Demystify The Origin Of Data: Part 2 Common Misconceptions About Master Data Collection

Share this:

Today, 1 in 3 business leaders say they DO NOT trust the data they use to make decisions (source: IBM). That’s a scary thought. There are systematic problems with data quality in the world of master data management. The path to improved data confidence starts when you ask your data vendor this simple question: “Where does your data come from?”

There are some common misunderstandings about data collection, that cause dirty data to recycle through the marketplace. We’d like to clear them up now…

MISCONCEPTION #1: Triangulation always leads to better data quality.

What Most Think

Most people believe that if data is triangulated for accuracy (cross-referenced with other neutral sources to determine what is most valid), then the information must be correct.


Real Talk

Currently, most data vendors are reselling information from a handful of the same original sources. Therefore, although triangulation occurs and information seems to be cross-checked for accuracy, in reality, each vendor is passing along the same source of data, causing dirty data to be recycled throughout the market masked as ‘verified’.


MISCONCEPTION #2: Manual verification leads to better data quality.

What Most Think

Many business leaders place a higher value on vendors that offer manual data verification (having a human directly verify the information as true).

Real Talk

The reality is that when data vendors use manual verification, they are spending costly resources in the wrong areas. Instead, machine learning and artificial intelligence (AI) can augment human intelligence, to scour publicly available information at scale, gleaning new insights and continuously building upon them.


MISCONCEPTION #3: More is better.

What Most Think

Many believe that the more data sources involved in the validation process, the better more reliable the answer will be.

Real Talk

Merging data together is a complex process. Each source of data has its own complexity and biases. The more data sources being merged, the more complicated the process becomes, as illustrated by the figure on the right. Joining together two, three or even five sources of data can be fairly straightforward. But imagine merging twenty or thirty sources together. Combining many different data sets together requires high-cost data intelligence systems to understand the nuances of each data source and attempt to unify them. You might merge things together that are inaccurate or incorrect.


3-Part Series: Demystify The Origin Of Business Data

Understand the truth behind the origin of your data and take control of your data strategy.

Download now
Share this: