Indexing Retrieve Documents100% of the Time22 Mar, 2010 By: Jim Thumma imageSource
Indexing Retrieve Documents100% of the Time
The old adage “seek and ye will find” would be great for workers if their
software would consistently deliver the information they need, where and when
they want it. Too often, the bridge between what’s needed and found is
incomplete, especially as organizations grow, the amount of stored information
increases, and more people are involved. A sea of documents, disparate legacy
and line-of-business systems, and the paths each document travels during the
business lifecycle makes some searches as efficient as sifting through dark file
folders in a dim basement while wearing sunglasses. Service, reputations and
profits can suffer irreparably when documents are demanded and can’t be located.
The challenges aren’t always the result of inferior technology. Often they
arise from poorly devised indexing plans.
Ideally, employees would determine what they need and have it delivered to
their cerebral desktops. Not only is this impossible, but security
considerations necessitate filtering what each user is (and is not) allowed to
access. Web-based enterprise content management (ECM), supported by an
enterprise approach to indexing, delivers information workers need to succeed,
unearthing relevant files within seconds. To ensure specific, meaningful, and
consistent search results rather than extensive, irrelevant lists of files (or
none at all), you should follow these basic indexing rules.
1. Understand that data—not documents—determines the success of ECM
Finding, flowing, securing and distributing documents are critical for any
business. All of these processes are driven by data within your organization,
and they are only achieved by leveraging the data effectively across your
enterprise. Since data is the axis that drives processes, knowledge, and
decision making within and across your organization, the indexing plan is the
single most important piece of the ECM puzzle.
2. Conduct a detailed document inventory.
Unless your company is a one-person operation, you probably don’t know the
ins and outs of every document type in your company or all of the people who
need to access its contents. Communication is a critical part of analyzing and
understanding the documents within your organization. Talk with IT staff,
managers, and end users from each department so you become aware of the business
value of each document type and its touch points within your business.
3. Create meaningful groupings of information to address common search
Although ECM doesn’t follow the color-coded file process used in paper-based
offices, it segments information in a similar way—only much faster, and
consistently following your established rules. Understand the types of
questions people will ask about your files. Know what kind of routine reports
are needed, and which data is drawn from specific document types. This will
help you to determine what to index. A university admissions office might index
by a unique student ID number, name, birth date, year admitted, area of study,
campus or college, and high school name or code. An insurer might index
policyholders by first name, last name, street address, document type, date
insured, claim number, and more. Separating each criterion into its own indexed
field is vital for quick, targeted, and easy retrieval.
4. Maintain indexing consistency across multiple departments.
In large organizations, multiple departments may use the same software
independently to build indexing schemes. When someone conducts an enterprise
search by a commonly used criterion, this can lead to complicated, ineffective
returns if departments use different fields to capture the same data. Having a
clear vision of your organization will help you create a plan that meets common
needs while allowing departments the flexibility to index additional data unique
to their area. Understanding proprietary information versus data that’s needed
by multiple areas is critical, but it’s often less obvious than one might think.
5. Streamline electronic filing with auto-indexing wherever possible.
Document indexing can be time consuming. Although indexing data or
interpreting questionable input typically involves humans, their involvement can
be streamlined by using automation. OCR, bar code capture, screen scraping from
other software applications, and other methods can be used to automate the
indexing process. The ability to take full advantage of these technologies,
however, typically is determined by the nature of the documents themselves, the
degree to which you understand your data repositories, and your access to the
Like most successful initiatives, indexing across the enterprise requires
methodical planning, meticulous analysis, collaboration, and solid
communication. It’s important to involve the right people from the start to
ensure buy-in to the project and acquire access to the details you will need to
make the project a success. Investing time up front to prime your employees to
embrace the solution and achieve maximum success is time well spent. With
solid indexing, your need for information can be fed with results rather than
being nourished by countless delays. A clear vision, an eye for detail, faith
in your team, and a solid solution will ensure that you seek AND find….every
Jim Thumma is Vice President of Sales & Marketing at Optical Image
Technology. This article is part of a five-part series called Enterprise ECM:
Strategies for Success. For more information on document imaging, document
management, and workflow products visit
www.docfinity.com. Thumma speaks on ECM at ITEX 2010.