A Retail Tale – When Online Met Offline
Most apparel retailers were quick to see the growing relevance of cross-channel transactions. Regulars at their high street stores, they noticed, first browsed through the online racks. When customers did shop online, they sometimes asked to return outfits at a nearby store. And retailers started to rejig their systems to support this cross-channel shopping. But the challenges, they discovered, ran deeper.
When customers, having shopped online, returned items of clothing at high street stores, demand planners often had no visibility into this event – especially not in real-time or even near-real-time.
So, their systems, taking into consideration only online purchases (not the returns in-store) forecasted an inaccurate high demand for the particular line of clothing. As a result, store managers often received a larger inventory than they really needed or were able to consume.
Sometimes, the situation played out in the reverse. Customers having reserved products online, showed up at high street stores to pick their shopping. But store managers, with little or no visibility into items that are reserved online, often lacked the inventory to deliver – especially for fast-moving popular lines of apparel.
Making matters worse, the distribution centers often maintained two separate inventories – one for online and the other for offline sales – without the flexibility to systemically interchange stocks between inventories, if needed. Obviously, revenues took a hit.
Online and offline promotions, and pricing – each had a life of its own. Customers once in-store, on discovering that the same product was priced lower online, requested that a price adjustment be made at the store.
Most store managers felt compelled to comply. They treated this as a return and new in-store sale. But, in reality, margins were being eroded with in-store inventory being sold at online prices.
Finding and Framing the Real Problem
When retailers treat their online and offline channels as separate entities, but bring these together superficially to quickly solve the problems posed by evolving customer demands, it – at best – offers brief symptomatic relief. The core of their supply chain is still ill-prepared to adapt to the new realities of omnichannel retailing, and leverage the opportunities it has to offer. Finding a way for retailers to have a holistic view of transactions across all channels is the real problem that must be solved.
Solving the Problem
Create a data lake that presents retailer with a single, unified 360 degree view of all channels in near-real-time. This can help retailers effectively address pain points around inventory planning, improve sales and demand forecasting, and sync online-in-store promotions. Here’s how:
- Anytime-anywhere visibility into ‘available-to-sell’ stock: Build a model that takes into account the entire inventory - both online and offline. With orders that have been placed, orders that are ready for shipment, cancelled orders, orders reserved at the store, and returns called out, the ‘available-to-sell’ inventory is published for every store manager’s perusal, and order planning. Order routing, thus, becomes more intelligent.
- Single source of insights for sales planning: Merge the online and in-store data sets to create one unified data lake that provides a rich source of insights for sales projection - without the additional effort of an operations team working to unify and prepare data for sales planning, week after week.
- Improvements to demand forecasting: Several customers buy products online, especially during promotions, but return some of it thereafter. Often, retailers rely on the number of products shipped to arrive at the demand metric. The proposed model helps retailers understand demand more holistically. It offers better visibility into orders placed, orders cancelled and orders shipped – thus leading to better demand forecasting.
The Outcomes
- Inventory optimization led to significant savings of millions of dollars for one retailer that deployed this solution.
- Time-to-insights for sales projection is slashed from 2-3 days to minutes; This, in the case of one retailer, was the equivalent of 600X latency reductions. The data lake also gives insights into the number of price adjustments made in-store.
- The data lake allows promotions across online and in-store entities to be executed in sync, leading to better margin management for retailers.
