by W H Inmon, Forest Rim Technology

Warranties are a fact of life, especially in the world of large retail purchases. Customers increase their confidence when there is a warranty.

In some cases customers will not even consider making a purchase unless there is a warranty. There are entire lines of products that could not effectively be sold without warranties.

But honoring the warranty becomes a liability for the manufacturing corporation. The cost of repair, the cost of replacement, and even the cost of servicing the warranty is an issue that the manufacturer must deal with.

For some organizations the volume of products and the ensuing number of warranties does not create a problem. But when the volume of products sold is high, and where the products serviced under a warranty is voluminous, the processing of the warranty becomes an issue.

Some companies can afford to process their warranty claims manually. But when the volume of warranties mounts, the manual processing of warranties simply becomes its own burden. For a large, modern company, processing warranties in an automated fashion is a superior choice over trying to process warranties manually.

The challenge with warranties is that most of the processing of warranties is done in the form of text. Text and narrative make up most of the important parts of warranties. And computers have traditionally had a hard time dealing with text and narrative.

The textual component of warranties often includes such information as what went wrong with the product, when it went wrong, why it went wrong, and other descriptive and diagnostic descriptions. In short it is normal to have narrative as part of a warranty.

Traditionally computers did not deal with narrative. Traditionally there was structured data that could be automated. But when it came to narrative, the computer was not of much help. And unfortunately much of the warranty was narrative.

But now there is Textual ETL by Forest Rim Technology. With Textual ETL organizations can read and process text and narrative in an automated manner. With Textual ETL organizations can read and process much of the important information in warranties.

Textual ETL reads unstructured data and places it in a structured, data base format. Textual ETL can place its output into any standard data base format such as Oracle, Teradata, DB2, SQL Server and Hadoop. Once the textual data is placed into a structured format, it can be read and processed by standard analytical software such as Tableau, Qlikview, SAS, Cognos, and others.

At a very high level Textual ETL sets the stage for analytical processing that would otherwise not be able to be done. Stated differently, without Textual ETL it would not be possible to do analytical processing on textual and narrative data.

And there is real value in being able to do analytical processing against warranty data. By analyzing warranty data, organizations can determine how to improve their manufacturing process. By analyzing warranty data, organizations quickly discover the weak points in their manufacturing process. And in the long run, this is some of the most important information a manufacturing organization can get.

But there are other possibilities for the automated processing of the narrative found in warranties. One of those possibilities is to use textual ETL as a front end interface for processing the warranty data, and then allowing the data to be processed by an application. In doing so, such questions as what warranty was in place, was the product still under warranty, what defect has been discovered, and so forth can be processed in an automated manner.


Forest Rim Technology was formed by Bill Inmon in order to provide technology to bridge the gap between structured and unstructured data. Forest Rim Technology is located in Castle Rock, Colorado.