Call center logs and warranty claims analysis

For most products, especially technologically complex ones and household appliances, calls are placed to service centers by consumers having difficulties. Sometimes these are simple issues and easily handled (e.g., read the manual) but other times they are complex and challenging. In almost all cases, service call centers create a detailed call center log of the call: time of call, the service representative on the call, the issue, the resolution, and any other notes on the call. Many are even recorded, sometimes with the statement “for training purposes only.” In all cases, these call center logs can provide a wealth of information and insight into issues that could lead to new product ideas. The logs are the same type of information as product reviews, but in a different form. The text analysis described above can be used to glean ideas for new products.

The same holds for warranty claims. Most, if not all, manufacturers issue (limited) warranties and some issue extended warranties, usually for a high price. Third-party providers also offer extended warranties. In essence, warranties are insurance policies against product failures within a reasonable period after the product is purchased, usually one year. The manufacturer clearly incurs a cost if the product fails so it has an incentive to make the product as fail proof as possible and also limit the coverage of the warranty.

Warranty data can be used to predict warranty costs from product failures and take action to fix the issue(s) causing the failure. The latter is important because too many failures would lead to a loss of customer support expressed in poor product reviews. These reviews, however, as discussed above, are a treasure trove of new ideas. Regardless, predicting failures is an important function. For this discussion, however, such predictions are not important. It is the use of warranty claims for new ideas that is important because the warranty claims should explain the nature of the failure.

There are two types of warranty claim analysis: a simple count of the number of claims per product (i.e., SKU) perhaps by time since purchase (e.g., 0-1 month, 1-2 months, etc.) and text analysis of problems and reasons for the claim (e.g., “the dishwasher does not drain”; “the oven temperature is not calibrated correctly”; “the smoke detector keeps beeping even with a battery.”). Thus, there are two types of warranty data: structured and unstructured. The warranty text data can be analyzed for clues for new products, perhaps as refinements of existing products, the same way product reviews can be used.

Sentiment analysis and opinion mining

An aspect of text data analysis that is more complex but that is gaining in importance is sentiment analysis and opinion mining. Sentiments are negative, neutral, or positive emotional statements that are made about an object. In our case of text data, the statements would be product reviews, call center logs, warranty claims, and so forth. Opinions are views expressed as pro or con statements about an object. Some analysts believe there is little to no conceptual difference between the two.15

Although sentiment analysis and opinion mining are important topics for product ideas, I will relegate discussion of them to Chapter 7 which is concerned with tracking the new product post-launch. Sentiments and opinions would be more important at the stage of the new product development process because they would tell you what is wrong - or right - about the product which may require further action. These actions could, of course, lead to another new product but here, nonetheless, I will restrict any discussion of sentiments and opinions to the tracking function recognizing that they could be used in the ideation phase as well.

Market research: voice of the customer (VOC)

A major theme of this chapter is that new product ideas can come from customers themselves. The market research and marketing literature have emphasized the need to hear “ The Voice of the Customer” for a long time. See Campos [2012] for an extensive discussion about the voice of the customer and ways to hear that voice. This section focuses on some ways to “hear” that voice using market research surveys.

< Prev   CONTENTS   Source   Next >