Big Data in Retail
Retail big data offers some big payback potential. The McKinsey Global Institute estimates that big data can grow profits in the retail industry by a whopping 60%.
Big data in retail is all about mining the volume, variety and velocity of consumer data streams in order to generate insights which lead to quick recognition of shopping patterns and demand trends, more personalized marketing, more successful product launches, optimized assortment and merchandising, improved shopping experiences and better consumer relationships.
Or you can go further and leverage big data such as social media trending and customer sentiment to influence demand planning, determine price elasticity or recommend merchandising and inventory optimization by channel. The point here is that big data use cases are as varied as the consumer data itself.
Most retail executives recognize the transformational impact Big Data can bring to their businesses. But they sometimes struggle in identifying the use cases and supporting processes that can convert untapped data into improved decision making. Here are some examples that may stimulate that thinking.
Marketing Advancements: Big data can be leveraged to develop micro customer segmentation, geo or proximity based marketing, real-time relevant offers, high propensity cross-sell recommendations and sentiment analysis by store, product, geography and channel. For example, sentiment analysis can inform retailers how consumers perceive their actions, offers and products—extremely valuable information for improving sales and marketing performance.
By cross-analyzing store and online interactions and conversions, and further cross-referencing the results by consumer demographical and geographical data, retailers will discover with far greater accuracy how to pinpoint the ideal customers for select products, deliver messaging for improved engagement and create offers for improved conversions.
Merchandising Enhancements: Harnessing big data from social channels or in-store behaviors can improve product assortment, placement and pricing which results in smarter shopping experiences and positively influences purchase decisions. Uncovering patterns which may include pop culture events, online buzz or weather data can be shown to improve merchandise placement, bundling opportunities, promotions and pricing.
Going further, correlating buyer interactions across channels, such as in-store merchandising placement with e-commerce product categorization or placement can deliver far more empirical data to show how product placement, product bundles or product cross sell promotions are optimally positioned.
Supply Chain Optimization: Applying big data to demand planning can aid just-in-time inventory distribution and improved logistics to help get the right products to the right destinations at the right times – and reduce both overstocks and stockouts. For example, with improved market demand models which go beyond looking at seasonal fluctuations and historical purchase patterns and further consider fluid market conditions and real-time customer demand gathered from online and social channels, retailers can optimize product shipments of top-selling merchandise, reallocate inventory to locations incurring higher demand, know exactly when to mark up or mark down item prices (by channel and location) and get advanced notice of when product demand will recede.
Customer Experiences: The analysis of what is being said by consumers online can provide retailers with valuable insights to enhance customer service and customer experiences by store or across mobile and online channels. Using a retail CRM system with social listening tools and integration to consumer social profiles can combine the consumer's demographics, firmographics and purchase history with far more revealing personal information gathered from social media, mobile app utilization, online and offline browsing patterns, and loyalty program interactions. The result is a far better understanding of the consumer’s persona and preferences along with what it takes to satisfy and delight the consumer.
In a recent retail big data project that I'm pretty proud of, we developed a big data model to perform product pricing elasticity by customer sentiment, segment, region, timeframe and competitors.
The client's consumer, inventory, pricing and POS data was spread across multiple systems and different formats. We aggregated the data to the retail CRM system so that it could be appended with consumer social data, used with workflow rules and displayed in dashboards, reports and a data warehouse. The data was revealing on many fronts, however, the biggest payback came when applying the price elasticity model with various consumer dimensions.
In one scenario, the pricing data was combined with a digital marketing campaign segmented by combinations of consumer profiles. The campaign yielded a 4.5% increase in conversion rate, and more importantly a 7.6% sales uplift with a 13% rise in gross margin. We later used the data to improve our Next Best Offer (NBO) algorithm which has increased cross-sell conversions from high single digits to low double digits and continues to grow sales uplift.
For more big data examples that deliver big paybacks, refer to my prior post of 5 Retail Big Data Examples.
The Retail Big Data ROI
Collating unstructured online data from sources such as social media trending, web browsing patterns, online communities, niche forums and digital media along with more structured data from POS systems, the loyalty system and the CRM application identifies hidden patterns, enables more specific business decision making and facilitates predictive models which give retailers advance notice of product demand and optimal methods to satisfy that demand. Interrogating and subdividing the data by store, region, demographics, behaviors and customer segment brings more specificity to promotions, merchandising strategies, bundling opportunities, price optimization models, supply chain planning and the many factors needed to get the right products to the right places at the right time.
For big data design and deployment guidance, check out my prior post on big data.
Next - Retail Store Immersion >>