Gain Power from your EDMS: Data Mining and Predictive Analytics – Part 1

Once your organization has implemented an electronic document storage solution (storing documents, paper images, and/or emails in a repository) what can you do with all this new-found data? Surely, there must be some sort of business value that can extracted from this huge pile of information that you’ve stored up, right?

This blog series will explore some of the values you can extract from your repository – the first being gaining market power and intelligence by performing data mining and predictive analytics on your document repository.

In order to data mine these documents, you will need to perform what is known as “text mining”, since most of your data will be unstructured data – that is, it will not be stored in separate organized columns in a database that could be easily access. Instead, your data is in free-form text in a document, such as a sales quote to a customer, a description of a customer complaint, or even in a legal briefing regarding a collection action.

Text mining can provide a business with benefits in a variety of ways. The most common is in the area of market intelligence, which is where companies mine their data to to drive better understand of their brands, brand awareness, product reputation, competitive position, and customer satisfaction.

Some industry examples are:

– Gaylord Hotels mined their customer satisfaction survey data and found that their biggest problems were not related to specific customer complaints, enabling them to create a real-time customer satisfaction dashboard

– Cablecom reduced customer churn by 12% by mining their text data

Using these techniques, your company can also perform fraud detection, risk modeling, legal compliance, and capture the knowledge that is in employee’s minds before it is lost.

In our next article, we’ll cover some of the methods being used to make text mining a reality.

Start typing and press Enter to search