Bad data quality can have serious consequences

Organizations are collecting increasing stockpiles of data to achieve a wide range of goals that include highly personalized promotional ads on ecommerce sites, loans that can be extended with little risk and street lights that improve the flow of traffic. With the 2.5 exabytes of information that are being generated on a daily basis and flowing into users' platforms, it seems like a reasonable assumption that some of that information will be affected by poor data quality, such as inaccuracies or content that is slightly off the mark. 

In many circumstances, a certain margin of error is acceptable. The United States Food and Drug Administration (FDA) backs this idea with its Defect Levels Handbook, which lists the amount of acceptable contaminants in food products. According to the FDA, 100 grams of peanut butter (about six tablespoons) will average 30 or more insect fragments, one or more rodent hairs and 25 mg of grit. Some companies can also generate a suitable end-product from ingredients that contain a small number of errors. However, there are other circumstances in which data quality is of the utmost importance. 

When bad data quality kills
Fatality may be the worst possible outcome of poor data quality, and while it might seem histrionic to say that bad information can lead to death, that's exactly what can happen in the pharmaceutical industry if the right checks and balances are not implemented, according to InformationWeek. The FDA is responsible for reviewing proposed medication names to prevent any two prescriptions from having monikers that are too similar and could lead to life-threatening mix-ups. 

The organization uses data matching techniques to find instances in which the sounds and spellings are too alike, the source explains. However, there are still examples like Methadone and Metadate or Taxcol and Toxotere. These drugs have been administered incorrectly and killed patients. 

When bad data quality hurts
There are many situations that fall in between the two extremes, in which data quality may be crucial but isn't life-threatening. Financial institutions, for example, have turned to data to inform their lending decisions in an attempt to remove the human margin of error. A recent study by Fenergo and Knowledgement found that banks are having difficulty improving their customer experiences and complying with regulations because of their data quality. 

Siloed information and lack of transparency were two particularly troublesome areas, the study found. Fortunately, there are tools that can help companies make the most of the information they collect. Data quality platforms ensure data is complete and organized before analysis.