Following on from her first guest post for WhichPLM, here Javeria Gauhar Khan of Data Ladder shares some tips for preventing bad data from crippling your business.
Customer data to retail businesses is like fuel to a vehicle. Hyper-personalization (the use of data to create super-personalized experiences, like using cookies to deliver special promotions) requests the need for data – but not just any data. It needs high-quality, structured data. Unfortunately for most retail businesses, high-quality data is a distant dream. Over 50% of retail companies we’ve worked with complain about the impact of bad data on their business that includes:
- Poor operational efficiency caused by poor to non-existent data management
- Reduced employee productivity caused by bad data (employees having to spend half their day in fixing data errors such as spelling mistakes, incomplete information etc.)
- Ineffective marketing campaigns and missed opportunities
- Reduced customer satisfaction
Bad data is crippling retail businesses.
If you’re still wondering how and what can be done about it, read along.
How is bad data impacting your business?
Data is the backbone of any retail business. From marketing to advertising, customer service to customer retention, almost every department relies on the credibility of this data. Corrupt data, therefore, affects every downstream application and process, resulting in chaos and flawed decision-making.
One retail business shared with us a chilling example of how bad data has resulted in losing significant opportunities for growth.
The business was still using a legacy system to hold increasingly complex customer information. The system was storing contact information, purchase information, credit card information and much more, but it had no mechanism in place to ensure the quality of this information. Hundreds of thousands of fields had missing address data, fake credit card numbers, duplicated customer information and – more importantly – this data was streaming in from multiple sources. The system was connected to a CRM, which was then connected to the billing service and each department had its data entry operator. Sometimes, it was the sales manager adding more information into the CRM, sometimes it was the IT manager fixing data in the ERP, sometimes it was the customer service rep jotting down information. There was no sync between departments. The legacy system itself was so troublesome, business users had to keep filing tickets to IT for any data extraction activity.
At year-end, the company was unable to generate accurate reports. The duplicate customer information, the high marketing costs with poor ROI, the incessant customer complaints, the high unsubscribe rate made the company question its processes.
Sadly, this isn’t the story of just one company. Poor data costs businesses $3 trillion annually. Businesses are pouring millions into data acquisition, but only a handful of them are recognizing the problem of bad data and doing something about it.
How do data conscious companies handle poor data?
We’ve worked with our fair share of super-focused, data-conscious companies and the commonality we found in these businesses was the prioritizing of data quality as a mandatory part of organizational processes. These companies employed several methods to handle poor data:
- Preventing Bad Data at Data Collection: Companies that follow the 1-10-100 business rule understand that it’s easier and less costly to invest in preventive measures than in reactive measures. For instance, it costs $1 to create a web form that has strict data entry controls in place, preventing the occurrence of duplicates. If quality is not prioritized during data collection, the business will have to spend $10 in validating information, as reps spend their time in hunting for information. Business decisions are halted until the data is sorted. Lastly, bad data will cause failed migration and transformation plans. A $100 loss all because of a list of duplicates!Preventing bad data is the right way to do business. Sure, you can’t always guarantee 100% perfection when it comes to data. Somehow, duplicates will happen, human errors will be made, but adopting preventive measures makes it manageable rather than overwhelming.
- Prioritizing Data Quality: Smart businesses focus on their data quality first then think of investing millions in new systems or technologies. They know that their AI and ML-based initiatives will fail if they don’t possess accurate data. For instance, a bank developing an AI model to help vet candidates for a mortgage will need accurate customer lists to be able to perform comparisons and matches. This list will need to be free of duplicates, contain the correct information, follow a standardized format, and be error-free to be useful.Companies that invest in optimizing their quality have seen a 2X increase in ROI. Those that used data quality management tools have seen a significant impact on their productivity and operational efficiency rates.
- Leverage Smart Solutions: Data-conscious companies know that data quality is a consistent, regular effort that needs to be automated. Instead of spending millions in hiring expensive teams to develop in-house solutions that fail, these companies focus on using automated data quality management solutions that take the burden off them.Best-in-line solutions don’t just clean and prepare data, they also allow for data matching and data enrichment capabilities. This enables the organization to have complete control over their data quality, without having to spend millions on resources, talent, or expensive solutions.
- Empowering Business Owners to be Data Owners: Business users are the true owners of customer data, so it makes sense to empower them to handle data quality. Most companies leave data quality problems to IT, forcing business users to depend on IT users. Because this dependency results in unnecessary delays, business users often end up taking matters into their own hands, fixing data directly without consulting IT. Any issues that arise after become the cause of conflict.Companies invested in data know that customer data is vital to business users, therefore, they make IT and business departments share responsibilities. Business users are trained to manage the quality of their data by working in line with IT managers to ensure the accuracy, completeness, timeliness, and uniqueness of data (benchmarks defining the quality of data).
- Emphasizing on Data Quality Training: Workforces need to be prepared to handle data. The involvement should not just be limited to executive leadership as we saw with one company that improved its data quality significantly by implementing a series of training within the organization. From the intern to the marketing manager, everyone was trained on the multiple facets of data quality. Companies that want their employees to be more data-centric, must have programs in place to enable a data-driven culture. Merely bringing on new technologies or systems will not serve the purpose when the workforce is still unprepared to handle challenges.
So bad data is ruining your business, but it can be helped…
Retail businesses in particular are the most affected when it comes to data quality challenges, but, given the right combination of tools, technology, and people, this can be prevented. As in our experience with different retail businesses, those that tackled data quality challenges head-on were far better prepared to bring on big data, AI, and ML-based technologies.
So, ask yourself this: is your retail business ready to focus on data quality?