Many companies found themselves in a predicament when they filled their storage facilities with excessive products, only to face financial trouble as inflation and economic downturns occurred. The importance of monitoring stock levels and cash flow cannot be underestimated. Without a comprehensive system that takes into account historical data, customer demand, and industry trends, businesses risk making costly mistakes in overstocking and storage expenses. This is where SKU-level demand forecasting comes into play, offering valuable insights and benefiting businesses greatly. By utilizing a combination of forecasting methods, data analysis, and key metrics, SKU-level demand forecasting empowers organizations to optimize their supply chain and make informed, data-driven decisions regarding pricing and inventory management.

Importance of SKU-Level Forecasting

Understanding the significance of SKU-level forecasting can revolutionize the operations of companies. Here are some useful tips on how to effectively utilize this forecasting method to drive success. SKU forecasting is a method that predicts the demand for individual products within a company's inventory. This process involves analysing data, including past sales and consumer trends, to help businesses anticipate future product demand and maintain the right levels of stock without overspending on storage or tying up cash in excess inventory. Ultimately, SKU forecasting allows companies to meet consumer demand without experiencing stockouts while also optimizing storage costs. Two primary methods of SKU forecasting are qualitative and quantitative.

Qualitative Forecasting

According to expert analysis and opinions, qualitative forecasting is conducted by individuals who rely on their judgment to make predictions in instances where data is limited. To enhance the accuracy of their forecasts, experts leverage market research, customer feedback, BETA launches, and product surveys.

Quantitative Forecasting

This approach is more straightforward. Quantitative forecasting relies on statistical analysis and data modelling, resulting in more data-driven results with less potential for human bias or errors. Techniques such as machine learning, artificial intelligence (AI), and time series analysis fall under this type of SKU forecasting. It is often advantageous to combine qualitative and quantitative methods when approaching SKU forecasting. While machines may outperform humans in accuracy, they may overlook certain personal aspects of forecasting such as new product introductions or unique events.

Benefits of SKU Forecasting

When it comes to businesses, especially those with online operations or extensive stock in warehouses, SKU forecasting plays a crucial role in managing inventory levels efficiently and reducing expenses. Below are outlined four compelling reasons for incorporating SKU forecasting into your operations.

Optimized Inventory

Managing inventory and optimizing stock levels continues to be a challenge for businesses that sell physical products. The rising costs of warehouse space, driven by shortages in skilled labour and increased prices of building materials, have added to the complexity. Utilizing SKU forecasting can assist companies in effectively managing inventory levels, ensuring that they maintain the appropriate amount of stock to meet customer demands without carrying excess inventory that incurs storage costs.

Improved Supply Chain

Forecasting SKU provides immediate inventory data that can be shared with suppliers to maintain adequate stock levels. Notifying suppliers promptly when products are low allows for production planning in advance, decreasing the risk of stockouts and ensuring customer satisfaction.

Market Trends Insights

Predicting SKUs can provide valuable insights into customer behaviour and market trends, enabling your business to make informed decisions regarding pricing and product demand. Think of it as a crystal ball that gives you a glimpse into what your customers’ desire, based on sales data, industry trends, and competitor offerings, allowing your business to align its offerings with consumer demand.

Better Cash Flow

SKU forecasting is essential for cost-saving measures through warehouse space optimization and preventing excess inventory of unnecessary products. The saved funds can then be reinvested in your business for strategies such as marketing campaigns or product development to drive growth.

SKU-level analysis involves examining individual products or SKUs in your inventory to assess their performance and potential demand. This analysis considers various factors such as stock levels, seasonality, lead times, and other variables that can influence product sales. By conducting SKU-level analysis, businesses can pinpoint which products are successful and which ones may need improvement. This type of analysis provides valuable data that can inform decisions on product discontinuation and the development of new offerings based on customer behaviour and market trends. In many cases, a combination of qualitative and quantitative approaches is the most effective method for SKU forecasting.

Forecast Inventory Levels with SKUs

In order to accurately forecast SKU demand, it is important to consider three main factors: your data, your method, and your metrics. Let's delve deeper into each of these components to ensure a precise forecast.

When developing a SKU forecasting strategy, it is essential to take into account various factors such as sales data, seasonality, and lead times. It's important to recognize that each industry has its own unique characteristics. For example, clothing retailers and manufacturers cannot approach SKU forecasting in the same manner. The initial key step in creating an effective SKU forecasting plan is to gather and assess data from your own sales to identify any trends or patterns.

After acquiring data, various forecasting methods can be employed to predict SKU demand effectively. Each method offers unique advantages; therefore, it is crucial to select a method that aligns with your industry, inventory objectives, and technological infrastructure.

Time Series Forecasting

Time series forecasting is a straightforward method that utilizes historical sales data to project future sales. This method assumes that previous sales patterns and levels will persist in the future, while also considering seasonal peaks and troughs. It is particularly beneficial for businesses that are new to SKU forecasting, as it can effectively analyze basic patterns such as seasonality and holiday demand. By utilizing time series forecasting, companies can anticipate product demand and optimize their inventory levels to meet customer needs. For instance, a retailer specializing in winter apparel can use this method to forecast demand for winter jackets during different seasons, allowing them to strategically manage their stock levels.

This approach is most reliable with a substantial volume of past sales data. However, its straightforward nature may limit its capability to conduct advanced analyses such as competitor or consumer behaviour studies.

Nirmalya Business Intelligence platform compiles historical sales data for individual product SKUs and generates forecasts. It employs algorithms such as the Autoregressive Integrated Moving Average (ARIMA) model to analyse data and predict future demand.

Machine Learning Forecasting

Machine learning forecasting is a more advanced method of SKU-level demand forecasting that uses algorithms to analyse data and spot patterns humans might miss. This method can be more accurate than time series forecasting because it can identify complex factors, like consumer trends, that impact product demand.

As machine learning forecasts analyse linear and non-linear relationships, it can handle more complex data than just consumer buying trends and seasonality. For example, a company that sells Detergents & Toiletries might find that demand for a particular item is influenced not just by the time of year, but also by the release date. Machine learning can use this information to create a more accurate forecast about product and consumer demand for the company.

Nirmalya Business Intelligence platform gathers extensive data on individual SKUs, encompassing sales figures, demographic information, market trends, and consumer preferences. This data is then inputted into a machine learning algorithm, typically specialized in SKU-level demand forecasting, to analyse the information and produce forecasts.

Regression Analysis

Regression analysis compares dependent variables, such as sales and inventory, with other fluctuating factors like price, consumer spending, and supplier costs in order to enhance the accuracy of predictions. This type of analysis is considered one of the most comprehensive methods for SKU forecasting as it can accommodate numerous independent variables. For businesses that deal with multiple variables such as various suppliers, seasonal variations, or fluctuations in the supply chain, utilizing regression analysis may be a strategic decision.

Nirmalya Business Intelligence platform gathers data on sales, manufacturing costs, and marketing costs for every SKU in your inventory. This data can then be used in conjunction with a platform such as Nirmalya BI to conduct regression analysis and produce forecasts based on the relationships between these variables.

Measure SKU Level Forecasts

For optimal utilization of your SKU data, it is essential to monitor key metrics and establish KPIs to assess the accuracy of forecasts. Below are three vital metrics to evaluate in SKU-level demand forecasting:

Stock Turnover Ratio

The stock turnover ratio assesses the speed at which products are sold and replenished in a company's inventory. It is calculated by dividing the cost of goods sold (COGS) by the average inventory held during a specific period, indicating how efficiently inventory is being managed. For instance, consider a FMCG retailer with an initial inventory value of 200,000 rupees for Surf Excel products and an average inventory of 10,000 rupees. The stock turnover ratio can be computed as shown below:

The stock turnover ratio is calculated as the cost of goods sold divided by the average inventory. In this case, the ratio is determined as follows: COGS ÷ average inventory = 200,000 ÷ 10,000 = 20.

The FMCG seller has a stock turnover ratio of 20, which means the average level of inventory is sold through about 20 times a year, or nearly once a fortnight. Having a high stock turnover typically indicates efficient inventory management and lower storage costs. Conversely, a low stock turnover may suggest inefficient use of funds for storage or difficulty selling overpriced items.

Gross Margin ROI

The Gross Margin Return on Investment (GMROI) quantifies the income a company generates from its inventory. It is determined by dividing the gross profit by the average inventory cost and is valuable for identifying SKUs with favourable profit margins. For instance, if a Retailer sells a Horlicks product earning 50,000 rupees in gross profit with an average inventory cost of 10,000 rupees, the GMROI for this SKU can be easily computed.

The Gross Margin Return on Investment (GMROI) formula calculates the relationship between gross profit and average inventory cost. In this case, the GMROI is calculated as 50,000 ÷ 10,000, resulting in a value of 5 Rupees.

The Gross Margin Return on Investment (GMROI) of five indicates that the CPG manufacturer earns 5 rupees for every rupee invested in Horlicks. Monitoring the GMROI for each stock-keeping unit (SKU) enables the business to pinpoint top-performing products, analyse demand fluctuations, and make appropriate production adjustments.

Weeks of Supply

It is advisable to monitor the inventory levels of your company to ensure optimal stock management. Having excessive stock on hand may indicate a wasteful use of resources, while low stock levels can lead to potential stockouts and customer dissatisfaction. For example, if a retailer has 145 units of a particular SKU in inventory and sells an average of 10 units per week, the following formula can be used to calculate the weeks of inventory supply.

The calculation for weeks of supply is determined by dividing the current inventory by the average weekly sales. In this case, the calculation is as follows: 145 ÷ 10 = 14.5 weeks.

The retailer currently has 14 weeks of supply remaining for this SKU in its inventory. The reorder level for this item will determine when to place another order with the supplier. It is important for the retailer to consider if maintaining a 14-week stock of a single product is necessary.

Best Practices

Keeping a close eye on key metrics like these can offer valuable insights into the precision of SKU predictions and help identify the optimal strategy for your particular industry and needs. Nirmalya BI assists organizations in smart stock management, continuous monitoring for adjustments as necessary, and expediting the decision-making process.

Effectively Manage Inventory

Avoid depleting your shelves completely, while also preventing them from overflowing. In addition to implementing a reliable SKU forecasting strategy, it is important to maintain safety stock and establish reorder points for each SKU to ensure optimal product levels. This entails:

• Safety stock, also known as buffer stock, refers to additional inventory maintained to safeguard against unexpected spikes in demand or disruptions in the supply chain. The optimal levels of safety stock for each SKU are determined by various factors, including stock turnover rates, supplier lead times, past customer demand, and storage expenses.
• Determine the reorder point for every SKU in your inventory, considering the cost of each SKU, safety stock levels, seasonality, supplier lead times, and GMROI. The goal is to strategically calculate the reorder point to efficiently manage inventory levels, meet sudden increases in demand, and reduce storage costs. This analysis will help optimize your inventory management process.
• The Economic Order Quantity (EOQ) for every SKU you carry is the most efficient order size that reduces the overall cost of inventory. This figure is determined by analysing expenses such as ordering, holding, and storing inventory. By computing the EOQ for each SKU, you can procure the optimal amount of inventory at the appropriate time based on projected demand and minimize the amount of excess stock being held. It is essential to constantly monitor and adjust your system, even if it is already in place, especially as your business expands or customer preferences evolve. This leads us to the next recommended practice.

Monitor and Adjust Accordingly

Regularly monitor sales and inventory levels for each SKU and compare them to the reports generated by your selected SKU-level demand forecasts. It is crucial to adjust your forecasts and reorder points as needed based on any unforeseen changes in demand or inventory levels to maintain accuracy. To achieve this, analyse actual sales, spikes in demand, and inventory levels to evaluate the accuracy of the forecasts and determine if any adjustments to inputs or outcomes are necessary. Ensure thorough monitoring of every SKU in your product line to consistently meet customer demand without negatively impacting your profitability.

Nirmalya Business Intelligence

Successful SKU forecasting relies on having the right platform in place. Monitoring stock levels, predicting changes in demand, and recommending optimal reorder quantities and stock numbers is a complex task that is beyond human capability. However, with the assistance of forecasting software, all of this can be achieved in real-time. Platforms like Nirmalya BI can optimize cash flow by reducing excess stock and improving availability to meet consumer demands. AI-driven solutions, such as our SKU-level demand forecasting app, Reorder, utilize various data sources like transactions, product information, pricing, and warehouse data to provide businesses with advanced AI models and accurate demand forecasts.

Inflation is a primary concern for business owners currently, leading to increased costs in warehouse storage, a critical component of the supply chain. Enhance your organization's efficiency by keeping track of minimum order quantities, seasonality forecasts, and warehouse capacity in real-time as customers purchase products or SKUs run low. Instead of sifting through endless spread-sheets, your team can focus on what truly matters: growing your business. For more information on Nirmalya BI, reach out to us today.