Today’s guest columnist is Michael Elkins, an independent consultant with over 17 years of experience helping global clients with the design and deployment of enterprise wide content management systems. His clients benefit from his ECM experience including strategic guidance, information architecture and ECM/data governance programs. His company, Kestral Group LLC is based in Denver, Colorado.
8 Things you need to know about developing an ECM Information Architecture
Information is our most important corporate asset, but the value of that information can only be realized if users can quickly find and use it. All too often, companies are handcuffed by numerous departmental or standalone content management systems, each with unique or incomplete information architectures. Getting to enterprise information architecture requires careful consideration during the design process including:
1. Requirements and not just technology should dictate the architecture.
Solutions developed in a vacuum are far more likely to fail. Users must be involved in the definition process as much as possible. At the end of the effort, the desired result is that users can intuitively find what they need quickly and easily. The best solution is not always out of the box from the ECM vendor.
2. Start with what you know.
Nearly a third of all companies have implemented Master Data Management (MDM) technology in order to gain control of their structured data sources. Unfortunately, few companies are utilizing that data to assist with the development of their ECM information architectures. In most cases, the ECM system will be integrated with other systems such as ERP, Customer Relationship Management (CRM), and other core business systems. Take advantage of existing folder structures to understand how users currently categorize content and look for any industry standards that may exist. In most cases, one will find existing industry standards for elements such as metadata, thesauri and taxonomies. Taking the time to map out data sources will provide a considerable cost savings both during development and ongoing support of the information architecture.
3. Don’t create something unwieldy.
It’s important to have a comprehensive information architecture that provides benefit to the company, but there can be a fine line between adding value and creating overhead. Users will gladly tag content if they know it will benefit them, but when tagging becomes a chore, the value is gone. The same can be said from a retrieval perspective. Even though an advanced search may help users narrow down their search results more quickly and effectively, many users still prefer the simplicity of a basic keyword search. Sometimes, less is indeed more.
4. Inheritance is a good thing.
So, how much is too much? The more that can be done for the user, the happier they will be and the better the data quality. There are a number of ways to default metadata values including information based upon user profiles or folder locations. All efforts should be made to limit the impact of tagging on the user.
5. Consider implementing a Thesaurus to improve search results.
Every industry, company and even discipline has its own language including acronyms and core terminology. Authors often revert to common acronyms without spelling them out. This simple act can affect a user’s ability to find the documents they are looking for based on the terms they search by. With mergers or acquisitions, the acquiring company’s terminology will tend to become dominant, and the acquired company’s legacy terminology will slowly fade away, often causing the documents from the acquired company to become “invisible” via the search process. Using a thesaurus can greatly improve the user’s ability to find the information they need.
6. Leave all options on the table.
Typing a term into a search box is not always the best path for a user to get what they need. I don’t know how many times I’ve heard that folders are not recommended or that they are outdated. In reality, there are situations where folders provide a very valid way for users to access information. Folders may offer an opportunity to default information which improves information quality and end user satisfaction. Multi-faceted taxonomies are not “out of the box” for most ECM vendors, but they do provide tremendous value for specific business applications. A strong information architecture should provide flexibility in how information is accessed.
7. Data governance is critical.
Developing the information architecture is not the end of the process, but rather just the beginning. The information architecture will surely change over time. Understanding what will change and controlling the impact of those changes is critical to the architecture’s long term viability. Data governance should be a core component of any ECM governance plan. From an information architect perspective, it is important to identify who is responsible for the various components of the architecture including core data sources, metadata standards, and taxonomies.
8. Think big and think portable.
All too often, ECM deployments are departmental in nature. As a result, only the implementing department’s needs are taking into consideration at the time of the design. Other departments may rely on the information and should be a part of the information architecture design process. For example, the Finance department may be implementing a solution, but Internal Audit department may benefit simply by being a part of the discussion to ensure that their needs are met as well. Beyond the multiple departments, companies need to think multiple systems. Many companies have more than one ECM system and, without consolidation, those systems should be aligned. Developing information architecture standards will drive improved access to information across all platforms by providing a more consistent user experience.
-----
Some other posts in the series...
Very good post !
Posted by: Martine laroche | March 03, 2010 at 07:35 AM
Great Read!
Posted by: Jim | May 28, 2010 at 01:27 PM