Retail customer attrition is at an all-time high and so are customer experience expectations. Today’s customers want to be valued as individuals which many retailers are not in a position to do. In fact, 68% of retail customers leave because they believe the retailer is indifferent to them and 66% leave due to poor customer service. The bottom line is that today’s customers are less loyal to brands and more loyal to customer experiences.
Both online and offline customer experiences are directly impacted by data quality. High data quality enables a single, accurate view of the customer. Low data quality introduces inconsistent variables such as different names or addresses for a single customer. While there are many good reasons why retailers don’t yet have a single view of their customers, when poor data quality becomes obvious to customers, the result can be very personal. In this worst-case scenario, customers flee to other retailers who make accurate customer information a priority.
Retailers that recognize the competitive importance of data quality capture the highest quality data upfront. This eBook explains the challenges retailers face and how they can ensure high quality data quickly, accurately, consistently and cost-effectively.
Customers don’t think of brands in terms of touchpoints, they think in terms of one, continuous experience. Retailers’ various touchpoints may include a mix of:
Multiple touchpoints are often involved in a single purchase. Failing to provide a consistent brand experience across channels and failing to recognize customers as individuals throughout their journeys causes churn.
Good data quality is the foundation upon which great customer experiences are built.
Retailers endeavor to provide ever more personalized shopping experiences as they gather additional types of data from mobile apps, loyalty programs, and IoT devices. For competitive reasons, retailers want to take maximum advantage of that data using advanced data analytics and, most recently, machine learning.
While most businesses have made the connection between data quality and analytics, they have not necessarily made the same connection between data quality and machine learning. How data quality impacts analytics and machine learning is addressed later in this eBook. For now, it’s important to recognize that bad data can negatively impact decision making based on analytics or machine learning.
Meanwhile, despite retailers’ dedication to emerging technologies and IT modernization efforts, many are still unable to access all of the data they need. Siloed data may be trapped in:
Each data source provides a view of the customer which may be incomplete. Between different systems a single customer’s data may be inconsistent because different applications use different fields or for different field structure s: (name versus first name and last name) for data collection. Further, the data may be stored in different formats (J. Smith versus John Smith). When the data from disparate sources are combined, retailers often discover they have redundant or inconsistent data.
Despite the lingering data quality issues, retailers continue to add more data sources from which to gain customer insights. For example, some ar e combining transaction histories an d loyalty card data with external demographic, geolocation and market data to achieve a “360-degree view” of individual customers. Yet, 81% of marketers say they lack a single customer view.
In most cases, “360 degrees” is merely aspirational since data elements are often missing or of poor quality. The result is a partial or erroneous customer view that can lead to false inferences. False inferences undermine efforts to build deeper relationships with customers.
Retailers don’t strive for bad quality data, it happens by default in the following ways:
The retailers that have overcome these obstacles have achieved a single customer view that brings together high quality master data.
Establishing a single customer view is important. However, it’s even more critical to ensure the quality of the data first so the customer views are accurate. For example, type-ahead addressing minimizes data entry errors made by customers making online purchases, CSRs in the call center, and sales associates at the POS. Type-ahead addressing can also immediately tag geolocation data in or attach geolocation data to customer information.
Accurate geolocation data is important because it helps ensure the timely delivery of purchases. Without it, retailers deliver some orders to the wrong address or the delivery otherwise fails. Most retailers absorb the cost of delivery failures so if a customer is required to pay for shipping, they only pay for shipping once. If the delivery is free or if the retailer wants to waive the delivery fee for the inconvenience of a failed delivery, that retailer absorbs both the cost of delivery and the cost of redelivery. Some retailers may offer additional incentives to preserve a positive customer relationship. Regardless, retailers pay for mistakes caused by data quality errors that could have been avoided in the first place.
Ensuring high data quality helps minimize failed deliveries and their associated costs. However, achieving high data quality can be very time-consuming and therefore expensive because the data must be cleansed and validated. In addition, duplicate records must be removed.
Third-party solutions automate some data cleansing and validation tasks, which saves time and money. In addition, data enrichment via third-party customer verification data and geolocation data helps ensure high quality master data.
Online retailer and financial services provider SHOP DIRECT Retail Ltd wanted to enable a single view of its customers which are located in the United Kingdom and Ireland. However, over time, the company’s database infrastructure and tools became too outdated to support modern marketing requirements.
To achieve a single customer view, SHOP DIRECT constructed an integrated prospect, inquiry, and customer account data warehouse that linked to the development and implementation of a customer relationship management (CRM) solution. The centralized customer data warehouse supports a single view of any individual who is currently, previously, or potentially trading with SHOP DIRECT.
SHOP DIRECT anticipated processing 12.7 million prospects, 6.5 million inquiries and 1.8 million name and address changes in the first year of the new data warehouse’s operation. The company used Trillium DQ to incrementally process new and changed records from source systems for inclusion in the data warehouse. To accelerate run time and reduce processing loads on individual machines, Trillium DQ discovers, standardizes, and enhances customer information and then identifies and matches group and individual relationships within cleansed data.
Using Trillium DQ, SHOP DIRECT achieved complete, accurate, and relevant customer views across its home shopping business while reducing bad debt. It was also able to consolidate multiple customer accounts and streamline data quality processes.
While there is a general misconception that analytics software or machine learning algorithms using raw data alone yield meaningful business insights, the truth is the quality of information and the accuracy of downstream analytics depend on quality data.
Data quality is also critical for machine learning because the data is used as machine learning training data. According to 451 Research, 49% of retailers are now running a machine learning proof of concept (PoC) compared to just 40% across all other verticals. Customer engagement is among the current use cases for 45% of retailers.
Importantly, high quality data enables reliable analytics and machine learning insights. Poor quality data yields dubious results. Practically speaking, the level of data quality results in such differences as:
High quality data enables higher marketing ROI, more accurate decisions, and the best actions for customer retention. And, when retailers can apply the broadest range of match algorithms to all parts of data across cloud and distributed platforms, they can achieve the best single view of the customer.
Operational analytics enables retailers to run their businesses more efficiently and profitably. Using geolocation data, they can improve operations in the following measurable ways:
Having accurate data is critical. Having accurate, timely data is even more important. Trillium Global Locator is ideal for high-volume environments such as call centers, marketing registration pages and ecommerce sites because it reduces the time to complete transactions while ensuring accurate geolocation data.
More retailers have created loyalty programs to gain additional insight into customer behavior. Some loyalty programs are very profitable and successful, while others fall short of expectations. Their success and lack of success are the direct result of data quality.
For example, Iceland Frozen Foods, one of the UK’s top 10 grocery chains increased store performance, checkout interaction and customer experience ratings through its loyalty card program. It used that information to generate discount coupons and personalize direct marketing offers, but the sheer transaction volume was a problem. Iceland needed to capture customer data from hundreds of thousands of card swipes per week across the store network, cleanse it, and match it with each cardholder’s records in the customer data warehouse. It then needed to update those records to create a current and accurate single customer view quickly and on an ongoing basis to avoid checkout delays, home delivery delays and customer frustration.
Rather than trying to optimize the existing loyalty card, Iceland decided to introduce a new loyalty card based on accurate views of over 3 million customers. Using Trillium DQ, it has improved the accuracy of home delivery and targeted marketing efforts. Using the data visualization tools, IT and the business can now collectively review and improve data quality performance, processes, and rules.
Good quality data helps retailers improve customer loyalty, customer engagement, share of wallet and customer lifetime value.
According to Marketing Land research, 66% of retailers say the accuracy of address details is critical to their business, but 80% say that customers often don’t realize that failed deliveries occur because the customer mistyped their own address. Type-ahead addressing and geolocation data together ensure fewer failed and late deliveries regardless of who made the data entry error.
In today’s hyper-competitive global business environment, merchandise delivery speed is critical to competitiveness. That’s why Amazon.com introduced one-day delivery for Prime Members. While Amazon isn’t always able to meet the one-day delivery target, the company realizes that delivery speed can make the difference between a sale and an abandoned cart as well as satisfied customers versus unsatisfied customers.
While most other retailers are not in a position to match Amazon’s one-day free delivery move, except Walmart for the time being, they’re accelerating deliveries in other ways such as allowing customers to order online and pick up in store, order online and have the order shipped from the store(s) nearest to the customer’s location, and order from an in-store kiosk for in-store pickup or home delivery. Having accurate customer and geolocation data helps ensure the fastest possible delivery times and also helps enable more seamless and frictionless multichannel and omnichannel experiences.
Conversely, failed and late deliveries reduce sales margins and negatively impact customer loyalty. They can also result in brand reputation issues when customers report them via social media channels. Savvy retailers are avoiding these results by prioritizing data quality and the timely availability of it.
It’s important to be able to assess, improve, and monitor the quality of your data to ensure the accuracy and completeness of it. Your company’s margins quite literally depend on it.
Accurate data facilitates positive customer interactions, accurate data analytics and machine learning outcomes, and timely merchandise deliveries. From a competitive standpoint, data quality is a differentiator. However, from the customer’s point of view, good data quality is table stakes.
With Trillium DQ and Trillium Global Locator you can: