Keeping Up With Customer Demand Through Data-Driven Forecasting

Fine-tuning inventory levels to meet changing customer preferences is a constant battle for wholesalers, and today’s fast-moving omnichannel environment is making that calculus even more complex. As they develop their purchasing plans, businesses are facing a bevy of uncertainties:

• E-commerce now accounts for 10 percent of total retail sales, requiring organizations across nearly every industry to manage both online and brick-and-mortar demand.
• Shoppers return 30 percent of online purchases, three times the number of in-person returns, which complicates inventory management further.
• Suppliers are servicing an increasingly complex web of online and offline channels, with 90 percent relying on field sales, 54 percent on their own websites, and 28 percent on third-party marketplaces beyond Amazon.com.
• External factors such as continued uncertainty over U.S.-China tariffs are adding to the pressure and forcing businesses to re-examine their inventory strategies.

It’s no surprise, then, that wholesalers cite increasingly volatile demand as their top challenge. In a recent survey, 75 percent say that developments like new competitors, new customers and the rise of e-commerce make inventory management a struggle.

As e-commerce continues to grow and customers have more ways than ever to shop, a solid strategy for predicting demand patterns and related inventory needs is a must. Yet many wholesalers have been slow to embrace digital transformation, relying on outdated tools that can’t stand up to fluid demand patterns. For businesses that want to win the omnichannel race, the time to shift to a data-driven strategy is now.

No Safety in Stockpiling

Today, the idea of inventory optimization for many businesses is “stock as many as we can get our hands on.” Stockpiling is a popular response to guard against sudden changes in customer demand, but more isn’t always more. Inventory increases create additional holding costs that erode profitability, make it difficult to reconcile plans or understand inventory changes, and often have little impact on customer service levels.

In a recent survey, 63 percent of wholesalers report having more than one month of inventory on hand. Those figures are even higher among those who cite volatile demand as a challenge, with seven in 10 reporting at least a month of inventory to spare. At the same time, however, 27 percent of wholesalers missed sales of more than 4 percent, an 8 percent increase from 2018. Inventory turns and profit are top metrics for these businesses, yet many are falling short — and inaccurate forecasting is largely to blame.

Many wholesalers still use elementary modeling methods or even gut instinct to predict and plan their inventory needs, a recipe for failure as customer expectations continue to rise. The wholesale industry has lagged behind other sectors when it comes to adopting advanced forecasting tools, with half of businesses reporting they haven’t used machine learning in their inventory management yet. By leveraging advanced systems and forecasting models, wholesalers can gain the insight they need to make smarter purchasing decisions and keep up with customer demand.

Better Systems, Better Strategies

Demand planning technology enables businesses to perform advanced modeling techniques — e.g., economically optimized replenishment cycles, cost of service analysis and safety stock cycle — that give them a clear view into the stocking needs for every SKU. Armed with the right tools, businesses can improve forecasting accuracy by doing the following:

• Refining their service-level strategies: Ramping up inventory levels across the board is a surefire way to end up with surplus. That’s especially true for slow- or low-moving items, which 66 percent of businesses say are challenging to manage. As they juggle a growing number of channels, using data insights to assign product levels precisely can help businesses improve inventory turns and customer service.
• Accounting for external factors: Variables like natural disasters, increasing competition or the latest consumer trend can wreck a purchasing plan fast. For example, 45 percent of wholesalers say predicting seasonal needs is an ongoing challenge. Data-driven forecasting can help businesses see further down the line, enabling them to make adjustments based on changing customer preferences and the evolving macro environment.
• Streamlining execution: Tightening transportation capacity and rising costs are putting the squeeze on many wholesaler budgets, prompting two out of three to adjust their ordering frequency. However, while ordering less frequently can help businesses trim transportation costs, it can also have unintended negative consequences on inventory levels. Building these costs into forecasting models can help businesses optimize order frequency based on transit costs as well as service-line needs, resulting in higher profits.

The right tools combine historical sales data with information like customer demographics, competitor analyses, upcoming promotions and even weather, producing more accurate forecasts that enable wholesalers to right-size their inventory. While customer behavior is bound to be unpredictable, a data-driven approach can help wholesalers put predictability back into inventory management.

Rod Daugherty is the vice president of product strategy for Blue Ridge, a supply chain solution.