Demand Forecasting for the Modern Supply Chain
Demand forecasting refers to the process of planning and predicting goods and materials demand to help businesses stay as profitable as possible. Without strong demand forecasting, companies risk carrying wasteful and costly surplus – or losing opportunities because they have failed to anticipate customer needs, preferences, and purchasing intent.
Demand forecasting professionals have specialized skills and experience. When those skills are augmented with modern supply chain technologies and predictive analytics, supply chains can become more competitive and streamlined than ever.
Why is demand forecasting important for modern supply chains?
In the wake of the pandemic, companies are in an exceptionally fast-moving business climate. Customer behaviors and expectations are evolving quickly and as more and more businesses adopt optimized supply chain practices and cloud-connected business networks, competition is getting fierce. Demand forecasting is important to the supply chain because it helps to inform core operational processes such as demand-driven material resource planning (DDMRP), inbound logistics, manufacturing, financial planning, and risk assessment.
How does demand forecasting work?
At its best, demand forecasting combines both qualitative and quantitative forecasting, both of which rely upon the ability to gather insights from different data sources along the supply chain. Qualitative data can be curated from external sources such as news reports, cultural and social media trends, and competitor and market research. Internally-sourced data – such as customer feedback and preferences – also contributes greatly to an accurate forecasting picture.
Quantitative data is typically mostly internal and can be gathered from sales numbers, peak shopping periods, and Web and search analytics. Modern technologies employ advanced analytics, powerful databases, and use artificial intelligence (AI) and machine learning to analyze and process deep and complex data sets. When modern technology is applied to qualitative and quantitative forecasting and predictive analytics, supply chain managers can provide ever-increasing levels of accuracy and resilience.
Demand forecasting methods
Depending upon the industry, the customer base, and the volatility of the product, demand planning professionals use the following forecasting methods:
- Demand forecasting – macro-level: Macro-level demand forecasting looks at general economic conditions, external forces, and other broad influences that may disrupt or affect the business. These factors help inform businesses of regional and global risks or opportunities, and keep them be aware of general cultural and market shifts.
- Demand forecasting – micro-level: Demand forecasting at the micro level can be specific to a particular product, region, or customer segment. Micro-level forecasting is especially attuned to one-off or unexpected market shifts that might lead to a sudden spike or plunge in demand. For example, if experts are predicting a heat wave in New York and your company makes portable air conditioners, it may be worth the calculated risk of preemptively bumping up your inventory buffers in that area.
- Demand forecasting – short-term: Short-term demand forecasting can be at the micro or macro level. It is usually done for a period of fewer than 12 months to inform day-to-day operations. For example, it may involve consulting with the company’s sales and marketing teams to see if they’re planning any promotional or sales events that might cause a demand spike.
- Demand forecasting – long-term: Long-term demand forecasting can also be micro or macro, but typically looks ahead longer than one year. This helps businesses make better-informed decisions about things like expansion, enterprise investments, acquisitions, or new partnerships. When businesses give themselves a year or more to analyze and test markets, they can get a more robust picture of what kind of demand trends they can expect when they set up shop or launch products in new countries or regions.
Factors influencing demand planning and forecasting
Silos are the enemy of accurate demand planning and forecasting. To be at its most accurate and efficient, supply chain planning requires very different areas of the business to be connected in real time and to be continually contributing data and insights. When armed with as much data as possible, demand forecasters are better equipped to grapple with these factors:
Seasonality and inventory forecasting
Products like sunscreen or Christmas trees have a very obvious seasonal ramp-up. But seasonality can also apply to anything that causes customers’ behavior to change during the year. This could include unexpected weather events or even something like the pandemic, which caused people to stay home and be indoors more than they normally would during the summer months.
Competition as it relates to demand forecasting
In the 2020s, businesses are operating in a competitive and complex market. Customer expectations are changing fast and include demands for shorter product lifecycles, faster delivery, and more personalized services. With its spike in online shopping, the pandemic saw a drop in customer brand loyalty, which has also contributed to greater competitive forces.
Types of goods and demand estimates
Demand forecasting can vary wildly from product to product, even within the same product category. For example, demand for black t-shirts may change and suddenly start outstripping demand for white t-shirts. The trick is not to spot that it changed, but to spot why it changed. Lifetime customer value, average order value, and product purchase combinations also vary greatly and sometimes change suddenly.
With demand forecasting tools, you can better understand and predict these trends and their causes. This helps businesses learn how to customize, promote, or bundle items to drive more recurring revenue and to better see how one SKU affects or drives demand for another.
Traditionally, many businesses have managed with only a few regional warehouses and distribution centers serving wide geographical areas. However, largely due to the Amazon Effect, customers now expect same- or next-day deliveries. This means that businesses have had to put fulfillment centers all over the country to achieve the proximity necessary for these new demands. Furthermore, this is no longer exclusively a B2C challenge. Increasingly, B2B businesses are also feeling the pinch of delivery-speed pressures.
This phenomenon has caused enormous upheaval in traditional demand forecasting processes. Where once supply chain planners had only to worry about inventory levels at a few locations, they now must establish accurate buffers and stock levels at sometimes hundreds of small distribution centers. And obviously, this leads to increased risk and potential loss. It also means that demand-planning professionals are more reliant than ever on cloud-connected supply chain solutions to deliver the intel and informed real-time data to help them be super accurate with their now smaller and more widely dispersed inventories.
Three steps to get started with demand forecasting
Here are three simple steps to help you establish good supply chain planning strategies and demand forecasting best practices:
- Let demand forecasting be what it is.
Demand forecasting is an important backbone in the supply chain planning process and underpins a lot of other processes. It can therefore be tempting for businesses to let demand forecasting become a catch-all practice that is bent and wedged in to support various other supply chain planning functions.
When used properly, demand forecasting has clear purpose: it predicts what, how much, and when customers will purchase. Other supply chain functions – like S&OP, inventory optimization, and response and supply planning – deliver complementary capabilities within an integrated business planning system. If these tools are used for the specific functions they’ve been designed for, demand forecasting tools can get on with what they do best.
- Demand forecasting software loves data, data, and more data.
When supply chain technologies – particularly those dealing with demand and inventory forecasting – are powered with AI and machine learning, they get better, more accurate, and more insightful the more data you feed them. Don’t only rely on backward looking data like past sales or past product performance. Look to additional sources like news, politics, social trends, and customer insights.
Today, data doesn’t have to be linear and simple to be analyzed effectively. Modern data management tools can curate and process large and complex data sets. And AI and machine learning bring speed and intelligence that not only allows for advanced and predictive analytics, but also learns from experience and cumulative data input.
- Budget and plan accordingly to optimize demand forecasting.
Supply chain planning requires a realistic and strategic approach to be at its best. Legacy practices and workflows are hard to adjust, and people tend to resist change. But in the end, improved demand forecasting and supply chain planning can increase profitability and reduce risk and loss while providing your supply chain team members with a more streamlined and efficient working experience. By earmarking budgets and team resources early on, businesses can help support better buy-in and a smoother rollout of their supply chain optimization plans.
Get more competitive with predictive analytics and demand forecasting
Every step you take towards the digital transformation of your supply chain gets you that much closer to the visibility and efficiency you require in today’s competitive business climate. Work with supply chain managers and team leaders across your business to start breaking down silos and learning where the biggest risks may be hiding – as well as the greatest opportunities for long- and short-term wins. Then speak to your software vendor to learn more about integrating supply chain planning solutions into your operations.