Republished with permission from Publishers Weekly.

This article originally appeared on August 31, 1998, page 26.

At the beginning of the 1990s, publishing appeared, compared to many other businesses, to be grossly unwieldy, unpredictable, unprofitable, out of control — in short, chaotic.

Here was an industry afflicted by huge advances that led to disappointing sales, skyrocketing returns, loss of vision, and loss of control in the distribution channel, which meant that some important trading partners were forced to wait as long as 180 days to be paid for their products.

Top management of the large publishing houses, many of whom came from other industries, felt that things must change.

One of the chief forces driving that opinion was the development of computer systems that promised Electronic Data Interchange (EDI): the exchange of completely electronic information among trading partners regarding job specifications, purchase orders, job status notifications, invoices, advance-ship notices, paper usage, and component inventory. Data about customer behavior, channel costs, production processes, and the like would flow easily and instantaneously. By tapping those rivers of data strategically, and analyzing their patterns, it would be possible for publishers to create new and better products, marketing, and channels of distribution, which in turn would increase sales, revenues, and profits.

Peter Olson, CEO of Random House, the largest publishing house in the world, was quoted in a U.S. News & World Report article: "We saw a need (five years ago) to have an integrated system that would provide, at the push of a button, information that would go throughout the company and be available to customers if necessary."

If that is the dream, how do we get there? The answer that major publishers are currently embracing is: re-engineering through the adoption of "enterprise-wide" software based on relational databases. Relational databases can be very powerful tools, and an enterprise-wide approach should generate the greatest strategic control and efficiency.

The relational databases with which most people are familiar are those checkbook programs that automatically add to your balance after you enter every deposit, subtract after you enter every check or ATM withdrawal, and keep your balance information always accurate, up-to-date and instantly available.

Can the hugely complex business of publishing, with all its facets of creation, production, and distribution, be similarly modeled? As Jonathan Newcomb, CEO of Simon & Schuster, explains: "If you don't understand where exactly you are in all your copyrights, how much you put into each project, and how much has to go back out for rights and permissions and advances and royalties, and how much you're going to invest in plant and paper and printing and binding, and what the likely returns are going to be from each project — and each project is different — if you can't track that incredibly complex matrix, then you aren't going to be able to be successful in this business.

Knight in Armor

Where is the white knight for the princesses of publishing who find themselves in dire economic straits? Who can solve so many problems and answer so many concerns?

The answer tried by a number of major publishing houses: a multi-billion-dollar German superstar software company called SAP. Since the early 1990s, SAP has been much touted by the business press. It has been portrayed as the darling of such corporations as Coca Cola, IBM, Microsoft, General Motors, and scores of other companies worldwide that have adopted SAP systems in the course of their efforts to make themselves more efficient, profitable, leaner, meaner, quicker-to-market, and customer savvy — in short, everything a company would want to be entering the new millennium.

SAP's premier product is a huge system called SAP R/3, a suite of interlocking software programs that, according to SAP's glossy descriptive materials, integrates an entire complex business cycle "into a workflow of business events and processes across departments and functional areas," as well as customers and suppliers. More specifically, the software elements include functions such as financial accounting, treasury, production planning, materials management, plant maintenance and service management, sales and distribution, and human-resource management.

In addition, "these systems automatically distill operational data into executive information" that permits "monitoring and controlling critical factors at all corporate levels." The real key is the promise that not only does the system offer efficient software to support every major corporate function, it also "overcomes the traditional hierarchical and function-oriented structures." In short, it brings the company together into a truly manageable whole.

"Re-engineering a business as complex as publishing is not going to be easy, cheap, or quick"

Happily Ever After?

And are the princesses of publishing being saved from their distress? Are they riding happily off into the millennium with this high-maintenance software hero? Not exactly, or perhaps more fairly, not quite yet. Like the knights of old, those computerized crusaders have caused a few new problems while they were changing the world.

For, in opening their companies to SAP, publishers also opened their purses to the tune of tens, and in some cases, hundreds of millions of dollars; and they subjected the patience of their employees to months (turning into years) of hard, frustrating work.

Re-engineering a business as complex as publishing with enterprise-wide systems is not going to be easy, cheap, or quick, no matter what. Change that big is always messy. As one executive commented about the experience of SAP implementation at his well-known publishing house, "This is so hard you cannot be ready enough."

Problems arise in adapting enterprise-wide software systems to any business, not just publishing. One industry analyst estimates that less than 10% of major systems projects are implemented successfully.

Horror stories abound: the former CEO and CIO of Oxford University Press were let go following a failed attempt at implementing SAP; bookstores were unable to order from one of the major companies during a SAP installation because the company's systems crashed for an entire week.

As is well known in the industry, Bantam Doubleday Dell faced significant hurdles with its SAP implementation. It was alleged that overruns brought the costs up to twice what was originally budgeted, and that there have been at least three major crashes. The first was for three weeks shortly after the system went "live;" there have been two one-week crashes since that time. BBD declined to comment for this article.

Seemed Like a Good Idea...

The essential business model of enterprise-wide systems is a sound one. Generally speaking, the SAP software is well adapted either to high-cost, low-volume sales (like computer systems), or to sales of large volumes of identical products sold rapidly (like Coke).

Unlike the automotive, computer hardware, and software industries, publishing is a business sector whose understanding and use of computers might generously be described as average. One executive who came to publishing from another industry for a SAP implementation noted that publishing is well behind other sectors, technologically speaking, in terms of management practices and structures.

Therefore, the sudden imposition of a huge computer-related, software-involved task on existing personnel has led to understandable breakdowns in the personnel and technological infrastructure at some houses.

Given the fact that SAP/R3 is generally imposed from on high, i.e., by the CEO or CFO, and that the financial systems are among the best in the SAP suite, early warnings from lower-level people that there appeared to be problems with other aspects of the publishing process, such as ordering, warehousing or distribution, were ignored until a significant breakdown occurred.

Despite the potential of the SAP/R3 software, there appears to have been overstatement in terms of the readiness of the software to deliver on all the requirements of the publishing business. There have even been rumblings of class-action suits based on assertions of fraud, that the product did not perform as claimed.

For a period of time, SAP was reportedly unwilling to make its software compatible with the many function-specific software programs still operating in publishing houses. Now, with its new R/3 Business Framework approach, SAP is inviting its customers to work with it in the design and development of software.

Now the Good News...

The good news is that progress is being made.

A whole new business has grown up, translation software, that allows existing software systems in a publishing company to interface with various components of the SAP/R3 or other enterprise-wide systems. Not only are those translators important in allowing a company to move one step at a time in implementation, they become essential for areas such as rights and royalties management. Fitting translation software to an existing system and making all the other changes so that a publisher's system can function smoothly and effectively with the SAP software remains, however, a time-consuming and expensive process.

Companies that have taken advantage of translators include Simon & Schuster, which built its Order and Shipment Information System (OASIS) because SAP didn't have a unit for that function. Later Houghton Mifflin and IDG Books Worldwide announced that they are installing REAL (pronounced Ray-al, Spanish for royal) Software's Alliant royalty-tracking system.

"Plan the work, and work the plan"

IDG's choice, according to director of distributed computing Mark Breslauer, has as much to do with who IDG is as with the software itself: "We're growing fast," he explains. "We needed royalty software to handle that growth and to provide the flexibility to handle a wide range of contracts."

Publishers might do well to keep an eye on the experience of other industries. As reported in a page 1 Wall Street Journal article last April: "The dark side of the computer-information revolution is coming into view. Companies across the country are seeing their cutting-edge computer systems fail to live up to expectations — or fail altogether  .  . . the backlash is beginning. Senior executives are starting to say 'No' to vast technology overhauls."

Appropriate Use

The key to implementing SAP or any large system, notes Chris Moschovitis, head of New York City-based publishing consultants The Moschovitis Group, is to recognize that the heart of the publishing process — editing, production, tracking, shipping, fulfillment — is much more complex than the accounting side. Therefore, he suggests that publishers first ask some enterprise-wide strategic questions:

  • What are your company's basic publishing goals?
  • Where does technology fit in? Editorial? Operations? Finances?
  • How much money can you afford to spend, in terms of people and infrastructure, on the process?

Once you have the answers to those questions, it's possible to look for technology solutions and partners that have a better chance to work to your advantage.

Simon & Schuster, for instance, has implemented a significant part of the SAP/R3 system with its eyes wide open to the experience at other firms. Moving very carefully, S&S reports success. The S&S approach was extremely deliberate, according to Michael Packer, executive vice president, strategy, technology, operations; and Gloria Samuels, senior vice president, technology, systems, operations. It involved a decision to focus on financials and to exclude functions such as warehousing, editorial, production, and royalty issues.

S&S's basic strategy, "plan the work, and work the plan," meant that the company made certain it had the right number of people with the right skills in place and prepared long before the implementation began. In addition, S&S took the trouble to get to know the people on the SAP side very well. The key to a successful process was the decision to do a phase-by-phase implementation, rather than a "big bang," taking one discrete function at a time "live," and lavishing it with attention. They moved instantly to redress the smallest problem, until they were certain the system was stable.

S&S financials went live in July 1997 and, according to Packer, is stable and functioning well, as is accounts receivable, which went live in March last year. Subsequently they tackled order processing for the K-12 division ("a big one"), followed by the balance of K-12 business processes. Other planned migrations include Prentice Hall Canada's order processing and accounts receivable, with higher education's order processing scheduled for later this year.

One of the surprises that S&S experienced in its relationship with SAP was the limitations of the SAP R/3 software's ability to handle publishing-specific tasks like backorders, returns processing, and transportation requirements. As a result, the software had to be modified to handle those functions, which added to the cost.

HarperCollins, too, has written a lot of its own software and grafted it to commercial packages, in particular the business packages from U.K.-based Vista Computer Services, which perform functions such as order entry, purchasing, inventory management, paper and raw materials tracking, and the like.

With regard to SAP, Allen Hoffman, manager of trading-partner relations, said that HarperCollins examined the potential two years ago and decided against that approach to enterprise-wide systems for two reasons: 1) the company was already doing so well with its distributors, and 2) the cost versus the benefits did not seem to make it worthwhile. Hoffman did allow that there are again some rumblings about the possibility of a fully integrated HarperCollins system. According to Information Week, HarperCollins has already built the biggest digital data warehouse in the industry, including complete data on years of sales history. That allows the house to provide decision support to management on almost any question that could be posed.

With regard to EDI, HarperCollins has implemented a book-ordering system based on software from St. Paul's Software, which, in their view, saves time and money and improves relationships with trading partners. In fact, Harper feels that in many ways the house is ahead of the industry in terms of the services it offers. For example, HarperCollins was the first to do advance shipping notices as part of its EDI package.

According to Hoffman, the EDI system has allowed HarperCollins to cut costs from $2-4 per order for paper-based ordering to less than a penny per order using EDI. Among the benefits are:

  • error reduction from reduced handling;
  • paperwork reduction (no more faxes);
  • reduced cycle times where the span from order creation to receipt of goods was dropped from three weeks to seven days; and
  • better business relations through a "valued partner" approach, in which the company works with each of its partners in the implementation of EDI, rather than promulgating procedures and demanding compliance.

Unquestionably, at some point as they enter the millennium, all publishers will have to move toward this brave new world of enterprise-wide systems. Hopefully, the experience of those who have boldly gone before will speed the process and smooth the ride.

As president of LIGHTSPEED, LLC, based in New York City, James Lichtenberg provides strategic counsel to clients in publishing, information technology, and higher education. His focus is the transformation of publishing and higher education due to the impact of new information technologies. He contributes regularly to Publishers Weekly on that issue. He is a former senior vice president at Hill and Knowlton, and was vice president of the Higher Education Division of the Association of American Publishers. His clients include Houghton Mifflin, Prudential, Nestlé, John Wiley, Harvard, Yale, The Chronicle of Higher Education, the Council on Family Health, AT&T, and DuPont. He was graduated from Harvard College with a B.A. in English literature, and he received a M.S. in sociology/socioeconomics from the New School for Social Research. You may contact him by e-mail at