Skip to main content

Identifying enterprises' top causes of data errors

Paul Newman

June 13, 2014

Archive

It's sad but true: When you're working with large volumes of customer data for improving your operations, and you're trying to do so at a high velocity to serve a great many customers, you're often going to see mishaps in data quality that seep into the process. Like with any endeavor in life, when you work too quickly, you often end up making mistakes.

Companies often have long, convoluted processes in place for gathering data, storing it in the right places and accessing it later when necessary. Frequently, these processes are inefficient, siloed among many departments or simply fraught with human error.

According to TechRadar, there are numerous potential ways that companies can get themselves into trouble with inaccurate data. Jeffery Brown, product manager at data specialist Infogix, struggled even to list them all.

"The causes for data errors within an enterprise are truly endless," Brown told the news source.

That doesn't mean they're imperceptible. In reality, every company is capable of investigating its troubles with data and finding out exactly where the process derails. Here are a few potential problem points:

Bad data from the source
A great deal of the data companies collect comes directly from primary sources - i.e., they get people's names and addresses when consumers are willing to enter that information in a form. Sometimes there are mistakes in those forms, and often, those errors are difficult to find and correct.

Sloppy manual data
There's also the trouble that arises when data is entered manually by a company's employees, who of course are human and always might make mistakes. Imagine, for example, that hundreds of potential customers filled out forms at a trade show using pen and paper, and they all have to be transcribed later. This process offers numerous opportunities for human error, so be wary.

Poor aggregation by multiple systems
Another issue is that of overcoming the silo effect - say a company has one database in its marketing department, another from the sales team and a third from customer service. What happens when you try to merge them all, but they're all in different formats and might contain duplicates? That's when things might get ugly.

Incorrect processing or cleansing
Companies engage in data-driven processes that might go wrong all the time. Consider a company's efforts to move data from silo to silo, to process it into a different format or to clean it by hand, using judgment calls. Every time businesses get "hands on" with their data, they risk tarnishing its quality. Caution is therefore vital.

Copyright ©, 2014-2017. All rights reserved.

125 Summer St Ste 1910, Boston MA 02110-1615, US