• by Todd Hinton , July 26, 2016

Not all customer data is created equal, and poor data quality is money down the drain. With poor data, you’re basically driving a business with terrible gas mileage.

The reasons behind poor data quality are primarily the “three Vs:” volume, variety and velocity. There are tremendous volumes of data, too many types of data, and the speed with which organizations must act on that data continues to increase. Data has a shelf life, just like motor oil. Never act on data that isn’t timely.

A study by 451 Research of 200 C-level and senior IT leaders found that only 40% were “very confident” of their organization’s data quality, and yet 94% recognized the impact it has on business incomes. So how can organizations achieve data precision?

Know where you’re going

First of all, organizations must determine the goals of their data. Without a clear picture of why you are collecting information, it’s impossible to enforce data governance policies or derive actionable insight.

Data governance establishes the rules by which a company uses its data. Its primary role is to ensure that data informs critical business functions.

Invest in a good vehicle

Second, organizations must invest in technology that automates data preparation to remove the human element. Business decisions increasingly must be made in real time, which is nearly impossible when people are doing most of the legwork.

For example, retailers compare store locations with addresses of customers to determine who has to drive more than 30 miles to visit a store. This helps them decide where to open the next location. But customer address data that’s outdated or incorrect, could have serious negative effects on the bottom line. The retailer must be able to identify things like customers who visit multiple stores, or those who live in the same household.

Most analysts today spend 80% of their time cleaning and prepping data, and 20% on actual analytics. This should be flipped. With automation, analysts can reduce the time spent manually entering and prepping data, and more time actually conducting analyses.

Software tools can flag issues, like missing last names or invalid phone numbers, and alert a human that they need to check certain columns or tables from a source. To be a data-driven organization, it’s essential to have these types of tools in place to be able to make real-time decisions.

When choosing data quality tools, ask four questions:

Is it future-proofed? Does it have the capability to quickly scale and handle new types and larger volumes of data?

How effective is its matching accuracy? The accuracy (and the scarcity of false positives) is imperative to data quality.

How powerful is the data standardization? It’s imperative to cleanse, standardize and normalize data prior to performing analysis or a data migration project.

Who are the tool’s competitors? See how your vendor stacks up.

Know the rules of the road

Finally, organizations (and members at all levels) must follow data governance policies and place a high priority on data quality. Analysts must voice their needs and present issues they’ve run into that could be corrected. And everyone from salespeople to customer service reps must be aware of the importance of accuracy.

Poor data quality is the biggest obstacle to actionable insights. You don’t just want to know that the same person who visited your store last week is the same customer that called customer service yesterday, and made a purchase on the Web site three months ago. You need to know, and customers expect you to. Without concrete goals, well-thought-out policies and the right technology tools in place, an organization is driving aimlessly.