Editor’s Note: We’re thrilled to welcome Shirley Zhao of Experian Data Quality, one of Mi9 Retail’s valued partners, to our articles page to share some expert insights about the importance of quality data for retailers.

Dirty data in retail is dangerous, and costly, business. When seventy-six percent of retailers believe their bottom line is affected by inaccurate and incomplete contact data, and when twelve percent of revenue is believed to be wasted, you want to make sure you’re getting as accurate, and as clean a dataset as possible to make those tight margins.

So what exactly is Dirty Data? Data is inevitably going to be “dirty,” mainly due to human error that results in obsolete, inaccurate or missing information. If data were defined normally as being “fit for purpose” for whatever business objective you have in mind, then dirty data is its complete opposite—that some aspect of the data prevents it from being used as was intended.

In the retail industry, data quality is at the heart of what both troubles and brings life to the business. 2.5 billion gigabytes of data were generated every day in 2012 and that volume is expected to double every two years until 2020. But what happens when the quality of that data is compromised due to human error, technological error or what have you?

These inaccuracies not only affect your bottom line, but also restrict your ability to properly use that data for initiatives like enhancing customer engagement and successful decision-making, as shown below by a 2014 survey conducted by Experian Data Quality.


When exceptional customer service is at the heart of everything you do, understanding that retail is detail and that poor detail leads to poorer customer service is the first step to tackling a data quality problem. Congratulations, and welcome to Data Quality Anonymous.

So, what are some strategies you can put in place to mitigate the impact of poor data?

Strategy #1: Validating Data At Its Source

The farther incorrect data travels down your data lifecycle, the more damage gets done. Without validating the accuracy and completeness of data at the source, changing your data management processes and systems would have little effect—kind of like trying to stick one Band-Aid on fifteen paper cuts. One step to correct this is to implement real-time APIs for data validation corrects data at its source, right at the point-of-sale.

Strategy #2: Prioritizing Your Top Data Quality Challenges

You can’t tackle every data quality problem everywhere. A staggering 84 percent of businesses recognize myriad data quality challenges, the biggest among them being deliverability, and enhancing insight and loyalty. Confronting all of these challenges at once can only complicate and dilute the potency of your data quality initiatives. Begin by defining the broadest, most noticeable problems and work your way towards a pinpointed focus. Channel Napoleon, and divide and conquer your data demons.

Strategy #3: Be Systemic and Holistic

Approach your data quality initiative not as a single project with new goals and definitions to be set every time, but rather a chance to analyze and restructure the way data flows within your organizational structure. Much of the problem lies in the fact that data is siloed across departments, where one department may define, categorize and communicate data in an entirely different way than another. Another problem that causes poor data quality is an overemphasis on technology rather than on the people and processes.

Oh no! What to do?

      1) Define and outline your processes. Make sure that there are clearly defined steps in place to organize and understand business information at all data touchpoints—that includes all departments and the people that handle the data as well as defining a common language that can be shared inter-departmentally.

2) Create a template for the future. A good, nailed-down process can make things flow smoother. If there are kinks in the process, even the act of documenting it down somewhere makes it easier to go back and improve. The initial effort taken to produce comprehensive template is worth the time saved on future projects.

3) Get A Chief Data Officer (CDO). These days, there are more and more distinctions between Chief Information Officers (CIO) and Chief Technology Officers (CTO), and the Chief Data Officer (CDO). Like some gargantuan marriage between the CIO and CTO, the CDO’s primary responsibilities lay solely within information compliance and information innovation, strategy and governance.

Realize that improving data quality is an ongoing battle, but being overwhelmed by the task at hand and neglecting improvement for other aspects of your business is a costly mistake. By recognizing the implications of poor data quality and beginning the conversation to better data collection methods and processes, you can prevent those mistakes from cutting into your company’s expenses.

Want to continue the conversation with a data quality expert? Check out Experian Data Quality!