Documents can be stored for:

• Retrieving them for business decision making, or
• Complying with statutory regulations, or
• Creating a historical archive, or
• Going along with general practice or personal preference.

The last of the reasons mentioned above is not a particularly valid reason in the case of businesses. Businesses have to incur costs storing documents and administering the stored documents. Hence they must be clear why documents are being stored.

When Do We Retrieve Documents?

Documents are retrieved for such purposes as:

• Referring to them to understand the background before taking some action or decision. A supplier bill might thus be retrieved before a payment is made, for example
• Identifying meaningful trends and patterns by analyzing the information in the retrieved documents. Such trends and patterns can be valuable decision support information
• Satisfying some regulatory authority about compliance with law, or about the volume of turnover, etc.
• Writing the history of the business, or extracting some historical information that can be used for marketing or brand-building purposes

It will be noticed that the need to retrieve is a universal requirement whatever be the original purpose that prompted storing a document.

The more efficiently this retrieval need can be met, the better the purposes can be served. This includes speeding up business processes so that higher sales volumes can be achieved. Let us look at how document retrieval is improved.

How Do We Improve Document Retrieval?

In the paper documents era, document retrieval was sought to be improved by decentralizing document storage whenever possible, by using special storage devices such as index card holder and by implementing such practices as keeping current documents in easily accessible storage. For example, the accounting department might have maintained an ‘Unpaid Bills’ file keeping all bills till they are paid (and then remove them to the filing section).

In today’s digital age, document retrieval methods are being constantly improved. Random access disks that could access any file extremely faster replaced magnetic tapes that had to be read serially to locate a particular document. Tagged indexing where each document had to be tagged with a unique id can be replaced by full text indexing that enables retrieval of documents in a more flexible manner.

We mentioned about full text indexing above. Full text indexing is not always the ideal solution. It is best used when the documents in the storage are diverse and unstructured in content. Tagging them with unique tags might also prove not worth the cost. In such cases, full text indexing enables looking for any document by words and phrases in its content.

However, we have to pay a price for this convenience in the form of a big index file. It is the index file that enables quick retrieval by linking all those words with relevant files. It can become a huge document when the volume of documents is large.

Document Retrieval Elements

Documents are made quickly recoverable through:

• Using the best storage media possible. For example, paper documents can be scanned into digital documents and stored in computer media that improves retrievability dramatically
• Using a scheme of classification and indexing that is in line with the ways people tend to look for documents
• Using unique ids wherever possible and providing a lookup table or other device to enable users narrow down to the desired id
• Training users to adopt the best practices for searching.