Data wrangling, also known as data mungling, shapes data, found in different formats, restructuring it to provide business value. Companies implement a variety of data systems to guide business decisions and solve problems to optimize a situation. Over time business context evolves, with newer technologies, additional data sources, updated regulations, and different stakeholder requirements. This change results in larger, messier outputs of unstructured and structured data that do not meet information needs.

Data analysts, data scientists, and others who search for information in data systems need to adapt transform data to meet the current context and requirements. Often this ends up as a semi-automated approach to clean the data, taking time and money from other processes. Businesses want better Data Management, integration, and automation to reduce the need for data wrangling and formatting data.

Other Definitions Data Wrangling Include:

• “The process of transforming data from one shape into another to prepare it for analysis and deliver some unified results. It is also known as ‘data munging.’” (Pete Aven)
• “Copying data to different data silos that could be in the cloud or elsewhere.” (Jennifer Zaino)
• “Steps that convert data from its raw to tidy form.” (Harvard University)
• “Taking data in one format and put it in another format that you need.” (MIT)
• The process of bringing different data sets together like cowboys corralling cattle. (Forbes)

Data Wrangling Examples Include:

• Data scientists wrangle data to get clean data sets for analysis
• A customer visits a retailer and wants a report of his or her spending; however, the retailer’s purchase information, spread out in different systems, does not mesh. So it takes extra time for the client to get the information
• Collecting and cleaning data to train artificial intelligence

Businesses Data Wrangle to:

• Unify information from disparate sources
• Create a single view of a customer or product
• Get essential business insights manually and repetitively due to a lack of automation
• Answer research questions

Image used under license from Shutterstock.com