In today’s volatile and uncertain market, companies can no longer afford to operate without a structured, data-driven approach to forecasting demand. Demand planning is more than just predicting sales—it’s about building an integrated, agile business that can respond to customer needs while managing resources efficiently. Despite its importance, many organizations still rely on outdated tools like spreadsheets or allow bias and siloed decision-making to corrupt their forecast accuracy.

Employing predictive analytics and integrated demand planning can significantly streamline decision-making processes, create new insights, and save several business functions a lot of time and money.

This article explains why businesses need to leverage demand planning to improve their operations and explains the quantifiable value of doing it so that it can be sold within an organization.

Why Focus on Demand Planning?

Most companies that decide to invest in demand planning or improve their process are primarily driven by one or more of the following:

• Forecast accuracy challenges
• A highly variable process that needs improvement
• Need for a more efficient manufacturing or distribution system
• Downstream inventory problems driven by unseen variability
• Desire to improve cooperation between sales and operations

At its core, demand planning synchronizes operations. It allows marketing, sales, supply chain, finance, and production to operate from a common set of assumptions. Without an accurate demand plan, supply planning becomes reactive, finance struggles with forecasting revenue, and customer service deteriorates from stockouts or excess inventory.

Consumer behaviors have become increasingly unpredictable. Economic shifts, global disruptions, and rapid product cycles mean relying on historical sales alone is no longer sufficient. Demand planning introduces a proactive lens incorporating internal drivers (promotions, price changes) and external signals (market trends, customer insights) to create adaptive forecasts.

Inaccurate demand forecasts translate to costly outcomes: expedited shipping, excess working capital, lost sales, and markdowns. Improved demand planning helps reduce forecast error, allowing for better inventory placement, production planning, and supplier coordination. Even a five to ten percent improvement in forecast accuracy can have a significant bottom-line impact.

Potential Improvements Resulting From Demand Planning

Organizations that invest in improving demand planning benefit from:

• Reduced inventory costs through better alignment of supply and demand.
• Improved service levels by placing the right product in the right place at the right time.
• Higher forecast accuracy can lead to more reliable plans across finance and supply.
• Faster decision-making is enabled by real-time data and scenario analysis.
• Greater agility because of the ability to quickly adjust to shifts in demand or supply.

A mountain of research today shows that a mature demand planning process helps improve forecast accuracy and deliver a high return on investment (ROI). Improved forecast accuracy, when combined with software that translates the forecast into meaningful actions, will decrease inventory and operating costs, increase service and sales, improve cash flow and gross margin return on inventory investment (GMROI), and increase pre-tax profitability. The forecasting error, no matter how small it is, significantly affects the bottom line. In our experience, a 15 percent forecast accuracy improvement will deliver a 3 percent or higher pre-tax improvement.

In a previous IBF study of 15 U.S. companies, we found that even a one-percentage-point improvement in under-forecasting at a $1 billion company delivers a savings of as much as $1.52 million, and for the same amount of improvement in over-forecasting, $1.28 million.[i]

The reduction in downstream finished goods inventory resulting from a well-established process and forecast accuracy improvements provides a one-time saving, as well as recurring savings arising from reduced carrying costs. There are great benefits in a make-to-stock or distribution company, the downstream inventory reduction could range from 10 to 20 percent since forecasting inaccuracies typically drive around 75 percent of the required safety stock.

Building and Investing in Demand Planning

Here are some best practices when it comes to demand planning.

• Build an unbiased, unconstrained, consensus-based forecast. Organizations often confuse the demand plan with the sales target. Sales may overestimate to push for stretch goals, while operations may buffer to protect service. Demand planning needs to separate judgment from aspiration. Instituting a formal demand consensus process ensures all voices are heard, but forecasts remain grounded in data and evaluated against actual performance.
• Upgrade from static spreadsheets to dynamic models. Many companies still use Excel as their primary planning tool. While familiar, spreadsheets lack scalability, version control, and real-time integration. Upgrading to a dedicated demand planning system (or enhancing existing tools with forecasting models) introduces automation, improves collaboration, and enables real-time adjustments. It also supports more advanced techniques such as decomposition models or AI-based forecasts.
• Understand and match models to patterns. Not all items follow the same demand pattern. Some are seasonal, some have trends, and others are highly volatile. Applying a one-size-fits-all model can lead to overfitting or underperformance. Instead, classify SKUs by their demand characteristics and apply the appropriate model, whether that’s exponential smoothing, moving average, or more complex causal models.
• Focus on data quality and forecastability. Forecasting is only as good as the data behind it. Cleanse data for outliers, missing periods, and promotions. Measure forecastability using the Coefficient of Variation (CV) or Demand Intermittency. The demand planner becomes the integrator, ensuring inputs from various departments are translated into a structured forecast. Establish accountability through KPIs like bias, MAPE, and forecast value add (FVA).
• Invest in training and improving skills with IBF. Leverage IBF’s certifications, workshops, and learning resources to empower your teams with proven forecasting and planning knowledge, building internal capability that drives consistent, confident decision-making.

Taking steps to practice demand planning optimally will increase the bottom-line benefits you gain from it.

Bottom Line Benefits for Practicing Demand Planning

Many companies leave money on the table with lost sales or poor service levels. An integrated demand planning process can translate to increased revenue of 0.5 percent to 3 percent with improved inventory availability or demand shaping capabilities. Total annual direct material purchase, along with logistics-related expenses arising from demand variability and lost opportunities, can see direct improvements of 3 to 5 percent. We can also benefit from a 20 percent reduction in airfreight costs. The figure below shows the anticipated benefits from a 15 percent improvement in forecast accuracy. (These are averages and individual results for organizations. They are dependent on many other variables and can be higher or lower.)

This illustrates the possible benefits from a 15 percent improvement in forecast accuracy

It is essential to understand that these are average savings amounts. It is up to you to determine what savings you believe you can drive with a mature predictive analytics and demand planning process. Sometimes you need to know what finance and executive leadership anticipate in terms of benefits; you need to be on the same page in terms of expectations. It is here that the Institute of Business Forecasting Advisory Services can shed some light on what is realistic based on past implementations.

Demand planning is not just a supply chain function; it’s a strategic business process that empowers smarter, faster decisions. In an environment where disruption is the norm and expectations are high, companies that implement disciplined, data-driven demand planning will not only survive but also lead.

The Benefits of Demand Planning: The Final Word

The path forward is clear: Separate judgment from strategy, invest in tools and talent, and build a collaborative process that evolves with your business.

In a world of uncertainty, demand planning offers clarity. It’s not just about predicting the future, it’s about preparing for it. Companies that invest in robust, unbiased, and collaborative demand planning are the ones that outperform, outmaneuver, and outlast their competition.

But you don’t have to do it alone.

The Institute of Business Forecasting (IBF) has been the trusted authority in forecasting, demand planning, and S&OP for over four decades. Whether you’re just starting your planning journey or looking to refine and elevate your process, IBF offers the training, certification, tools, and global community to help you succeed.

Join IBF and take the next step:

• Get certified with globally recognized credentials
• Attend industry-leading conferences and events
• Access exclusive research, case studies, and best practices
• Learn from and connect with top planning professionals around the world.

[i] Chaman L. Jain (2018). The Impact of People and Processes on Forecast Error in S&OP. IBF research report #18. August 31, 2018

Do you have questions about demand planning? This guide explains everything you need to know about this complex topic in a simple and understandable way.

What is Demand Planning?

Demand planning is the process of using analytics, data, insights, and experience to make predictions and respond to various business needs. It leverages demand forecasts—not as an end in themselves—but as a tool to highlight opportunities and risks, establish business goals, and support proactive planning across functions.

While demand planning often reports into the supply chain, it is not solely a supply chain function. It is a cross-functional discipline that integrates insights from sales, marketing, finance, and operations to create a consensus plan—a unified view of what is most likely to happen in the market.

Demand planning combines historical sales analysis, market intelligence, consumer behavior trends, and business knowledge to guide actions across the organization. It enables companies to anticipate demand shifts, align resources accordingly, and avoid stockouts and excess inventory, especially in an environment where customer expectations and market conditions constantly evolve.

At its core, demand planning drives better business performance by ensuring that decisions are based on relevant, timely, and collaborative inputs, not guesswork or isolated projections.

Bottom line: An accurate demand forecast provides the information your operations and sales teams need to plan how much product to buy or manufacture to meet projected demand as efficiently as possible with limited waste.

What is the Difference between Demand Planning and Demand Forecasting?

Demand forecasting and demand planning are closely connected but serve different purposes:

• Demand forecasting is the analytical process of using data, statistical models, and judgment to predict future demand. It’s a probability-based estimate of what might happen and forms the foundation for planning decisions?
• Demand planning takes the forecast further by integrating it into the business strategy, aligning stakeholders around a shared set of expectations, and determining the actions needed to respond to that demand.

What Purpose Does Demand Planning Serve?

Demand planning aims to create a realistic and actionable view of future demand so the organization can align supply, resources, and investments accordingly. It addresses critical issues including:

• Strategic planning and assessing risk (long-term planning and S&OP/IBP)
• Finance and accounting (budgets and cost controls)
• Marketing (consumer behavior, life cycle management, pricing)
• Operations and supply chain (resource planning, production, logistics, inventory)

Why Is Demand Planning Important?

In a world of increasing uncertainty, demand planning helps companies stay ahead. Done well, it enables organizations to:

• Improve service levels and customer satisfaction
• Minimize inventory carrying costs and waste
• Respond quickly to supply chain disruptions or market shifts
• Enhance collaboration between departments
• Increase profitability and operational efficiency

Demand planning is not a one-time task but an ongoing, iterative process that requires the correct data, tools, and cross-functional collaboration. It must also be flexible enough to adapt to volatility, whether driven by global events, consumer trends, or economic shifts.

Poor demand planning leads to the outcomes businesses aim to avoid: lost sales, excess inventory, wasted capital, and disconnected teams working from different assumptions. A well-executed demand planning process, on the other hand, builds organizational alignment, reduces bias, and leads to better business outcomes.

Why is it Critical for Businesses to Practice Demand Planning?

Demand planning is not just a supply chain function; it’s a core business process that drives strategic alignment, financial performance, and customer satisfaction. Based on IBF research from 34 organizations across different industries, companies that invest in structured, data-driven demand planning realize tangible benefits across key areas of the business.

1. Improve service and protect revenue: A strong demand planning process helps businesses meet customer needs with greater reliability, ensuring on-time and in-full (OTIF) performance, even in the face of market volatility or promotional spikes. The result? Improved customer satisfaction, stronger brand loyalty, and higher top-line revenue.

Fact: A 15-point improvement in forecast accuracy has been shown to drive on to two percent in top-line sales growth and improve OTIF performance, meaning fewer stockouts and more happy customers.

Demand planning ensures your business doesn’t miss out on sales because of poor availability. It provides the early visibility needed to make smarter inventory and production decisions, keeping your shelves stocked and your customers returning.

2. Increase operational efficiency and reduce cost: Demand planning enables organizations to run more efficiently by minimizing waste, improving resource utilization, and allowing smarter scheduling of production, logistics, and labor. It transforms decision-making from reactive to proactive—letting teams plan ahead rather than scramble in response.

Fact: A 15-point improvement in forecast accuracy can deliver a 2.3 percent or more increase in pre-tax net profit, driven by better operational alignment and cost control.

By aligning cross-functional teams around a consensus forecast, organizations reduce duplication of effort, optimize capacity, and ensure the right resources are available at the right time—leading to smoother operations and lower costs.

3. Manage assets and free up cash: Effective demand planning significantly improves companies’ inventory and working capital management. With clearer insight into what’s actually needed—and when—businesses can reduce excess inventory, lower carrying costs, and avoid the pitfalls of overproduction or fire-sale markdowns.

Fact: For every 15-point improvement in forecast accuracy, companies can realize a 12 percent reduction in inventory, freeing up valuable cash and minimizing waste.

Demand planning ensures businesses are not over-invested in supply, storage, staffing, or space. It helps unlock capital tied up in inventory and directs it toward more strategic, value-added investments.

Demand planning is no longer optional. It’s a strategic necessity. Organizations that invest in robust demand planning processes not only gain greater visibility and control but also position themselves to thrive in a constantly evolving marketplace. With the proper training, structure, and leadership, demand planning becomes a competitive advantage that enables more resilient, data-driven organizations.

Where Does Demand Planning Fit Within an Organization?

Demand planning is a strategic, cross-functional process that touches nearly every part of the organization—from supply chain and operations to sales, marketing, and finance. While its reporting structure can vary, what matters most is how the function is structured, supported, and empowered, not simply where it reports.

Based on IBF research and industry benchmarks:

• 48 percent of demand planning functions report into supply chain or operations
• 23 percent report into the commercial side of the business, such as sales or marketing
• 8 percent report into finance
• 10 percent operate as an independent function or report directly to a business unit owner
• The remaining 11 percent follow other models depending on organizational design.

These variations reflect the flexibility of demand planning—it can reside within different departments, depending on the company’s structure, maturity, and strategic priorities.

But here’s the key: regardless of the reporting line, demand planning must operate as a cross-functional, collaborative, and unbiased process. Its success depends on its ability to engage multiple stakeholders, reconcile competing priorities, and drive consensus to produce a unified, realistic view of future demand.

The Complexities of Demand Planning

Finding and maintaining the perfect balance between sufficiency and surplus can prove especially tricky. It isn’t a once-and-done task. Economic conditions change, and competitive environments constantly evolve.

To address this, demand planning typically requires using demand forecasting to predict future demand trends. This has added benefits, most importantly, heightened company efficiency and increased customer satisfaction.

What are the Key Components of Demand Planning?

Here are the critical parts of demand planning:

Product portfolio management

Effective demand management requires a clear understanding of product lifecycles, from launch to phase-out. Product portfolio management supports this by tracking each product’s stage and showing how changes in demand can impact related items. It also plays a key role in planning new product introductions, helping teams anticipate demand, allocate resources, and support successful launches. With strong portfolio management, companies can better manage transitions, reduce risk, and respond more effectively to market changes.

Statistical forecasting

Statistical forecasting is based on the concept that past history best predicts future performance. It uses complex algorithms to analyze historical data to develop demand forecasts. This exacting process demands accurate data, including eliminating outliers, exclusions, and baseless or inaccurate assumptions.

Sales forecast and overrides

As a process champion, the demand planner plays a critical role in driving consistency, structure, and accountability across the forecasting process. One of the key responsibilities is managing sales inputs and overrides—ensuring that adjustments to the statistical forecast are based on valid insights rather than bias. This involves working closely with sales teams to understand market intelligence, promotions, and customer expectations while also challenging assumptions when needed. The goal is to balance collaboration with discipline, ensuring that overrides improve forecast accuracy and align with broader business objectives.

Trade promotion management

In today’s highly competitive environment, it can be challenging to spark the interest of prospective customers. That’s why sales and other promotions are becoming increasingly common. They often result in increased consumer demand. Trade promotion management helps ensure that these types of programs are properly executed, that there is adequate product supply, and that they deliver all expected benefits to a company.

Demand Planning Methods

Quantitative forecasting methods are the foundation of most forecasting processes, with approximately 74 percent of companies relying on historical data to project future demand. Standard demand forecasting methods are:

• Time series models, used by nearly half of organizations (48 percent), are the most common approach and focus on identifying patterns, trends, and seasonality in historical data.
• Cause-and-effect models, used by 17 percent of companies, link external or internal variables—like price changes or promotions—to shifts in demand behavior.
• Machine learning and AI are emerging tools in forecasting. Currently, about six percent of organizations use them, and as adoption grows, they offer the potential for more adaptive and automated insights.
• Judgmental forecasting, reported by 17 percent of companies, is a qualitative method incorporating expert knowledge, market intelligence, and human insight when data is limited or context is needed.

What is Required to Do Demand Planning Effectively?

Effective demand planning is more than just generating a forecast. It’s about creating a reliable, unbiased view of future demand that drives smarter decisions across the organization. Done right, it improves service levels, optimizes inventory, enhances collaboration, and ultimately boosts profitability. However, to achieve these outcomes, companies must establish the proper foundation. Here’s what’s truly required to do demand planning effectively:

• A clearly defined process: An effective demand planning process must be structured, repeatable, and aligned with business goals. It should define all stakeholders’ roles, responsibilities, timelines, and expectations. The process should incorporate steps for data collection, model development, consensus building, evaluation, and communication—ensuring that each cycle produces more accurate and actionable insights than the last.
• High-quality, clean data: The best forecasts are built on relevant, clean, and complete data. That includes historical sales, customer orders, promotional activity, and external factors like market trends and economic indicators. Without trustworthy inputs, even the most sophisticated models will produce unreliable outputs. Demand planners must work with IT and business teams to ensure data integrity, consistency, and standardization.
• Forecasting approach: An effective demand planning process requires selecting the right forecasting approach, whether it’s bottom-up (built from item-level inputs), top-down (driven by high-level business targets), or middle-out (a blend of both, used to reconcile plans across levels). Planners must also determine the appropriate level of aggregation, such as by item, customer, location, or time, based on how the forecast will be used and the level of noise in the data. The planning horizon must match the decision being supported—ranging from strategic (long-term capacity and investments) to tactical (monthly or quarterly planning) to operational (weekly or daily execution). Since no forecast is perfectly accurate, planners should establish acceptable and expected error thresholds and measure forecast performance to continuously improve.
• Cross-functional collaboration: Demand planning is inherently cross-functional, involving input from sales, marketing, supply chain, finance, and operations. To be effective, the process must include a consensus step, where teams align on a final, agreed-upon forecast. This collaboration minimizes bias, integrates commercial intelligence, and ensures the forecast reflects both statistical outputs and business realities.
• Skilled demand planners: The demand planner plays a critical role as a process champion and cross-functional influencer. Strong planners possess analytical capabilities, organizational awareness, communication skills, and the ability to challenge assumptions objectively. They must manage statistical models, evaluate overrides, monitor forecast accuracy, and facilitate dialogue between departments.
• Focus on continuous improvement: No forecast will be perfect, but the goal is to improve continuously. That means measuring forecast accuracy and bias, tracking value-added steps, and adjusting models and inputs over time. Each forecasting cycle should yield better insights and inform more intelligent decisions.
• Executive support and integration into business strategy: Demand planning must be embedded in the organization’s decision-making processes with strong executive support to ensure it has the visibility, tools, and authority to drive change. Gaining buy-in from key stakeholders is equally critical, as it builds alignment, promotes cross-functional engagement, and reinforces the value demand planning brings through improved customer service, operational efficiency, and business performance.

Demand Planning: Best Practices

Once the foundational elements are in place, adopting these best practices ensures demand planning becomes a value-driving process that adapts to change and supports better business outcomes:

• Understand the purpose of forecasting: Clearly define why you are forecasting—whether for financial alignment, production planning, or service optimization—to tailor the process accordingly. Anchor the planning process in key business questions and explicitly state the assumptions driving forecast changes and decision-making.
• Identify demand drivers: Analyze internal and external factors—such as seasonality, promotions, economic trends, and customer behavior—that influence demand patterns.
• Gather relevant inputs across functions: Incorporate insights from sales, marketing, finance, and operations to ensure the forecast reflects a broad and informed perspective. Cleansing and structuring data is essential to ensure accuracy and consistency, providing a reliable foundation for effective forecasting and informed decision-making.
• Track forecast performance regularly: Measure and report forecast accuracy and bias at appropriate levels of aggregation to continuously improve planning effectiveness. Forecast errors and metrics help us identify uncertainty and bias, allowing us to communicate them clearly, prioritize errors in high-value products and items, and improve forecast accuracy through better inputs and process refinement.
• Schedule timely and recurring meetings: Regular forecast review meetings enable collaboration, resolve conflicts, and build consensus around the final demand plan. Demand planning should act as a hub for cross-functional alignment, bringing together departments to drive consensus and accountability.
• Communicate and manage results: Share insights and results across the organization, highlighting successes, identifying gaps, and reinforcing the value of the demand planning process.

What Skills Do Demand Planners Need?

Effective demand planners must combine analytical expertise with business acumen to interpret data and translate it into actionable insights. They need strong communication and collaboration skills to work cross-functionally with sales, marketing, finance, and operations, facilitating alignment and consensus. A deep understanding of forecasting techniques—from time series models to causal methods and emerging AI tools—is essential to building and evaluating accurate forecasts.

Demand planners must also be adept at managing uncertainty and bias, using metrics to identify errors, and continuously improving forecast performance. Critical thinking and problem-solving abilities are key to challenging assumptions, evaluating overrides, and navigating business complexity. Equally important is the ability to act as a process champion, ensuring the planning cycle is structured, repeatable, and aligned with strategic goals. Ultimately, demand planners serve as integrators across the organization, requiring a balance of technical skills, strategic thinking, and emotional intelligence to influence without authority.

The Future of Demand Planning

The future of demand planning is rapidly evolving into a more strategic, technology-enabled, and integrated function that drives value across the entire enterprise. Fueled by advancements in artificial intelligence (AI), machine learning, and predictive analytics, demand planning is becoming more precise, automated, and responsive. These technologies allow organizations to analyze vast amounts of real-time data from sources like point-of-sale systems, distributors, and suppliers, enabling more accurate forecasts and timely decisions that reduce waste and improve customer service.

As forecasting tools become more sophisticated, the demand planner’s role will shift from generating numbers to generating insights, focusing on scenario planning, cross-functional collaboration, and business alignment. Demand planning will continue to integrate with S&OP and IBP processes, connecting operational planning to financial and strategic goals. With global supply chains becoming more complex and volatile, demand planners will be expected to manage greater uncertainty while maintaining agility and discipline.

However, as Eric Wilson of the Institute of Business Forecasting (IBF) cautions, the successful integration of advanced technologies requires more than just investment—it demands alignment with business strategy, proper implementation, and upskilling teams to interpret and act on AI-driven insights. Without these, organizations risk underutilizing powerful tools or making misaligned decisions. Done right, the future of demand planning is not just digitality becoming a central pillar of strategy and competitive advantage.

Demand Planning 101: The Final Word

The world of demand planning is rapidly evolving. However, the reality is that companies that don’t practice it must jump on board. If they don’t, they risk losing out to competitors who do. Demand planning will help you satisfy consumers, run your organization efficiently, and drive dollars to your bottom line.

Leverage the information in this guide—and the other resources available through IBF—to launch and optimize a demand planning practice at your company.

This article is taken from the book, Practical Guide to Sales & Operations Planning (S&OP/IBP). It’s currently available at a special introductory price. Get a copy here before the price increases.

As companies advance in their S&OP journey, the depth and breadth of planning expand, necessitating greater involvement from finance professionals. Finance expertise becomes indispensable as it adds nuanced insights beyond mere numerical analysis to the S&OP framework.

This integration fortifies businesses, enhancing their adaptability in today’s dynamic market landscape. Integrating finance into S&OP transforms the process from an ancillary function to a pivotal component of organizational strategy. Managers gain visibility into real costs, steering decisions towards tangible outcomes rather than theoretical conjectures. From procurement to production scheduling and marketing strategies, finance-informed decisions align with overarching business objectives and financial plans.

“Finance transforms S&OP from an ancillary function to a pivotal component of organizational strategy”

Excluding finance from S&OP, planning is limited to input-output dynamics, overlooking crucial aspects of business operations. Finance’s involvement is essential for a holistic understanding of the business ecosystem, as it ensures that decisions are based on financial realities. Collaboration between S&OP and finance bridges gaps in comprehending interdepartmental synergies, facilitating informed decision-making.

A synchronized approach that seamlessly integrates forward-looking sales and operations plans with financial considerations is essential for effective budgeting. This alignment ensures that planning decisions directly impact financial outcomes, fostering organizational coherence and fiscal responsibility. By fostering collaboration between finance and S&OP stakeholders, businesses cultivate a comprehensive understanding of operations, enabling informed cost management decisions essential for sustained growth and profitability.

Benefits of Integration

It’s important to know how to talk about business in the Executive S&OP step by turning plans for numbers into plans for money. Executives decide what to do based on how the results will affect the organization’s health and its bottom line. They talk about EBITDA, P&L, and cash flow, not estimate error, units, and capacity. When Finance and S&OP work together, you can talk to leaders in their own language and make sure that everyone has a better understanding of the plans.

1.Alignment with Financial Goals: Finance being a part of the S&OP process makes sure that decisions made by the business are in line with its overall financial goals. When financial factors are taken into account in strategic planning, the S&OP process becomes a unified force that drives the company’s goals.

“Finance’s role in S&OP gives it a strategic view”

2. Strategic View: Finance’s role in S&OP gives it a strategic view by consistently predicting key business drivers, using predictive analytics, and incorporating up-to-date sales data. By adding financial information, the organization’s strategic path, risk assessment, and upcoming opportunities can be shown more accurately.

3. Collaboration: S&OP is naturally a process that involves people from different departments, and adding Finance breaks down silos and encourages people to work together. Different departments can make choices that are in line with bigger financial goals if they work together. Finance helps make it easier for people from different departments to work together and takes budget limits into account when making long-term plans.

4. Flexibility: A flexible financial plan is important in today’s fast-paced business world where market conditions, customer tastes, and world events are always changing. Adding Finance to S&OP, which focuses on ongoing planning and rolling forecasts, helps businesses respond quickly to changing conditions by making sure that monthly predictions are in line with financial plans.

“When Finance is added to S&OP, it makes the whole company take the same approach”

5. Unity Across the Enterprise: When Finance is added to the S&OP process, it makes the whole company take the same approach. By making a monthly Profit and Loss (P&L) and rolling forecast, businesses can better understand the factors and drivers that affect different areas. This breaks down barriers and promotes a more unified work culture. Long-term resilience in the face of uncertainty is helped by this unified method.

How to Integrate Finance Into S&OP

• Monetizing S&OP Plans: The first step towards integration is to monetize S&OP plans. This involves translating plans into financial terms and ensuring that all participating S&OP functional leaders have monetized plans to run their respective areas effectively.
• Alignment of S&OP Design with Financial Management: It is very important that the S&OP design is in line with how the business handles and reports its finances. This alignment ensures the easy addition of financial factors to the S&OP structure.
• Speaking the Same Language: The S&OP structure needs to be able to communicate with the finance team. We must change S&OP measures from volumes to values, inventory to working capital, and resource use to return on assets in order to achieve this.
• Finance-Led Variance Discussions: Finance can be very helpful when it comes to leading budget difference conversations, asking important questions, and speaking up when needed. This collaborative approach ensures the consideration of financial concerns during decision-making.
• Joint Objectives for Financial and Operational Departments: Setting shared goals for the finance and operations teams gives them a direction to follow, makes their goals more aligned, and encourages them to work together.

Conclusion

Including Finance in the S&OP process is not only the right thing to do; it is also a must for businesses that want to do well in today’s business world. For organizations striving for excellence, transitioning from understanding how to get finance in S&OP to effectively leveraging this partnership for strategic success is critical. By harnessing the synergies between finance and S&OP, organizations can achieve greater agility, profitability, and long-term resilience. This is an important step on the S&OP maturity journey.

BF’s new book Practical Guide to Sales & Operations Planning is a fantastic resource to learn best practices in S&OP and IBP from world-leading planning experts. You’ll learn how to start an S&OP/IBP process, progress it along the maturity curve, and use it to drive effective decision making that has a direct impact on KPIs like inventory turns, forecast accuracy, cash flow, customer service and more

Forecasting is part of everyday life. We watch the weather channel before we make weekend plans and text our family our ETA when we leave work. However, forecasting in the business context does not happen naturally. It is the responsibility of the demand planning team and as far as the rest of the company is concerned, how we arrive at these forecasts is a mystery. However, it doesn’t have to be this way.

For companies just starting out in demand forecasting, I would like to offer some very practical advice.

Forecasting anything is about determining mathematical dependencies between variables. The dependencies can be linear or nonlinear. Typically, having more data means you have a better chance at figuring out those dependencies and, as a result, developing a decent forecasting model.

This doesn’t mean that Neural Network or any other advanced algorithm. In fact, a simpler model is often better.

Simple Models Mean Buy-in From Stakeholders

First of all, if you have a choice between a linear and nonlinear model, choose linear even if it means losing a few percentage points on accuracy. The main reason is that a linear model (e.g. a regression)  is easy to translate into a formula and can be easily understood by stakeholders.

If your team can’t explain the predictions, they have no value. There will be no user buy-in, no follow-up questions, no discussions. The wise consumer of a forecast is not a trusting bystander but a participant and, above all, a critic. It is almost impossible to offer any meaningful criticism to something that is difficult to understand.

As your team develops a forecasting model their ultimate goal is to come up with something that stakeholders will digest and retain. Those should be simple things like “if X goes up 10% our sales are likely to go up 5%”.

Secondly, while having external drivers as components is a potential game-changer in the forecasting world, it is absolutely not required to get started. The best foundational model can be achieved by creating drivers (features) that are rooted in the time series data itself. Some examples of those features can be month number, quarter number, and a rolling average value for a number of previous periods.

Source data (price by month) and the three features engineered for the simplest forecasting model

Having 100 external drivers (oil prices, labor market statistics, search word frequency, etc.) might look good on paper and result in a higher accuracy, but the business stakeholders will likely be bewildered and close the deck, never to open it again. The optimal number of causal relationships is between three and five; this way stakeholders can actually remember what they are.

Aim to Be Directionally Correct, Not Perfectly Accurate

The third and final point is that being directionally correct is the most important forecast characteristic. It is a very intuitive one, but it is often overlooked in the data science world. To illustrate this, let’s evaluate three different forecasts, and we’ll use Root Mean Square Error (RMSE) – one of the most common forecast accuracy metrics – as a way to compare them.

Comparing three forecasts

In the world of data science, the lowest RMSE wins. So would we pick forecast 3 as  the best one in this case? Not so fast. Out of the three models here only one correctly indicates a downward trend for the month of February. There are lots of use cases where being directionally correct is far more valuable than landing on a value that’s closest to the actual one. For example, in a Demand Review, the knowledge about a downturn in the market is a powerful weapon to wield. With that being said, in this case we would choose Forecast 1 as the best one.

An actionable forecasting model has stakeholder buy-in, is explainable and directionally correct.

 To sum up: an actionable, practical forecasting model is not the one that uses the highest number of external drivers, has the most advanced mathematical algorithm or even the highest accuracy metric. It is the one that has stakeholder buy-in, is explainable and directionally correct. This way your team can be sure that it will drive meaningful discussions and result in actions that bring value to the business.

In many companies there often appears to be a difficult relationship between the sales and demand planning teams when it comes to the plan numbers. While some tension between these teams is inevitable, it seems to me that perhaps we are viewing this relationship incorrectly.

Based on my several decades of experience dealing with salespeople, I think many people – including Demand Planners and Managers – do not really understand that salespeople are not the enemy when it comes to planning.

They can, in fact, be your best ally. If you understand how the salespeople tend to think and operate, and the environment they operate in.

Every Salesperson’s Goal is to Make Plan

Salespeople are hired to sell a certain volume of product at a specified margin percent each year. If they meet the plan numbers for a year, they get another chance to do the same for another year. If they fail to meet the plan for a year, they are then under more pressure to make the numbers the following year.

Most plans increase incrementally each year, so every year the salesperson must be more creative, more focused and motivated to make the new numbers. Effective salespeople are always hunting for new opportunities to help them make plan.

Salespeople as Sharks

If we think of salespeople as sharks, we can get some insights into why they might seem like an enemy in planning. Sharks are hunters, apex predators with few enemies.

To be successful as a salesperson, a you must think and sometimes act like a shark.

To be successful as a salesperson, a you must think and sometimes act like a shark. Swimming slowly through the sea of sales opportunities, constantly searching for the next sale, exploring new areas, and acting quickly when an opportunity appears. Their only real competition is other sharks, that is, competing salespeople.

You Cannot Tame Sharks, But You Can Learn to Feed Them

I often see Demand Planners and salespeople arguing over the details of plan numbers, and in most cases, this is both useful and inevitable. However, this should not be viewed as something negative. We want plan numbers that are the result of honest deliberation. Where this process can derail is when each side sees the other as an opponent when, in fact, they both want the same goal – sales growth. So learn to feed the sharks.

Learn to feed the sharks.

Share every piece of relevant information you can with them. And do not limit yourself only to data available within the company. POS and inventory data are nice to have, and in fact necessary to guide a business. But include news about the companies that the salespeople are serving, and that they may not have time to review on their own.

Significant changes in location counts, staffing, programs of competing suppliers (including promotions), management changes and company performance are all useful pieces of tactical information that can help a salesperson judge when and how to approach a customer with a sales opportunity.

Ask to See the Math

While what I have said so far might seem like I think Demand Planners should always follow the sales team’s direction, there is one fact that Demand Planners need to ask when a salesperson proposes a new plan.

Show me the math.

Show me the data that you used to get to the numbers you want to use. Do not make me use your numbers just because they “feel” right, or because you need these specific numbers to make your plan.

Keep it real. After all, the Demand Planner’s key job is to make sure that what is planned actually gets sold. A demand plan is a request for product. A sales plan is a map to meet the sales goal. Both need to be based on realistic math that shows a clear path to the goal.

Above All, Build Solid Relationships with Salespeople

Effective sales are based on good relationships. We tend to buy mostly from people we know and trust. Effective planning is equally dependent on solid relationships. This means we can disagree with each other without becoming disagreeable.

We can disagree with each other without becoming disagreeable.

We can playfully challenge each other and play hardball when we see the other side gaming the numbers or hiding information. And never try to prove that the other side is “wrong”, as this can permanently damage the relationship and prevent future sharing of information.

Let the Sales Team Be Your Teachers

Good salespeople are in regular close contact with their customers. They know what drives their customers and what they need. If you are a Demand Planner, learn to regularly ask them about their customers and their business. They often know things about their customers that can help you with your planning.

Are their customers over inventory against their plan or open-to-buy? Are there buyer changes coming? Is the company in merger talks or under financial stress? Are they planning to repeat last year’s holiday promotions again this year? This kind of information can lead to especially useful discussions about how to plan future business.

Sales is a Game, and You Both Need to Win – But Not at Each Other’s Expense

Collaboration is often more difficult than merely playing to win. It requires more effort. However, in the long run, it produces more wins for more people, and helps support ongoing relationships.

So get to know the sharks that make your company successful. Feed them what they want and help them find the opportunities that will make you both successful.

I recently read Bob Stahl’s newest book, Sales and Operations Planning – An Executive Update, and I came away with a different perspective on a long-time problem; how to get consensus on a challenging forecast.  

Over the course of my long career, I have been part of, or facilitated, more than a thousand consensus meetings. And while most of these sessions generated little to no organizational tension, there have been times when it has been particularly difficult getting different parties (Sales, Marketing, Finance) to agree on a forecast. Under normal circumstances, early in the year and new product forecasts tend to cause the most tension because of the significant commercial ambitions loaded into these plans.

However, even these plans are often malleable with sufficient supporting data and conversation. The most difficult consensus challenges are always those forecasts that are most speculative, with little supporting data or with the greatest uncertainty.

Finding Consensus During Demand Chaos

COVID created forecasting chaos for many organizations. Tension increased during consensus meetings, especially during the early phases of the pandemic when, as an example, the fortunes of different product families were trending in opposite directions. Demand felt out of control.

“It was as if the pandemic froze us into inaction.”

The once-in-a-lifetime disruption confronting all of us made it hard to arrive at a forecast that everyone could agree on, despite having considerable supporting data. And for those product families for which orders and POS activity were down, arriving at consensus often seemed more difficult. No one wanted to “give-up” on the forecast so early in the year – especially given the unknown nature of consumer behavior in disruptive times. It was as if the pandemic froze us into inaction.

How I Handled Disagreements During COVID & What I’d Do Differently

When I was faced with the inability to arrive at consensus for many of the declining categories, I found myself proposing a simple approach that short-armed the forecast. I suggested looking at only at the next two to three months—acknowledging the reality of a short-term decline–while also holding the outermost forecast range to prior expectations.

We then provided a growth ramp back to the original forecast. It was a cheat of sorts. We did not “put the moose on the table” as Bob Stahl might have suggested but the short arming allowed us to move forward, effectively kicking the can down the road to the next month when better or more confirming information might be available.

While this tactic felt right in the moment, it also tossed out the window some time-honored S&OP concepts regarding managing the depth of horizon of a forecast. And while it is hard to call this approach a mistake, as we were in dark and unknown waters at the time, in hindsight it would have been better to press the issue more—to lean less on the crutch of uncertainty and rather push each member of the consensus group for their best (in this case, lowest) call.

“Start with a plan that everyone can roughly agree on, and then further challenge the assumptions.”

Instead, we did not so much collaborate on a plan; it was more like we ducked for cover. Which brings me to Bob’s book, in which he makes a pragmatic point that really resonated with me: Start with a plan that everyone can roughly agree on, and then further challenge the assumptions of that plan to get further clarity.

The ‘Greatest Common Denominator” Approach to Planning

Think of this as almost a “greatest common denominator” approach to planning. Effectively, the consensus facilitator starts by asking everyone their estimate and supporting data before trying to seek agreement. For example, in the face of double-digit declines ask, “Does everyone agree the forecast should come down for the year ?” Then follow that up by asking, “By how much, and how would you pace the decline?” By asking relatively open ended questions all voices and opinions are heard, and the range of perceived opportunities are dimensioned.

After reading Bob’s book, it became apparent to me that by putting in a short arm “device” we avoided much in the way of thoughtful conversation. We did not seek common ground. I know this because nearly everyone walked out of the consensus meeting thinking that the forecast should have been lower. We did not reach consensus – we only postponed the hard decision by four or five months when we finally made the hard calls needed to reset the forecast lower.

Some Conflict is Normal is S&OP – Embrace It

Most long-term S&OP practitioners know all too well that at least some level of tension, conflict, and disagreement are normal in consensus meetings. In fact, some disagreement within the S&OP process is to be expected and perhaps even encouraged. No one wants an S&OP plan put together via groupthink and without some rigor of organizational tension applied. Unfortunately, in the midst of COVID, we avoided this tension.

“No one wants an S&OP plan put together via groupthink.”

One of the most important lessons to come out of the COVID crisis (and Bob’s book) is to solicit more opinions and points of view as a way to put more voices into the forecasting process before trying to arrive at an agreed number. Let the opinions of the consensus team come out and bloom to see if there is a unifying or common perspective before trying to narrow the forecast. Disagreements over outcomes earlier in the pandemic would have helped avoid “chasing the forecast down” phenomena that ultimately occurred.

If history offers a lesson, avoiding tension is, well, wrong. If we believe disruption will become more common place, this is a lesson worth learning.

To get up to speed with the fundamentals of S&OP and IBP, join IBF for our 2- or 3-day Boot Camp in Miami, from Feb 6-8. You’ll receive training in best practices from leading experts, designed to make these processes a reality in your organization. Super Early Bird Pricing is open now. Details and registration.

As the saying goes, what gets measured gets improved. The same is the case with supply chain management. End to end supply chain operations related to planning, sourcing, production and logistics have specific metrics that help us improve processes and overall performance.

For every metric, there is a target number or goal that needs to be agreed upon. Then, each of those metrics is linked with a strategic objective or a key value driver (KVD), and is further supported by planned initiatives that help in achieving the target or goal.

Demand Planning Metrics

Demand planning metrics relate primarily to forecast error metrics. They include:

• Mean Absolute Percentage Error [MAPE]
• Weighted Absolute Percentage Error [WAPE]
• Mean Squared Error [MSE]
• Forecast Bias, and
• Forecast Value Added [FVA]

Of these, Forecast Bias and FVA are critical and should be tracked over the planning time horizon. Bias indicates the direction of the forecast errors and helps us understand whether we are under-forecasting or over-forecasting.

FVA is a measure of the value added during the forecasting cycle. It is measured by comparing forecast accuracy before and after each activity in the demand forecasting and planning process to determine if that activity improved the accuracy of the demand forecast. Tracking FVA during the forecasting cycle enables us to find the right mix of quantitative and qualitative approaches that result in fewer errors.

Supply Planning Metrics

Supply planning metrics are related to sourcing and production. They include:

• Order Confirmation to Delivery Lead Time
• OTIF % [On Time in Full]
• Material Availability for Monthly Production Schedule
• % Rejections and % Returns
• Spend Analysis [By Product Mix and Supplier]
• Direct and Indirect Category Analysis and
• Supplier Performance Index

These metrics allow us to understand supplier performance and to perform spend analysis.

• Production/Manufacturing metrics include:
• Daily schedule adherence
• Production loss
• SKUs with excess production
• SKUs with shortage in production
• Line Utilization %,
• OEE% (Overall Equipment Effectiveness)

It is essential to segregate the metrics into daily, weekly, monthly and quarterly buckets. However, the most crucial performance indicator is OEE % as it gives a snapshot of equipment productivity, product quality, utilization and speed.

S&OP Metrics

During S&OP meetings, the appropriate metrics are reviewed and discussed with the respective functional teams during each phase or step of the S&OP cycle.

Logistics & Distribution Metrics

This section covers metrics related to warehousing, transportation and distribution operations.

Warehousing metrics

•  Dock to Stock Cycle Time
• Inventory Accuracy %
• Operating Costs [Budget vs Actual]
• Detention Costs
• MHE Utilization %
•  Process Productivity Metrics
• Order-Delivery Cycle Time and
• Environment-Health-Safety Metrics

The metrics that are critical and need extra attention are the ones related to inventory management and customer order management. Accurate inventory management within the warehouse is critical to ensure correct order picking, packing, staging and loading operations. End to end order management from order receipt to fulfilment to proof of delivery are central to warehouse operations.

Transportation metrics

• Truck Placement Reliability
• Loading Time
• Unloading Time
• Transportation Lead Time
• Operating Costs
• Vehicle Capacity Utilization %

In the case of transportation, vehicle arrivals at the appointed times are critical for timely onward operations. Monitoring transportation lead times are also essential, especially for last mile deliveries.

Distribution metrics

• Despatch Schedule Adherence
• OTIF [On Time-In Full] %
• Customer Order Confirmation to Delivery Lead Time
• Product Returns %
• Customer Satisfaction Score

As an extension, customer satisfaction and product returns metrics need to be measured and monitored. Most 3PL and 4PL service providers invest in processes and systems related to customer issues. Order fill rates and customer satisfaction metrics need to be focused upon and prioritized.

Other Relevant Metrics

The advent of e-commerce and quick commerce has resulted in ever-increasing reverse logistics and material flows. Therefore, metrics related to reverse flows are as important as forward flows. Also, macro-economic factors, geopolitics, climate events, trade flows, fluctuating commodity prices, etc. have necessitated the measurement and monitoring of resiliency metrics.

Reverse logistics metrics

• Logistics Costs and Product Returns as a % of Sales

The challenge here is to minimize or reduce product returns and therefore reverse logistics costs. Just like the forward supply chain, designing the reverse supply chain is essential in ensuring that returned products can be inspected, segregated and reused, repurposed or refurbished depending on the type of product

Resiliency metrics

• Time to Survive (TTS)
• Time to Recover (TTR)
• Financial Impact (Sales and Profitability)

Resiliency metrics became extremely important for supply chains following the COVID-19 pandemic. Companies soon realized how every supply node in the chain could affect downstream activities due to material shortages and supply delays.

For firms that had a global network of suppliers, the ‘Time to Recover’ took several weeks and months. Moreover, supply related constraints had an adverse impact on demand that eventually affected sales and profitability numbers. Firms have started to invest in tools and systems that enable ‘what-if’ scenario analysis – what the financial impact might be in case a supply node is unable to perform as planned.

Linking Operational & Financial Metrics

The various operational metrics and performance indicators listed above have a direct or indirect impact on business performance – more specifically, financial performance. Therefore, it is essential to establish a link between operational metrics and financial metrics.

The key financial metrics to focus on include:

• Cash to Cash Cycle Gross and Net Working Capital
• Inventory Turnover Ratio
• Gross Margin and Net Margin %
• Return on Assets (ROA)
• Return on Capital Employed (ROCE)

Cash to Cash Cycle, Gross and Net Working Capital and Inventory Turnover Ratio are the three metrics that should be on the radar of the supply chain management team. All the demand, supply and distribution plans and their execution have a direct bearing on the cash conversion cycle and working capital levels. ROA and ROCE are impacted by supply chain operating costs. Any reduction or saving in procurement costs, inventory handling costs, production and logistics costs shall have a positive impact on ROA and ROCE. The operational metrics can be dovetailed into a relevant business/financial scorecard every month, quarter, and financial year.

People, Processes & Systems

Finally, for any performance management philosophy to work, there is a need to put in place the right team with suitable skills and understanding of the supply chain domain in addition to devising and deploying the appropriate processes and procedures for data management (data recording, reporting, monitoring and analysis) and selecting the right tools, software and systems for reporting, analysis and visual management.

Visual management tools such as dashboards, functional control towers and related systems could go a long way in monitoring key operational metrics on a real time basis to enable course correction and development of corrective and preventive action plans going forward.

To get up to speed with the fundamentals of S&OP and IBP, join IBF for our 2- or 3-day Boot Camp in Miami, from Feb 6-8. You’ll receive training in best practices from leading experts, designed to make these processes a reality in your organization. Super Early Bird Pricing is open now. Details and registration.

Effective demand planning requires ongoing collaboration with many other teams. Sales, Finance, Supply and Operations — among others — depend on reliable demand planning data. Demand Planners frequently exchange data with members of these teams in order to improve forecast and sales performance.

And in many cases, members of these other teams will request specific data from the demand planning team. Salespeople want to review future forecasts; Finance wants to know about promotional plans; Supply may question a future number that seems unusually high. So, in many cases, this exchanging of information is key to the process of improving future forecasts. And since data from the demand planning team can impact many other teams’ performance, it’s important that Demand Planners collaborate effectively with these teams.

But it’s also important that the demand planning team manage the many requests they may receive in a way that doesn’t place excess strain team members. Pam in Finance hears how you created a custom report for Jan in Operations, and now Pam is asking for help with another report. Paul in Sales asked you for an updated forecast for one of his customers, and now his sales manager wants similar reports for all his salespeople. Over time Demand Planners can find themselves doing work for other departments, and in some cases these requests are valid and help to support the business. But it also easy for these requests to grow to the point where the Planners are not able to devote the time needed to provide timely and accurate forecast data.

How do we balance collaboration with other teams with preserving time for quality planning? Here are my five recommendations for keeping your planning team from being overwhelmed by outside requests — for defragging your planning process to maintain a proper balance.

1. Communicate in advance your team’s priorities so that the members of all other teams know in advance what requests are appropriate
2. Always ask if demand planning is the correct team to be handling the request
3. Always ask how a request will help improve forecast performance
4. Beware of sticky requests
5. Estimate the time required and whether the request needs to be handled immediately
6. Communicate in advance your team’s priorities so that all other teams know in advance what requests are appropriate.

In my work in demand planning, I have often found that the members of other teams are not fully aware of the purpose of my role. So when I explain to them what I do and how it impacts all the other teams involved in Supply, Finance, Operation and Sales, they may still ask for help. But in these cases, they now know that their requests need to be aligned with my central role in planning, and they are not offended when I refer them to another team for help. And I have appreciated the managers who regularly reminded all the teams that we worked with that our central role was not reporting but research and planning, and that good planning requires a clear focus and undisturbed time, and that requests that distract us from our central role hurt every one that depends on our work.

Ask if Demand Planning is the Correct Team for the Request 

People from other departments who ask a demand planner for help with information are often looking for help with issues within their own department, and may not be aware that their request is unrelated to planning. So, for example, asking a Demand Planner to research issues with past purchase order data might seem like a legitimate request to someone in Finance or Operations. However, in most cases, Demand Planners are not concerned with past orders. In addition, asking a Demand Planner to take time to do this will not help with planning future forecasts. In this case, it makes sense for the Planner to gently recommend that the requester contact a department such as Customer Service for assistance.

Ask how a Request Will Help Improve Forecast Performance

Since demand planning’s role is to provide the best possible forecasts of future performance, it makes sense for members of this team to question how requests for help from outside teams will help Planners improve forecast accuracy. A salesperson who calls for help reviewing a customer forecast can provide insights that can help the Planner with future forecasts, and collaborating here is part of the Demand Planner’s role. However, requests for help with managing purchase orders, pricing or late shipments is mostly outside the Demand Planner’s realm, and such requests need to be referred to the appropriate team.

Beware Sticky Requests

Some requests may not be directly related to demand planning, but since they are easy for Planners to handle, they help without questioning if they should do so. Where this gets them into trouble is when a single request turns into an ongoing request to provide information or reporting. I call these sticky requests. While collaboration is important, taking up a Planner’s time with ongoing requests that distracts them from their key role — providing the best possible view of future requirements — is harmful to everyone who depends on reliable forecasts. I have personally fallen into this trap, and in some cases I have had to ask my manager to inform the requester that I am no longer able to provide the requested information. A willingness to be helpful must always be balanced against managing the core demand planning responsibilities for all the teams that depend on our work.

Ask if the Request Needs to be Handled Immediately & Estimate the Time Required

When people ask for help with appropriate issues, it’s easy to assume that they need help immediately. I have learned to ask requesters when they actually need the information. Sometimes they need it immediately, but very often their need is not urgent. So, knowing when they need the information helps me manage my workload while also helping them. For my own planning purposes, I also estimate how much time and what resources I would need to allocate to each request, so I can give the requester an honest answer as to when I might have what they need. And when a request truly is urgent, take the time up front to get as much detail as possible about what the requester really wants. There’s nothing worse than dropping everything to help someone only to hear afterward that it was not what they truly needed.

The Balancing Act 

In my experience, most Demand Planners are highly skilled, they know their business, and they want to do their best to improve their forecasting skills. They know that other teams are depending on them for reliable forecasts. And it’s natural for people on these other teams to look to these talented individuals for help. The challenge is to balance collaboration and helpfulness with tactfully defending your time and focus required for truly effective planning. Defragging your demand planning processes and keeping them free of distractions is key to maintaining this balance, and allowing the demand planning talent to drive ongoing improvement.

This article first appeared in the fall 2023 issue of the Journal of Business Forecasting. To access the Journal, become an IBF member and get it delivered to your door every quarter, along with a host of memberships benefits including discounted conferences and training, exclusive workshops, and access to the entire IBF knowledge library. 

We all know that machine learning (ML) and AI gets the analytics and data science community excited. Every self-respecting forecasting department is developing ML algorithms to predict who will click, buy, lie, or die (to borrow the title of Eric Siegel’s seminal work on the subject). All analytics conferences and publications are filled with AI buzz words.

But when it comes to real-life implementation, the majority of demand forecasters are somewhat cautious about implementing machine learning. Why is that? Isn’t machine learning all about predicting, which is literally a Forecasters job? Let’s explore the opportunities and pitfalls of applying machine learning in forecasting.

Demand Forecasters & Data Scientists Define ‘Prediction’ Differently

There is a subtle difference in the way forecasting and ML define ‘prediction’. When Forecasters say ‘prediction’ we mean a prediction about the future. Traditional forecasting prediction methods include Time Series modelling, algebraic equations and qualitative judgement calls. As a result traditional forecasting is somewhat manual and time consuming, and may be swayed by human judgement. However, the outputs are easily interpreted and it is an agile process; the Forecaster knows where the numbers are coming from and may easily make corrections as needed. Further, traditional forecasting may be done with limited data.

Machine learning or statistical model ‘prediction’ refers to predicting the past. This sounds a bit counterintuitive, but the idea is to compare the model ‘prediction’ with reality and measure the difference or error. These errors are used to finetune the model to predict the future. Consequently, model predictions are heavily driven by past performance and are almost impossible to finetune. Also, the interpretability of models is very limited. Another factor to consider is that by design ML requires a lot of data. On the upside, machine learning is quick and automated as well as objective, being free from human judgement.

Machine Learning was Built for the Digital World; Forecasters Work in the Real World

Machine learning and AI algorithms were created for a digital world with almost unlimited data on customer clicks, purchases and browsing data. As we know, these algorithms do an excellent job in luring us to make repeat purchases, buy complimentary items, and sign up for loyalty programs. The sunk cost of prediction error (lost sales) is relatively low. In addition, every error is an opportunity for the machine learning algorithm to improve itself.

The real world marketplace is quite different from the digital, marketplace, however. The data here might be limited to cash register sales, loyalty program data, or shipment data. The sunk cost of prediction error can be quite high as restaurants and retailers make procurements in bulk. Also, predictions cannot improve themselves as there is no automatic feedback loop. For these reasons, many brick-and-mortar retailers and their suppliers still rely on traditional forecasting methods. This does not mean that Machine Learning cannot offer opportunities in improving forecasting but there are a few considerations that need to be addressed before venturing into machine learning.

Machine Learning Requires Much More Data Than Time Series 

Any machine learning algorithm requires a lot of data. By a lot of data, I do not mean dates or variables. Machine learning models run on defined observation levels—this can be customer, store etc. You need at least a thousand of those (if not thousands) for machine learning to work. If the sample is limited to only 10 stores, it is probably better to refrain from machine learning and use Time Series techniques instead. Another factor to consider is the cost of maintaining the data. Is it readily available or does it need to be inputted manually? Does the data need to be engineered? Would that be a one-time effort or an ongoing process requiring human and computing resources? What would be the cost of storing data over the years?

Machine Learning is far Less Interpretable than Time Series 

By design machine learning is a black box. For example, predictions may be generated by a vote of thousands of decision trees. You can use colorful histograms to depict the weight of each factor in the model. These charts look very smart on presentation slides but are very far from intuitive. If the cost of a wrong prediction is millions of dollars, companies might be more comfortable with Time Series and arithmetic they can understand rather than a slick black box algorithm. This especially applies to new products with no sales data or limited test data.

There are a few workarounds for understanding machine learning. Playing with parameters might be a good indicator of the robustness of results. If one slight change to model inputs or specifications results in significant changes to predictions, this might be a red flag.

At the end of the day model trustworthiness may be proved by testing on new data. We don’t necessarily need to understand the ins and outs of algorithms if we are confident in the end result. Robustness of this argument may depend on your audience. Typically, analytics professionals are comfortable using machine learning predictions as long as they are tested. Supply chain leaders might be more cautious in making business decisions based on black box. A good sanity check is to run traditional forecasting methods in parallel to machine learning. If there is feasible difference between the results there might be either an issue with the model or an important consideration was left out when creating a traditional forecast.

The Cost Benefit of Machine Learning is not Always Clear

It goes without saying that when machine learning is set up right it is wonderfully efficient. All one needs to do is provide inputs and press the button. The ‘setting up right’ piece might be relatively straightforward or extremely difficult depending on prediction goal and data available. Repeat products with abundant history may be easily predicted using even out-of-the box ML packages such as SAS or Azure as long as the data is readily available. New product predictions may require intricate proxy algorithms to solve for limited data. This may require development of ML algorithms from scratch. In addition, there may also be a need to engineer data from different sources to feed the algorithm. This might require significant investment to either hire contractors, expand analytics team or put pressure on existing resources. Before ramping up a data science crew, companies would be well advised to consider how often the algorithm will be used, the efficiency gains, and the computing resources required for the project.

Impacts on Overall Business Planning 

Forecasting is the cornerstone of business planning. Any changes to the forecasting process may have an impact on other areas of the business such as Finance and Supply Chain. Typically, traditional forecasting methods rely on a top-down approach. A forecast is created in aggregate and then broken down by store/time period, etc. These breakdowns may be later used for financial targets or demand planning at store level. By design, ML Forecast utilizes a bottom-up approach. A prediction is created at store/time period level and later aggregated. When switching from traditional forecasting to ML, companies must ensure smooth transition at all stages of business planning. If not done right, this transition may result in discrepancies between the ML prediction vs the financial targets and supply plans.

To summarize, ML is a great instrument to streamline forecasting. As with any tool, it has its applications, benefits, cost, and risks. When utilizing ML for forecasting, companies should consider their data, business need, decision making culture, and planning workflow. A great place to start might be trying out ML on your data using online, off-the-shelf solutions such as Azure and SAS. Most of these solutions have step-by-step training videos that will help fit an ML algorithm to your data. Experimenting with these solutions may help decide whether ML is a good tool for your company’s forecasting, and whether an off-the-shelf solution is sufficient or there is a need for in-house development. Even if it turns out that for whatever reason ML is not a good fit for your company, there is no investment lost and some analytical knowledge will gained.

This article first appeared in the summer 2023 issue of the Journal of Business Forecasting. To access the Journal, become an IBF member and get it delivered to your door every quarter, along with a host of memberships benefits including discounted conferences and training, exclusive workshops, and access to the entire IBF knowledge library. 

The following is the second in a two-part series covering the art and science of  demand planning. Read the first part on the art of planning here.

Why do we Need to Forecast?

Companies need to understand what the future will bring to make the right decisions: What is my optimal footprint? What is the right price based on future expected volume? How much capacity is needed, and which inventory investments will yield the best returns? Using a statistical forecast provides a good starting point to answer all these questions.

However, not everyone has the background to understand the math behind the process, the underlying assumptions used in the forecast, or the best way to integrate this information into demand planning decision making. This is why it is important to simplify the statistical forecasting principles and assumptions at a high level so everyone actively participate in the discussion and help build the demand plan.

The Science Behind Statistical Forecasting

“The only way to predict the future is to understand the present.” – Isaac Asimov

The first question that needs to be answered is: What are you trying to predict? Forecasting order entry or sales will provide different outcomes and maybe both are needed, but for different reasons. My experience is that to drive the supply chain process you are better off using order entry as an input, since sales includes variability related to supply availability. But if there is a conscious thought process behind the decision of what is being used, it should not be a problem. Let’s review a quick example to clarify this situation.

A few years ago, during a demand discussion among colleagues, we were seeing a high level of volatility in a certain SKU. The Demand Planner was not able to explain it well as the item had steady sales in the past. Then the Supply Planning Manager stepped in and mentioned that we had a recent shortage in this product and the peak in sales should be due to a big lot coming in and being shipped out in the same month.

These circumstances lead us to review the process and the data we were using, as we were trying to determine what we could sell if there was not any supply constraint. The best set of data to do this was orders entered with the requested date from the customer. Once we used these figures, the pattern was much smoother as order entry had the information of what the customer needed, and the sales information had the information of when we were able to provide it. These subtle differences in the input of the process provide a very different output.

Now we can move to discuss the process. The intention of this section is to understand from a ten thousand foot view the components of the models and the high-level assumptions behind them. There will be complex calculations involved and we will not get into the details behind the math. After all, the experts can handle this very easily, but it is the understanding of these concepts in general that helps us to drive constructive discussion and decision-making.

There are a lot of different methods that try to predict the future including a simple moving average, exponential smoothing, econometric models, linear programming methods or machine learning algorithms. We will focus on time series – a series of sequential data points ordered by time. The two main methods we will review are exponential smoothing and linear regression because they are the easiest models to understand and the most widely used.

Exponential Smoothing

The first technique is called exponential smoothing. This method uses historical information to decide how much weight to give more recent history, trend or seasonality. These are the three main factors that you want to spend some time reviewing as they will give you some insight into the current demand pattern. The math will tell you how important each of these is, but it is up to people that understand the market to explain why.

This type of modeling is very helpful as it is data driven. I remember a meeting where a person on the marketing team mentioned that a certain product family was seasonal and after reviewing the data, it turned out that this was not statistically true. After digging into the details, it turned out that some products in that category had a strong seasonality component, but at the aggregate level you could not distinguish that pattern.

Another way to predict the future is to tag demand to macroeconomic variables and see if there is some correlation between them. A good indicator used in the plumbing industry is housing starts since it is a leading indicator of economic activity that tells you how many more homes are being built. There are a lot of different economic indicators out there and probably one of them is a good fit for your industry. Finding this correlation would be very useful since there are public forecasts available that can be used to predict the demand of your products.

Underlying Assumptions & Caveats

“History never repeats itself, but it does often rhyme” – Mark Twain.

What happens in an environment where there is volatility, uncertainty and complexity? Well, some time is spent discussing the data but this is usually not enough to understand all the caveats and assumptions.

One of the most important principles in forecasting is the underlying assumption that history will repeat itself and that past information can provide a good representation of what will happen next. The math tries to find certain patterns hiding in the data, like determining if the recent past is more useful or if there is a trend or seasonality involved. Early on in my career, when forecasting a high-volume item, we got a very high forecast for the near future, which did not made sense. It turned out that we had a big promotion in the recent past that was skewing the numbers. Once we took this out, the forecast corrected itself. The lesson we learned was that it is very important to scrub history to find outliers in the data.

Another important assumption is that external factors in the environment will remain constant, allowing the forecast to be developed under similar circumstances every time. Variations in the industry outlook, regulatory changes or economic growth fluctuations provide a challenge to the models described above. An alternate way to incorporate this information into the demand plan is required in these cases. A recent example of this is that after years of economic growth, the economy is stagnating or even shrinking. While this is known and discussed in the news and at the watercooler, there is always a lag from when this starts to happen and when the forecasting model picks it up.

Finally, a few important factors to consider in the assumptions are the horizon that you are using to plan and the level of aggregation. Think about the weather for a moment – usually the forecast for tomorrow is very accurate but looking at any day next month is not worth it. The same is true for any type of industry forecast; the further your look out, the less accurate it becomes.

It is a similar situation for the level of aggregation. It might be easy to predict how much you will sell in dollars for next month but it is harder to determine how many dollars per SKU or per customer you will sell next month. This is important to understand, because I have faced situations where the business asks how much we will sell in five years at SKU level by customer. It does not make sense to do a detailed analysis in this situation.  It is better to provide a directional number that is easy to explain and that supports the business to make the right decisions.

Measuring The Output Of The Process

Some of the most popular phrases in demand planning are “If I could predict the future, I would go to Vegas” or “The forecast is always wrong” and, my favorite one, “My crystal ball is broken”.  It is a fact that no forecast will be exact, but you can get within a decent range and with a good level of confidence. This is why it is good practice to measure the accuracy of your statistical forecast to understand the reasons behind your top misses.

From my perspective, it is important to understand how good the forecast accuracy is in terms of mix and volume. A good way of measuring mix is through MAPE (mean absolute percentage error), which basically tells you by how much you missed the forecast – regardless if you missed up or down – compared to what your actuals were. The advantage of this metric is that it is easy to understand (since it is a percentage) and it does not net out negative and positive values as it uses an absolute error.

The way to measure volume is through understanding if there is a bias at the aggregated level of your forecast. For example, when aggregating all your forecasts at the total dollar level by month, if you find out that the actual sales number has been below the forecast for some time, you might want to understand the reasons behind this. Ideally you want to oscillate between being a little above and then the following period a little below over the time horizon.

Discussing The Forecast During The Demand Planning Meeting

After understanding how a computer (or a Demand Planner) creates a forecast, it should not be a surprise why this needs to be reviewed in a group setting. Even if math is not your strength, a lot of value will come from having a thorough discussion of the historical data, the process used to come up with the numbers, the actual forecast, and metrics.

Below is a checklist of things to review. During this discussion, make sure that you balance time between the items that will move the needle, but covering in enough detail that allows you to understand the current situation.

• Start with the input. Are we using the right dataset and have we looked at history to review and scrub?
• How much importance do we give to more recent history versus the past? Is there a pattern in the data to be discerned, like trend or seasonality?
• Is there an economic indicator that we could peg to a group or family of SKUs that will help us determine what the future holds?
• Is there any external factor that could deter us from using our forecasting methods to determine the future values? Examples of this are changes in price (by us or the competition), changes in the economic or regulatory environment, or even new products that could cannibalize current demand.
• What are the forecast accuracy metrics telling us? Are we better than last month? What are the top offenders and why?

In conclusion, a good statistical forecast provides a good start for your demand planning process and a solid foundation. The data, assumptions and metrics need to be understood and discussed in depth. But keep in mind that this only the beginning of the process. There is an art component of the demand planning process that incorporates changes in the environment, market intelligence and in general a consensus between areas that should be aligned with the company strategy.