Big data and better retailing

The history of retailing has been narrated through several relatively stable periods, but interspersed with inflection points or - rather - disruptive changes. At the heart of these changes has invariably been data:

Firstly, the emergence of EPOS in the 1970s and 1980s played a major role in the development of category management;

Secondly, in the ensuing 1990s retailers' loyalty or membership card programs created a marketing industry that made better decisions based entirely on customer insights, with Kroger in the US and Tesco in the UK Kroger in the U.S. and Tesco in the U.K. led the global retail industry in this area;

Third, and most recently, the e-commerce revolution has provided retailers with data and insights that were previously unavailable - on customer decisions. Through the use of clickstream data, the customer is recognizable most of the time, and then categories can understand that when I bought product C, I actually looked at product A and product B as well - a huge breakthrough for industries with slow sales cycles. Additionally, omni-channel retailing and social media have opened up a new era that allows customers to take possession of a wealth of information to compare products, services, and prices - even though they may end up purchasing in a brick-and-mortar store. Once again, a whole industry has emerged - through customer retargeting technology and recommendation engines - where e-commerce can make real-time business decisions.

In the first few years of the 21st century, the term big data was used to describe a whole set of new concepts, such as lots of records (long data), lots of dimensions (wide data), text or images (unstructured data), and real-time or quasi-real-time (near real-time). The big data explosion triggered by the development of technology and social media has provided retailers and brands with more ways and means to stay highly connected with their customers and do more business.

At the heart of all retailing is the creation of a better value proposition for the customer: whether it's lower pricing that saves the customer money, more relevant product selection, better customer service, more effective promotions, or more efficient operations and distribution.

Because there are so many new technologies generating data, and so many new sources of data, retailers must have a framework for making sense of that data: the Shopping Trip Model. The Shopping Trip Model defines all of the customer touch points that generate data and the scenarios in which those touch points occur (e.g., search, visit/store, shelf pickup, payment, usage, etc.). In this way, any new data concept is understood as a function of some touch point, there are many touch points and each of them generates multiple data sets.

For example:

Shopping (online or offline) in the model - that is, the behavior of the customer that occurs during those moments when he/she is browsing on the website, in the app or in the physical store. Specifically, in the case of brick-and-mortar stores, emerging technologies allow retailers to understand how customers browse in the store and how they end up finding the items they're looking for. WIFI is widely available in airports, hotels and restaurants, and in China, it's starting to become more common in stores. This could change - and the ultimate winner could be some of these data collection technologies - perhaps beacon or other Bluetooth-based solutions. Regardless of who will dominate in terms of technology, the opportunity for data remains the same - as does the ability to understand how long customers stay and how they travel. The ability to link and analyze shopper insights through customer identifiable tokens (like an auto-login app) to behaviors such as pickup and payment could lead to real business transformation for retailers.

So how does this data actually end up helping retailers? In the past, retailers have typically tagged customers in a variety of ways (extrapolating or directly collecting data that describes an individual's wealth profile, family status, shopping behavior, etc.). However, this is only a glimpse of the pie. For data to be used in real time, enhancements are necessary in terms of time, location and goods or categories.

Customers may be price-sensitive about beans but willing to spend a lot of money on skincare. A customer in Shanghai may work a fast-paced job Monday through Friday but by the end of the week will take her family to a shopping center for a relaxing weekend and likes to shop from the comfort of her own home, so that customer should receive special treatment that day. Think about how you can tell store guides to use customer insights.

In the next five years, using multiple data sources to understand customers in multiple dimensions will be standard practice in retail. The companies that achieve a competitive advantage with big data will be those that can accurately characterize the customer in a variety of scenarios. Once the data is available, the possibilities for differentiation and personalization will be limited only by that company's imagination and ability to execute based on customer insights.

Thousands of stores or store differentiation (offering different promotions, personalized pricing, differentiated merchandise mixes or brand experiences, etc.) creates huge logistical and operational challenges. In the coming years, there will be retailers who have the data capabilities to personalize the shopping experience for their customers, but lack the operational capabilities to implement the changes suggested by data analytics. In fact, even today, many stores know that tweaking the evening's selections will result in better sales - the limitations of implementing merchandise adjustments lie not in the data insights that can guide the adjustments, but in the operational factors such as shelving, floor space, or inventory control. Just as important as investing in data is the need to smarten the store - giving the store real-time visibility into inventory status (perhaps via RFID), physical attributes (temperature, lighting, humidity, etc.), store associates (location, language, specialties, etc.), and, more importantly, the customers who visit the store (profiles, shopping preferences, etc.). By linking the "people (customers)" to the "place (store)" through data, we can provide customers with a personalized shopping experience that differentiates the store, both online and offline.

The above is what I share with you about big data and better retail, more information can be concerned about the Global Ivy to share more dry goods