Roundtable Discussion: Enterprise content management

KMWorld recently hosted a roundtable discussion on enterprise content management (ECM) that included: Charlie Sodano, IS manager, Berlex Laboratories

; Lubor Ptacek, senior director of product marketing, Documentum; and Connie Moore, VP, Forrester Research . KMWorld senior writer Judith Lamont spoke with the group about enterprise content management and Berlex's application of it.

Lamont: Enterprise content management is a relatively new term. What does it include?

Moore: We use enterprise content management as an overarching term that describes a number of different technologies that up until recently have been seen as discrete markets. It includes document management, Web content management, records management, document imaging and digital asset management, among other things. We shifted away from using the term "content management," because the term generally was used to refer to Web content management only, and did not cover all the content in an enterprise. ECM encompasses all of the unstructured content in an organization.

Lamont: What action do you expect in this market over the next few years?

Moore: We think the records management sector is going to show explosive growth because publicly held companies must comply with the Sarbanes Oxley Act, and certain companies must comply with industry-specific regulations such as the Health Insurance Portability and Accountability Act or HIPAA. In fact, we expect an annual growth rate of about 160% through 2006. We think the ECM market as a whole will grow at more than 20% per year over the same time period.

Lamont: Why should enterprises consider having an ECM strategy?

Moore: Moving from point solutions toward an integrated ECM strategy allows an enterprise to make better use of its intellectual assets, and lower its overall costs by consolidating its IT infrastructure. Organizations using this approach also have more leverage with key vendors and can get lower prices than when they have multiple solutions in place. An overall goal organizations should have is to reduce the number of repositories in use for unstructured content.

Lamont: What got Berlex Laboratories started on ECM?

Sodano: My role at Berlex includes responsibility for the library, records management and more broadly, how to deal with electronic records of all kinds. We wanted to initiate an effort to establish electronic notebooks for research being carried out at our facility here in Richmond, California, which is a primary research site for our parent company in Germany, Shering AG, and make them available to employees at all of our sites.

Lamont: What kind of tasks are the scientists carrying out?

Sodano: They are carrying out several types of tasks in their research. They synthesize thousands of compounds a day using automation—as many compounds as possible. Then they screen them through appropriate biological systems to come up with a few good candidates that the company can invest in. The research information needs to be captured from an intellectual property point of view for patent applications, so we wanted the notebooks to provide a record of this discovery research.

Lamont: How were the research records kept previously?

Sodano: Scientists kept handwritten lab notebooks. Sometimes they entered data on the computer, but they would still print it out and paste it into the notebooks. In order to document a scientist's activities, we would have to copy the notebook. Of course, there were problems with legibility in the handwritten information, and sometimes the notebooks were lost or damaged. Our attorneys had to rely on interviews with the scientists to get information for patent applications, because the notebooks were inadequate.

Lamont: Why did Berlex select Documentum's ECM for this project?

Sodano: We were already using the Documentum platform for our FDA drug submission program, starting in 1997. Although we did survey other options, Documentum was the most logical choice because of our previous investment. Also, although we were starting with a relatively small pilot program in one facility, we knew we wanted to scale up, and Documentum could handle the expansion. The system also supports our requirements for paper archives. There is not yet enough litigation to validate the acceptability of electronic records in patent law, so we need to produce and archive paper copies. In 1998 we began developing the R&D management system, using Microsoft Office as the front end and Documentum's ECM to manage the information.

Lamont: How did the scientists react to the new system?

Sodano: We allowed a gradual migration by making use of the electronic notebooks voluntary over the first few years. By 2001, 80% of the scientists in our facility were using it. At that point, the electronic notebooks became mandatory. We rolled the system out globally, and in July of 2003 the electronic notebooks became mandatory for our facilities in Europe and Japan too. We now have about a thousand scientists using the system.

Lamont: What do they see when they log on?

Sodano: They have a template in either Microsoft Word or Microsoft Excel, and enter basic information about their search such as its title, key words, tracking information and other parameters. They still cut and paste computer-generated data, but now they do it electronically, rather than cutting up a printout. So it saves them a lot of time. Also, sharing information has become much easier. Content stored on the local servers is replicated to a central server on a daily basis, which allows access by people in our facilities worldwide.

Lamont: What can the scientists do now that they couldn't before?

Sodano: Information related to projects is legible, accessible and searchable. We have about 50,000 documents in the system now, and when people start a new project, they look around and see what information is available already. Sometimes they find out that people in adjacent laboratories are working on similar projects. Before, they never would have known. Documentum has some wonderful query tools that you can use, and people can get good ideas about how to improve their research methods. Extracting information for patents has also become much easier.

Lamont: How does Documentum view the evolving ECM market?

Ptacek: Connie made a very important point, that ECM is no longer just a grouping of discrete systems for the different content types. That is really a view we agree with, and we believe we have taken a leadership role. Having separate repositories for different content types and having enterprise processes in separate systems does not serve a company well. In real life, content types are very often mixed together. You may have a scanned document that is accompanied by a description in Microsoft Word, a number of pictures, and maybe some PowerPoint slides. They all need to be presented together

Lamont: What does this integrated approach provide?

Ptacek: In order to accomplish something like the example of the scientist's notebook, you need to have all the different content types on a single screen and in a single repository. This goes beyond the definition of ECM as just different content types being managed with a single solution, to a solution that allows you to span different departments and different processes in the company.

Lamont: How would different departments interact with this content?

Ptacek: If we think about the example we just heard, scientists might be working on the content at one end, but the regulatory people work on the same content. Then there are marketing people who are creating brochures and some legally binding descriptions that go on the packaging of the drug—they also need access to the very same content. So the content is then being refurbished throughout the corporation and that is kind of the new evolution of ECM, where we see that the processes are now kind of spanning the entire value chain of the content.

Lamont: How will the structured or database information intersect the unstructured content we are talking about here?

Ptacek: We will see a lot more user friendliness, for one thing. For example, we see this happening in the ERP space, which traditionally has dealt with very structured databases, such as part numbers and customer records. With each part number, though, there is a data sheet, and with each customer record there is a contract or purchase orders that reside in a content repository. We now see these kinds of applications embracing and integrating with the content applications.

Lamont: What is the role of XML in this integration?

Ptacek: XML is a means to this end. It's a technology that makes a lot of this much easier, because it's easier to standardize on. What was preventing widespread adoption was the lack of authoring tools, but now Microsoft and Adobe are basically using XML as the default format, which will help companies to embrace XML on a much bigger scale.

Sodano: The XML solution is where we should be going, and as Lubor pointed out, Microsoft and Adobe are going down that route anyway, which makes it easier for us to do some long-term planning for what we are going to do with these files. We expect to retain the look and feel of the documents in Microsoft and Adobe, and capture the metadata in XML. Subsequently that can be wrapped with a signature to make it not subject to change over a period of time. So I think things are going in the right direction, but it's still slow as far as having the tools readily available to do it easily.

Lamont: Do you see any technical barriers to ECM in bringing the various content types into a single application and functional system?

Moore: I think it's more a question of time than difficulty. You do get into issues of performance in terms of retrieving certain content types, like video or high volumes of images, when implementing a universal repository. We believe dedicated repositories for high volumes of images and video will still be needed, but companies should strive for a universal repository as much as possible. But more importantly, at this stage of enterprise content management, companies that have been focused on a single content type are buying other products that have different repositories and workflow software modules that have to be rationalized. Enterprise content management today isn't what enterprise content management is going to be in two years. Right now, the ECM market is in an acquisition phase; in two years it will be in a rationalization phase as vendors eliminate duplicate functionality in their ECM offerings.

Lamont: What specific developments does Documentum expect in ECM over the next few years?

Ptacek: We see a clear trend that customers are standardizing their content-based applications across the enterprise, just as they are standardizing any other technology. The IT consolidation and the need to easily repurpose and synergize across the enterprise are drivers for this. We will see more of an integrated suite approach between applications so that things like records management will need to be an inherent part of any application that deals with content.

Lamont: Is there one mistake that end-user companies are making as they move into ECM?

Moore: I think it's a question of who is driving the investment and the decision. Sometimes the line of business is going to see things more narrowly, and the CIO more broadly in terms of what the total organizational needs are for content. If a decision is based primarily on what one department needs, the resulting system might not work for the organization as a whole.

Sodano: This brings us to another issue of records management in general. Most companies really haven't completely addressed this issue. We had this with paper, but with electronic records it's going to get even worse unless people have a better handle on what they are saving short term. Long term they are going to have a difficult time digging out of the mess after you have terabytes of information. I know many people haven't seriously thought about that, and are still just backing up their information as a way of protecting it. But it deserves a lot more attention because in the future, if we are going to rely on electronic records we are creating now, we need to pay attention to the format, the media and the software we can use in the future to retrieve and work on the documents.

Lamont: From the vendor's viewpoint, how do you see this market evolving?

Ptacek: We think that the movement of IBM and Microsoft into this space is a clear sign of its importance. A lot of vendors are going into acquisitions now, just as Documentum was acquired by EMC, so we see a consolidation of this space. Big vendors are moving in, and that is very important because the smaller vendors will be affected.

Lamont: Does Forrester concur with this assessment?

Moore: Yes. We expect significant consolidation over the next 24 months. Infrastructure companies such as IBM will be competing intensely with the independent ECM vendors. Prospective users should consider vendor viability as an important factor in their selection, especially in the case of smaller vendors. Users should look for a product roadmap from vendors involved in acquisitions to see how they plan to integrate the disparate content technologies into their product offerings.

About Berlex Laboratories

Berlex Laboratories, a subsidiary of Schering AG, develops specialty medicines such as cardiovascular, neurological, dermatological, oncological and fertility control products. It also has introduced several contrast agents for use in magnetic resonance imaging and X-ray procedures. The company has facilities in California, New Jersey and Washington.