Retailers have been hit hard since the fourth industrial revolution began blurring the lines between the physical and digital worlds. They have been forced to keep up and evolve at every turn or perish. Those that survived the so-called “Amazon Effect” were thrust into the global Covid-19 pandemic that shut down physical stores almost overnight.
The key to survival for these retailers is data, yet many retailers are still flying blind when it comes to maintaining and managing their inventory. According to Shopify, out-of-stock items, or stockouts, cost retailers over $1 trillion every year. But the impacts of those stockouts are accretive. Not only is there a lost opportunity for revenue and a high price to rush replacements, but customers often get frustrated and go elsewhere, which hurts brand reputation and loyalty. Overstocking is just as bad because it can turn into dead stock – the Achilles’ heel of any retailer.
When it comes to inventory management, retailers that can survive whatever is thrown at them next are those that empower district or store managers to make SKU-level decisions. The ability to make data-driven decisions that consider customer preferences along with geographic and demographic data is a critical differentiator in the industry. Those who don’t harness all the available data will be left behind; those who do will dominate the market.
What retailers are doing now
No two locations of any restaurant, grocer, pharmacy or apparel store sell exactly the same products. Yet, company administrators make inventory decisions based on research reports commissioned when that physical location was originally built. There may be yearly lookbacks to make a few adjustments based on total sales for different SKUs, but put simply, store managers are often left to stock whatever they receive. It doesn’t matter that one location never sells Pantene Pro V shampoo or always runs out of Dove body wash; corporate sends the same items every week.
Stockouts due to surge purchasing can have devastating effects on inventory in one region, but not another. A more efficient management system can accommodate for these demands.
Even when store managers have an avenue to make inventory adjustments, it often involves a lengthy, manual process to make their way through the email chains of approval. When a music festival is happening near a guitar store location in a month, that store manager should be able to increase inventory for guitar picks with an automated approval process. Better yet, a system can predict the exact effect of that music festival would automatically make the inventory adjustment.
What will it take to succeed
Out-of-the-box solutions promise solutions, but often at the cost of additional siloed applications for a business tech infrastructure.
A quick Google search returns a seemingly endless number of programs and platforms promising to help retailers optimize inventory and improve the customer experience. And because both are necessary to strengthen competitive advantage, retailers buy them. Often, though, adding yet another platform to a retailer’s repertoire simply adds another disparate dataset that is incompatible with most of the company’s other systems.
Rule No. 1 in business is you’ve got to know your numbers. But companies could be doing so much more to squeeze more value from their numbers — without buying another new platform and going through a time-intensive and costly implementation process. Companies simply need a modern data infrastructure that connects the disparate systems they’ve already invested in and pipes that data into machine learning models, artificial intelligence and RPA efforts, creating an intelligent inventory management system that ultimately increases sales and customer satisfaction. Setting up the proper infrastructure eliminates the need to sift through spreadsheets to track inventory, which often leads to errors. And it eliminates the dependence on researchers that take weeks, if not months, to plan the right inventory and layout for a store.
In one recent engagement, Blueprint used machine learning to take manual inventory counts out of store managers’ hands. Blueprint took a slice of historical inventory and sales data from a quarter of the company’s thousands of stores, incorporated data on outside factors and created models to analyze trends. Within six weeks, the algorithm outperformed the manual inventory count and produced accurate sales forecasts. That algorithm’s accuracy will continue to improve as it learns and is refined by the data scientists behind it and as more external data is added. Read how that engagement started here.
To begin making SKU-level inventory decisions on a store-by-store basis, companies must incorporate as much external data as possible with their existing universe of data. To help with this, Blueprint partners with Precisely, opening the door to enriched geospatial, socio-demographic and traffic pattern datasets. As a company grows and even more data becomes available, a modern cloud-based infrastructure can scale along with it, allowing for the flexibility to constantly consume new data as it becomes available without the heavy lifting that comes with more traditional methods of data ingestion.
Let’s have a conversation about how your company can make smarter, faster and more proactive inventory decisions that will increase sales and customer loyalty.
