Remember that we’ve had data lakes explained as a place to store data from different sources, in whatever format it comes, whether structured or unstructured. We’re exploring beyond the data lake definition, looking at practical uses of data lake analytics. So far, we’ve covered data lake analytics in politics and in supply chain management.
Now we’re honing in on data lake analytics in e-retail.
The Many Facets of a Customer
There is a lot to know about a retail customer. Are they shoppers who are just waiting to become customers of our businesses. How about the frequent customers? What about the one-time customer who we want to win back? Some customers are connected to us because of quality. Or others come back for the customer service.
How do we build the most targeted interactions for each customer? We need to look at a lot of different data sources.
Consider the case of online retailer Ru La La’s data lake experience. The company realized that its clickstream data, all of the information about what a customer did on the company’s website or with marketing emails, had been kept in one place. In a different data set, the company had been tracking sales and order history. By combining the two types of data into one place, a data lake, the company has seen more targeted and personalized marketing.
That isn’t the only analysis that come from the new combo data set. Ru La La is also using the information to track product performance. This means that planning and development are getting a boost as well.
The marketing benefits of seeing a customer from all sides are widespread. As a result, promotions, like customer loyalty programs, can become more tailored, more useful. Or the retailer can target the right products through the user experience.
Pierre Harand of the 55 Data Company describes an online fashion retailer’s data lake switch that accomplished just that. The retailer was able to use its detailed customer data to decide the order in which products show up on their website. That process wasn’t limited to customer data either. Instead, they could look to their inventory and price margins alongside customer preferences. The end results? More products added to customers’ carts and better product turnover.
The Many Facets of a Company
Even beyond a customer’s experience and history is the history of the company’s website, product listings, update schedules. This kind of data from the history of the company can paint another picture. It offers a 360 view of the company’s decisions that align with changing sales.
James Dixon talks about the potential for creating a sort of company time machine by using a data lake. With a data lake, a company can make a remote control with rewind and freeze frame options. Then, in ecommerce, he notes that a company can go back and see everything that was happening at a particular moment. The company can see “how many shopping carts where open, what was in them, which transactions were pending, which items were being boxed” and more.
This kind of freeze frame would let us deep dive into the processes that work. Now, it’s a way to finally see demand, supply, and operations all at once.
Intriguing isn’t it? Ready to think about a data lake for your marketing and operations? We’re here to help.