From soup to nuts, seasonal profile adjustment is an important – but often overlooked – part of inventory forecasting. Are your organization’s forecasting models keeping up?
For inventory managers, predicting sales for products with seasonal, slow or otherwise intermittent demand is notoriously challenging. In a recent survey of wholesale suppliers, nearly half (45%) called seasonal profile management a pain point for their businesses. From freezer case to OTC, seasonal items make up a significant portion of SKUs, which means accurate forecasting is critical. Yet many retailers are falling short, relying on outdated methods that don’t take seasonal dips and spikes into account.
As customer demand becomes increasingly complex in today’s flexible fulfillment environment, forward-thinking organizations are turning to new tools to help solve these challenges. By embracing a data-driven approach to inventory management, retailers can predict what they once considered unpredictable and manage their entire portfolio more effectively.
Outdated Tools, Unreliable Results
While new supply chain tools can help dramatically improve forecasting, many retailers haven’t yet moved beyond manual forecasting methods – or even old-fashioned gut instinct. According to the same survey noted above, half of suppliers haven’t implemented machine learning in their forecasting yet. By depending on outdated methods, retailers face challenges like:
- Forecasts that don’t account for seasonal adjustment appropriately
- Making seasonal adjustments that don’t improve forecast accuracy
- Not being able to calculate whether a seasonal adjustment will improve accuracy
For grocers, inaccurately forecasting seasonal items means getting stuck with excess holiday baking supplies in January or Easter hams in May. Stockpiling is a common tactic for organizations, the survey found, with more than 60% of wholesalers holding more than one month of inventory. Those high inventory levels don’t translate into higher service levels, however, with more than a quarter of organizations missing at least 4% of sales in 2019. Rather than ramping up inventory across the board, accurate forecasting requires optimizing levels by product – a practice that’s even more important for seasonal or other unpredictable items.
Going Beyond Guesswork With Machine Learning
To manage their inventory effectively, retailers first need to marry the optimal forecasting and replenishment strategy with each SKU, which requires a more advanced forecasting approach. Leveraging machine learning techniques can help retailers identify seasonal items more easily, generate more accurate forecasts and gain an edge on competitors still struggling with elementary demand models. To master seasonal profile management, here are four key areas to keep in mind.
- Determining which items to adjust. Do you know every SKU that should have a seasonal profile? The right demand classification techniques can help you understand how sales are likely to fluctuate across a wide variety of demand behaviors. Using probabilistic forecasting and advanced analytics, you can create an accurate forecasting model for each item – even the most challenging ones.
- Finding the right model. Many businesses rely on a classic exponential smoothing model to predict demand, which smooths sales activity throughout the year into a holistic forecast. But what happens if an item doesn’t have any sales for six months out of the year? Rather than smoothing periods of no sales into the forecast, which doesn’t produce the most accurate results, use advanced forecasting techniques for SKUs with intermittent demand. For example, aggregating sales to a higher level, such as category or location, can help provide a more precise forecast.
- Measuring accuracy. How confident are you that your seasonal forecasts are actually correct? Machine learning-enabled tools can help businesses remove the guesswork from seasonal adjustment. Using pattern-matching algorithms, these tools help ensure that seasonal adjustments will actually improve accuracy before making the change.
- Automating forecasts. The average retailer has thousands of SKUs, which makes it all but impossible to manually create forecasts for each seasonal product. Beyond improving accuracy, inventory forecasting tools can deliver big efficiency improvements by producing product-level forecasts, with no need for user intervention.
As technology continues to evolve, emerging techniques like unsupervised machine learning can help businesses improve accuracy further. Using near human-level intelligence, these tools can identify trends like seasonal pattern clustering, and adjust inventory levels accordingly.
For retailers just beginning to explore advanced forecasting tools, a supply chain partner can help organizations identify and implement the tools that offer the greatest value. By taking a data-driven approach to seasonal profile management, businesses can right-size inventory to meet dynamic, complex demand patterns all year long.