.

    What tools you use depends a lot on where you start. If you're already a Microsoft shop, you'll likely want to stay that way in the .NET world. If you're not, you will need to weigh the advantages of Microsoft's approach versus other vendors such as Sun or IBM, taking into account the cost of required infrastructure changes. The good news is that the steep performance improvement "curves" in the IT business drive significant cost-improvement opportunities in infrastructure over relatively short time horizons. Therefore, such infrastructure transformations are not only realistic but also often attractive.

    Parting Thoughts
    You've run focus groups; customers tell you they love the improvements you're proposing. You've benchmarked: none of your competitors would, or could, get close. You've mapped processes and found the "Herbie," or the bottleneck.

    You've architected a cutting-edge, Web services-based solution.

    On paper it sounds brilliant. But as we have all discovered, in practice there is a lot more to "getting the horse to drink." Chief among these factors is usability, both in the narrow sense of a tool itself, and viewed against a backdrop that reveals how well the tool would fit in a person's everyday life.

    In short, don't forget that while improving the user experience may start with major performance gains through Web services-enabled business process redesign, it's unlikely to end there. A smart investment program will consider ancillary usability issues to maximize available returns.

  • More Stories By Andrew Bibby

    As vice president of client services and alliances, Andrew’s responsibilities ensure that value is driven from major alliances throughout the client services organization. He has 16 years of experience developing and implementing both project and relationship management methodologies in the professional services industry. He possesses deep understanding of how to organize and prioritize programs of activities around real business needs and how to work within the political landscape of large complex corporations to ensure success at every level.
    Andrew ran the Information Technology side of the mutual fund and trust operations for the Bank of Bermuda, which exposed him to high-level global projects with complex implementation schedules and rollout plans. He began his career in the financial futures and options area with PA Consulting Group.

    Comments (0)

    Share your thoughts on this story.

    Add your comment
    You must be signed in to add a comment. Sign-in | Register

    In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.