Time Warner is one of the world€™s leading media and entertainment companies. Time Inc. publishes over 130 magazines including Time, People, Sports Illustrated, Entertainment Weekly, Southern Living, In Style, Fortune, What€™s On TV and Money. The Publishing division significantly expanded its international reach with the 2001 acquisition of IPC Media Limited, the United Kingdom€™s leading consumer magazine publisher.

Sourcing paper, a critical resource to produce magazines, has been moving away from domestic sources and is becoming increasingly international in scope. Domestically, Time Inc. prints over 1.5 billion paper copies annually. Over 500,000 tons of paper are purchased annually. Time Inc. must collaborate with paper suppliers and printers to be sure that the correct paper for all magazines is available at the right printer at the right time. Thirty paper supplier mills, including the larger paper suppliers like Stora Enso, International Paper, UPM Kymmene, and MeadWestvaco, produce the bulk of Time Inc. paper. This paper is shipped to over twenty global printing locations including RR Donnelley, Quad Graphics, Quebecor World and Brown Printing. At these printing locations paper is received, inventoried, and used.

The Publication Papers Supply Chain Figure 1 calls for close collaboration of paper companies, publishers or cataloguers, and printers.

TimeXchange, a Time Warner system, was built to facilitate effective, collaborative international paper sourcing. Our system is an innovative XML standards-based system for the purchase and inventorying of paper. It is a multi-currency, multi-metric inventory system that manages the paper inventory. It leverages an industry leading implementation of papiNet, an XML exchange standard for the paper industry. Its Java web based design for transactions and reports supports collaborative supply chain access by magazines, paper suppliers, and printers. It provides accurate projected inventory to support timely paper ordering.

Collaborative Systems have clear business impact and value. By increasing inventory visibility and facilitating standards-based paper purchasing, our system streamlines the paper buying and inventory processes. It allows the publisher to buy paper just in time with a full set of current information.

Building new systems like TimeXchange while participating in the design of new standards like papiNet work well together.

Designing the domain of a standard and the scope of a system are complementary. For example the database constructs for paper and the XML standard to detail paper characteristics interact nicely. Internationalization of the domain of a problem, fostered by a representative gathering of publishers, suppliers and printers, generalizes a solution. Be it multi-metrics, multi-currencies or varying financial requirements for invoicing, an international standard implemented well enforces a flexibility of design. A vertical standard helps to develop a common vocabulary. For example, papiNet uses British spelling, €˜colour€™, not color and British usage, a paper €˜reel€™, not a roll. A common vocabulary informs system design.

For our development a vertical standard of sufficient scope like papiNet acted as a €concrete abstraction€?that demonstrates what our collaborative system has to be able to handle. It is not a €œcomplete abstraction€?or meta-data that can be data about anything, but neither is it a narrowly defined specification that only focuses on certain paper product lines. It is concrete enough to build a data model that is resilient without being so abstracted that it cannot be implemented. As an example our system allows for flexible paper characteristics that allow any kind of paper to be ordered as a €managed product.€?This allows the system to be used for reels, for sheet paper, for coated and uncoated paper, containerboard, billboard sheets, etc. Our system implements the vertical domain of the standard, paper as defined in papiNet; it is not a Rosetta-net based system to be used for inventorying computers.

By overlapping system design and XML design, each activity informs the other. Database design interacts with message design. Transaction flows interact with XML messages. Collaborative systems at the publisher, printer and supplier use XML as the glue to combine disparate systems.

XML standards and their implementation are key opportunities to refine and generalize your underlying systems. Time Inc. was fortunate in that we were committed to rebuilding our paper inventory system. Making XML the core of this redesign made sense. Many of the printers and suppliers, grown by acquisition, have multiple legacy systems that need integration. Some other publishers, printers, or suppliers were committed to an ERP system approach for paper processing. Regardless of approach implementing XML standards is an opportunity point to consistently talk papiNet XML internally and externally. XML makes you think of the total system regardless of boundaries and inhibitors. XML, unlike EDI, is a potential solution to many internal and external integration problems.

XML standards efforts offer invaluable insights into varying business practices that give you models for how to change your supply practices. For example the paper packaging industry works more on a replenishment model with automatic reordering and delivery whenever paper stocks get to a certain level. European publishers like News International use a method where they store less paper at the printer but call off paper stock from the supplier when it is needed. Discussing these practices and designing XML messages for them promotes utilizing system development as an opportunity for effective workflow reengineering.

There are a few best practices that helped us be successful.

It is important that the XML standards team and of course the system designers have a high degree of continuity. Many standards efforts flounder because they are not representative, are too large, take too long, and incur frequent turnover. The vision skews before it can be fully attained. PapiNet was fortunate in that it had good representation from publishers, printers and suppliers and a remarkable degree of continuity. Our project at Time was also fortunate in that we had dedicated resources, standards savvy, to build a system and refine a standard concurrently.

It is critical to actively engage with trading partners throughout all phases of a project. Publishers, printers and paper suppliers need to develop a common understanding of what the standard means and how it is to be implemented. Trading partner buy in, trust and transparency are very important. For example, our joint detail test plans with suppliers like Stora Enso and printers like Quad Graphics and Brown Printing included initial message exchange, specific situation testing, and parallel testing; all testing had to be completed before implementation. We continue to work with European and North American paper teams on a common understanding of how we will all implement each standard message. The strength of a vertical standard like papiNet is that it is flexible but consistent and that takes an attention to detail.

Recognize that implementation experience is critical to refine XML design and system design. We began implementing XML messages starting with the Purchase Order with our collaborative supplier shortly after their design. We found that standards are often conceptual until you go through the work of implementing them with trading partners. As we completed the papiNet message design and implemented more messages and more trading partners we get better at seeing problems up front but nothing beats implementation to flesh out the reality of the XML message or the system.

Just a couple of caveats.

Don€™t rush to extend the standard or your system into related markets or processes until you have solid traction. At papiNet we started with publication papers, packaging, and fine papers. We have begun to expand the standard into related areas including book components, wood products, label stock and recovered paper. Expanding the domain of the problem yields many improvements, but it is at the risk of losing focus. We have developed better design principles that changes to our common elements must expand the concept of the element and be upwardly compatible from our existing standard. You must limit the number of releases of a standard and guarantee whenever possible that changes are upwardly compatible. A too active standard discourages implementation.

Leading a standards group has it benefits and its drawbacks. If you are facilitating a group, you can€™t always get your way. As an example in the 2.0 version of our standard we changed a key construct to more clearly denote how deliveries were packaged. I disagreed with the change but being the facilitator could not as actively argue my point. Collaborative standards development works but you must always remember that collaboration is not always getting your way.