Document management systems provide users the capability to index documents using metadata attributes. These index attributes, when applied in a consistent fashion, can provide users with the capability to quickly find and retrieve associated documents. The question is… how do you enforce consistent metadata indexing for all of your documents? Relying on free-form text to index documents in an environment consisting of multiple locations and numerous users will not work.
A classic example is when a user is prompted to enter a business name. Even a household name such as Dell has numerous potential spellings, like: “Dell”, “Dell, Inc.”, “Dell Computing”, or “Dell Corporation”. In each of the representations, the good news is the name Dell is consistently included which allows for semi-efficient LIKE searches against the system. For more complex company names such as Pfizer, simply ensuring a portion of the name is entered correctly is a challenge. Finding 100% of the related documents requires ensuring the attributes used to represent the same entity are entered identically each time.
The easiest approach to ensure consistent metadata is to provide users with intelligent drop-downs that allow them to quickly select from a list of approved entries. There are interface designs that people use on a regular basis that utilize drop-down tools. For example, when purchasing online, often there is a drop-down list containing all of the abbreviations of the states/provinces in the United States and Canada. Living in Oregon, I have learned that if I press the ‘O’ key 3 times, Oregon usually gets selected and I can quickly move on to the next field. Entering metadata for documents should be a comparable experience. The key is to provide users with the ability to enter accurate data with the “least amount of strokes” – golfers should love that analogy!
So the next logical questions are “where does the list of allowable index variables come from?” and “how does it get maintained?” In the state abbreviations example, the available list is small and rarely changes making storing and maintaining the list a simple assignment. However, with most implementations the indexing requirements include much larger populations of dynamic data including customer information, document categories, and project attributes. For most of my customers, this data already exists in one or more of their computer systems such as the CRM system, the ERP system, and/or the project management system. By leveraging the existing business applications’ existing data, users performing the role of indexing documents can get intelligent, real-time dynamic drop-downs without maintaining redundant sets of data.
Ensuring that index information is consistent across the repository is only half of the equation. The remainder is to ensure that the indexing information can be applied to the document in the most efficient manner. To that end, a robust document management system provides the capability to automatically index a significant portion of the metadata information using what ColumbiaSoft refers to as “nested profile properties”. Imagine if the process of indexing documents required only entering a customer’s identification number and the remaining index attributes such as customer name, region, and account manager were automatically populated. By linking attributes using ODBC connections and applying business intelligence rules, the data can be quickly and accurately indexed allowing your entire company to efficiently manage documents.
Efficiently applying consistent index information to your documents using a document management system will improve your organization’s ability to find, leverage, and manage your critical business documents. The first thing I hear from recent document management converts is the remarkable time savings and efficiencies their organizations realized by having properly indexed their documents. For example, time-intensive activities like responding to customer inquiries are minimized to quick tasks, which frees up resources to work on core business activities.
