The era of “Big Data” involves a seemingly endless buffet of both “good data” and “bad data” – just as food-preparation hygiene is a critical factor in your eating experiences, data hygiene and data scrubbing are instrumental to the success of your overall business intelligence strategy.
What Is Data Scrubbing?
Raw data is just like raw fruit, vegetables and meat in one inescapable way – it is not ready for “prime time” without suitable preparation and processing. The act of “data scrubbing” should be envisioned as the early steps required to repair or delete incorrect data before it is transferred to the next step such as data warehousing.
Data Scrubbing vs Data Cleaning – “Data Scrubbing” and “Data Cleaning” are often used interchangeably, but they are really not the same. Data cleansing is a simpler process of removing inconsistencies while data scrubbing involves several specialized processes such as decoding, merging, filtering and translating.
Data Quality and Business Process Optimization
Today’s “Big Data” environment by necessity involves a higher degree of automation, and “data mistakes” are an undesired side effect. How big is this potential problem? According to a study by Experian, 86 percent of companies recently reported that there might be “inaccuracies” in their data. The same research indicates that about 40 percent of businesses view incomplete information as their biggest data problem.
An effective data quality strategy is literally impossible for your company unless you aggressively attack “dirty data” before it causes even bigger problems. For example, the Experian study also reports that 40 percent of all business project failures involve poor quality data as the primary reason.
The problems due to dirty data are, of course, not totally new. The wise and practical computer axiom of GIGO – garbage in, garbage out – was postulated over 50 years ago due to the early recognition about the critical importance of data hygiene. About 99 percent of today’s businesses understand the importance of having clean and scrubbed data. While there is plenty of genuine awareness about the problem, the bigger challenge is actually “fixing” the problem of “bad data.”
Cost Implications of Bad Data
If you could identify a business intelligence strategy that would improve your bottom line by 10 percent or more, what would you do? According to the Data Warehousing Institute, dirty data costs U.S. businesses over $500 billion annually. Experian estimates that a typical company loses about 12 percent of net income from the bottom line due to inadequate data hygiene.
Sources and Causes of Poor Data Quality
In the search for where you should start looking for dirty data (database records that contain errors), here are the five primary sources and causes to investigate:
- Older systems with obsolete data.
- Inaccurate data entry (good, old-fashioned human error).
- Too many databases and channels to maintain properly.
- Lack of coding standards.
- Simple errors like missing data in database fields.
According to the Experian survey, human error is the leading cause of dirty data – and 50 percent of respondents mentioned call centers as their biggest source of problem data. Another 37 percent mentioned data from mail orders and retail physical branches. A major source of dirty data in the healthcare industry is inadequate coding standards.
Common Approaches to Improve Data Quality
Only 30 percent of organizations have a data quality strategy. The other 70 percent deal with data quality problems only as they appear or pursue simple cures like manual checking on Excel spreadsheets. Regardless of whether your company falls in the 30 percent or 70 percent categories, you should search for cost-effective ways to improve data hygiene.
5 Simple Data Scrubbing Steps for Your Team
If there was a simple way to eliminate dirty data in your company, would you use it? Here is a practical suggestion with five simple steps to consider:
- Determine at least one area in your business with a data quality problem.
- Enlist the participation of top management by demonstrating business benefits due to data quality improvements.
- Review current data quality in the “target” area.
- Use data scrubbing software and other data scrubbing tools to produce a sample report about the current data quality in the target area.
- Within your various teams, find “Data Quality Champions” to lead the way by spreading awareness and communicating both big and small achievements in improving data quality.
Remember – you are not alone in this effort. As noted above, well over 80 percent of companies have data problems lurking somewhere within the organization.
But this is a specialized business problem that will benefit from all of the expert help you can muster – do not hesitate to talk to data management experts like DataEntryOutsourced.
Please share your experiences with data scrubbing and dirty data by using the social media buttons and leaving a comment below.
– DataEntryOutsourced
Prev
