CSA Exclusive: Oracle Retail says data science is key to planning

Retailers seeking to manage uncertainty in merchandise planning and forecasting should apply advanced data analytics.

Jeff Warren, VP of strategy and solution management, Oracle Retail, recently spoke with Chain Store Age about the continuing challenges retailers face in properly stocking individual stores. Warren discussed how sophisticated analytical solutions make it possible for retailers to effectively sort through large volumes of data and adapt to disruptions such as COVID-19-driven demand surges and port bottlenecks.

How can retailers meet customer demand for "mothballed" products that have been pulled from shelves during COVID-19?
Over the past year, a season’s worth of merchandise moved from sitting in dark stores to waiting in warehouses for spring to come again. As these styles once again come into season, retailers need to create new location planning processes to ensure ‘mothballed’ inventory is in the right stores to recoup maximum sales and margin.

Retailers have always aimed to create store-specific assortments to best align inventory and buying decisions with customer demand, and they should take the same approach when allocating stock that was pulled back at the start of the pandemic. Understanding each store’s unique selling patterns is a critical step to meeting that unique customer demand, but can be overwhelming for the buyer and planner due to the massive amounts of data and labor-intensive tasks required. Data science tools can identify location-level selling patterns so retailers can systematically distribute holdover merchandise.

Retailers should also adjust assortment strategies with proactive in-season item management (based on the customer, trends, and item attributes) to achieve maximum margins. Seasonal goods that have a limited demand window should take precedence over goods that might have less seasonality or competition in terms of promotion and mark down budgets.

What strategies and solutions can retailers use to effectively react to unexpected demand surges and shifts?
In the past, retailers have reviewed inbound receipts and recent sales history to build inventory flows and sales plans. However, the past year is an anomaly, and retailers lack insight from recent history to guide them. As a result, the only way to make an informed decision is through the location plan, using a combination of data science and analytics, based upon an optimal and adjusted set of reference data.

With optimized history, a modern retail planning solution can analyze and correct data for the past year’s anomalies resulting from store and warehouse shutdowns, ensuring that the location plans’ foundation is accurate and trustworthy. This advanced use of retail science gives retailers the confidence to proceed with the plan.

How can technology help retailers deal with port bottlenecks, such as the Suez Canal backup?
Like any hold-up in the supply chain, the concern with port bottlenecks is a potential disruption to the flow of inventory into stores and warehouses. Whatever the cause, retailers should look to the inventory they do have and determine how best to reallocate and redistribute it across their operational footprints. Enabling a 360-degree view of available inventory, agnostic of location, empowers retailers to locate the right inventory, regardless of where it physically sits. 

If overall stock is running short, merchants must determine if it’s more advantageous to shift stock between stores or channels but potentially price it differently. Retailers should consider hypothetical ‘what-if’ situations and challenges to help determine the best strategy and price point for each item. Every channel and location poses its own unique potential challenges and rewards, so it is essential for brands to contemplate different scenarios to understand how to profitably place inventory.

As well, a modern order management system can provide an essential safety net to retailers in light of uncertainty. A sophisticated management system will leverage order brokering options, such as weighted assignments and probability rules, to fulfill orders based on multiple criteria, such as sale velocity, margin and proximity.

What changes in consumer behavior can retailers expect in the post-virus "new normal"?Our research indicates consumers are ready to make up for lost time and missed celebrations. As the economy reopens following the pandemic, 21% of consumers said they’d buy ‘casual/party attire’ first. Retailers should bear this in mind as they plan assortments for the summer and beyond, knowing shoppers are looking to gather and celebrate once it’s safe to do so.

We also know from our research that consumers have missed the tangible aspects of retail—such as trying things on (25%) and being able to touch and feel products (23%)—the most. Retailers should prepare for the operational impact of a return toward normal foot traffic as well as the reopening of fitting rooms, which may have been closed throughout the pandemic. They must prepare store associates for challenges like lines, crowds and discarded items in fitting rooms—all things staff hired in the past year may not have encountered.