How to Forecast Market Demand

Successfully forecasting market demand helps ensure you have sufficient stock for upcoming sales periods. Market demand forecasting involves using past sales data to predict future consumer needs. Accurate demand forecasting leads to more effective business operations, improved customer service, and reduced production time. It helps companies avoid high operational costs, poor customer service, and product shortages.

Identify Specific Products. Rather than focusing on an entire product line, identify the specific products you want to track. This makes organizing past data and forecasting future demand easier. For instance, if you have a winter clothing line, start by focusing on gloves instead of handling the whole product line.

• Focus on the products that generate the highest revenue. For example, many businesses follow the 80/20 rule, meaning 20% of their products or services generate 80% of their revenue. Identify and track demand for those.
• You may need to forecast demand for all existing products. However, forecasting demand for a small group of similar items like gloves, boots, and winter hats will be easier and more accurate.
• Consider forming a Sales and Operations Planning team, including representatives from each department, responsible for preparing product demand forecasts.

Review marketing plans. Any marketing or promotional campaigns can potentially boost the demand for your products. Look at past data and assess where you have been successful. Consider whether a special discount or a holiday promotion helped drive product demand. It is important to take all factors into account when forecasting demand, especially if you plan to repeat a sales strategy you have used before.

Review key metrics. Identify the causes of fluctuations in customer product demand. Key metrics include demographic and environmental factors. Demographics encompass age, gender, geographical location, and any other identifiable characteristic groups. Understanding the demand of key demographic groups helps narrow down the data range for forecasting. Environmental factors also influence demand. For instance, an extremely harsh winter might lead to a drop in sales.

Look at the market. Analyze the actions and statements of your competitors, customers, banks, and anyone else involved in your market. Consider whether your competitors are running a major promotion or sales campaign.

Consider past months. Look at previous months and the yearly sales variations, such as during holiday seasons. This will help you identify annual and seasonal trends. When reviewing past months, analyze the factors driving demand. Don’t overlook any price adjustments or marketing campaigns that helped attract new customers. There’s always a reason behind a surge in business activity, and a savvy entrepreneur will uncover it. For example, you may have run a 'buy one, get one free' campaign in August for 'Back-to-School' products. If you decide to reuse this strategy, factor it into your forecast.

Determine delivery time. Delivery time is the period from when an order is placed until the product is delivered. This information will assist you in forecasting demand. By knowing this, you can estimate product completion speed and meet market demand.

• If you are purchasing from another company, delivery time is measured from when the order is placed until the products are delivered to your location.
• You can also calculate delivery time by assessing raw materials and components needed for production. Knowing the production time required helps you forecast demand more accurately. Focusing on a specific item will allow you to estimate the amount of raw materials needed and the production time required.
• Once you have forecasted production quantities, consider the demand for each product. For example, if producing pencils, you will need to know the amount of wood, rubber, and other required materials based on your estimate.

Identify the method to be used. In general, there are four main methods for forecasting demand. These include judgment, experimental, correlational/causal, and time series methods. Based on the product history, choose the most appropriate method. For instance, the experimental method is often used for new products with no market data history. These methods represent how you will collect the majority of the data you need.

• You can combine several methods to forecast demand more accurately.

Consider the judgment method. This method determines demand based on general market knowledge observed by the sales team and management. With their experience and insights, they can make forecasts with a certain degree of accuracy, and in some cases, their forecasts may be highly accurate. However, the data gathered from this source may be unreliable as it depends on their personal viewpoints. Therefore, it should only be used for short-term demand forecasting.

• There are various ways to implement this, mainly depending on the team. However, you do not need to use the entire list of people. You can choose any combination to achieve the goal, depending on which expert group you believe will provide the most accurate judgment.

Decide the necessity of the experimental method. This method works best for new products and is not useful for products already on the market with an established demand history. It uses results from a small group of customers and extrapolates these conclusions to a larger customer base. For example, if you randomly contact 500 people in a city and 25% of them indicate they will buy a product in the next 6 months, you can assume this ratio applies to a population of 5,000.

• If the small group of target customers likes a new technology and responds well to trial marketing, you might conclude that this figure also forecasts national demand. The issue with this method is that it often gathers customer opinions rather than actual demand data.

Consider using the correlational/causal method. This method focuses on identifying why people buy your product. The idea is that if you understand why someone buys the product, you can build a demand forecast based on that motivation. For example, if you sell boots, you know that product demand is influenced by the weather. If the weather forecast predicts an extremely cold winter, you could conclude that the demand for your boots will increase.

• This group of methods also includes product life cycle and simulation models.

Calculate demand using the time series method. The time series method relies on using past data, trends, and mathematics to estimate demand. Specifically, you can use moving averages, weighted moving averages, and/or exponential smoothing to accurately forecast demand. While these methods offer stronger results, they must be combined with other methods and subjective evaluations to account for the impact of market changes or business plan adjustments.

Form an executive assessment. Gather a small group of senior managers within the company and ask them to provide their demand forecasts. Each member can offer valuable insights from their market experience. They can also help select marketing campaigns and quality suppliers. This method is not as expensive or time-consuming as other judgment-based methods. However, its weakness lies in the fact that it relies on expert opinions, which can be biased and may promote their own agendas.

Aggregate salespeople's opinions. Ask each salesperson to forecast their own sales. Salespeople are the closest to the market and have a deep understanding of customer needs. Combine these forecasts from each sales level by city, province, and region. The advantage of this method is its low cost and ease of data collection. The drawback is that it depends on customer opinions, which can easily change. Additionally, salespeople may inflate their numbers to secure their position within the company.

Hire individual market experts. Market experts observe industry trends and consult with your sales team to forecast demand. These may include business magazine writers, economists, bank directors, and professional consultants. However, the amount of information one individual can gather is limited. Therefore, you should assemble a team of market experts to collect as much data as possible.

• Compared to sales teams, these individuals can provide deeper, more refined insights into the market. However, as outsiders, they won't understand your product's specific needs as well as your company's employees. You should use these experts to forecast overall market demand and then rely on internal judgment to estimate your company's potential success in that market.

Use the Delphi Method. Start by forming a panel of experts. This can include managers, selected employees, or industry specialists. Ask each participant to provide their demand forecast. Have them respond to a set of questionnaires over two or more rounds. After each round, anonymously present the results of the previous round. Encourage experts to revise their answers based on the results obtained from earlier rounds. The goal is for the group to converge on a consensus forecast by the end of the process.

• Set a clear endpoint, such as a specific number of rounds, a consensus point, or stability in the results.

Conduct customer surveys. You can gather information from customers through various means: phone calls, emails, reviewing historical order statistics, or analyzing market trends. Ask about their purchasing plans and subjective buying behaviors. Use a large sample size to generalize the results. Inquire about the likelihood of them purchasing your product and then aggregate the responses.

• Customers are the best source of information on the demand for a specific product. The risk with this approach is that surveys often overestimate actual demand. While someone may express interest in a product, it doesn’t guarantee they will make a purchase.
• Keep in mind that surveys can be expensive, challenging, and time-consuming. They rarely provide a solid foundation for demand forecasting.

Use experimental marketing. This method is ideal in the early stages of product development. Find a small, isolated region with a demographic profile that matches your target market. Implement all the steps in your marketing plan, including advertising, promotions, and distribution strategy. Measure product awareness, market penetration, share of market, and total sales. Adjust your strategy based on the insights gathered to minimize issues when launching the product nationwide.

Invite a focus group of customers. Gather a small group of potential customers in a room, have them try your product, and engage in a discussion about their experience. Participants are typically compensated with money or small gifts for their involvement. Similar to surveys, the data obtained is more useful for analyzing the product itself rather than forming the basis for demand predictions.

Use scan data. Identify a large group of customers who agree to participate in an ongoing study about their shopping habits, such as at grocery stores. Convince these customers to provide information like household size, age, income, and any details relevant to your product. Each time they purchase groceries, their purchase information is recorded and analyzed. This data can be collected when they use loyalty cards. It provides a rich database for statistical modeling and uncovers relationships in the data.

• As with other experimental methods, forecasting demand from this data can be challenging.

Review monthly sales or seasonal trends from past years. Look at sales figures from previous years to identify periods with higher sales percentages. Are these periods stable? Are sales higher in winter or summer? Measure any fluctuations during those times. Were the changes greater or smaller in certain years? Next, consider potential causes behind these fluctuations. Use what you've learned to apply predictions for the current year.

• For example, if you sell boots, sales may have been particularly high during cold winters. If this year is predicted to have a similarly cold winter, you should adjust your demand forecast accordingly.

Examine customer reactions. This involves observing how changes in products or the market affect sales. Create a sales chart over time and highlight significant dates, such as price increases or the release of a competing product. It can also be broader, such as responding to economic shifts or changes in personal consumption patterns. Reading relevant business magazines and articles can provide additional insights. The more information you gather, the clearer you'll understand which factors may influence future product demand.

Build a product life cycle model. A life cycle represents the "life" of your product, from its initial introduction to the present. Analyze sales at each stage. Consider the characteristics of customers who purchased during these phases. For instance, you may have a group of early adopters (those who love the latest technology), a group of followers (those who wait for the product to be tested and reviewed), laggards (who only buy after the product has been on the market for a while), and other customer segments. These groups help identify life cycle trends and demand patterns for your product.

• This model is most commonly used in industries such as high-tech, fashion, and short-life-cycle products. What makes it unique is that the demand is directly tied to the product’s life cycle.

Use simulation models. Create a simulation model of the flow of components into the manufacturing plant based on the required material plan and finished goods distribution flow. For instance, calculate the time to receive each component, including shipping time (regardless of the source). This gives you a clearer understanding of the speed at which products are acquired to meet buyer demand.

• These models are known for their complexity and the difficulty in building and maintaining them.

Use the moving average method. This is a mathematical technique applied when there is little to no trend in the data. It provides a general overview of the data over time. To determine the demand of the last three months, add them together. Once you have the total, divide it by four (to calculate the next month). The formula is F4 = (D1 + D2 + D3) ÷ 4. In this equation, ‘F’ represents forecast, and ‘D’ denotes the month. This formula works well for stable demand.

• For example, forecast = (4,000 (January) + 6,000 (February) + 8,000 (March)) / 4 = 4,500.

Use weighted moving average (WMA). If the demand is unstable, this formula can be used to account for variance in sales. The WMA formula is WMA 4 = (W * D1) + (W * D2) + (W * D3). ‘D’ represents demand, and the corresponding numbers represent the months. ‘W’ is the weight constant, usually between 1 and 10, determined by historical sales data.

• Example: WMA = (4 * 100) + (4 * 250) + (4 * 300) = 2,600.
• Use a larger weight constant for newer data and a smaller one for older data. This is because newer data has a stronger influence on the forecast result.

Use exponential smoothing. This technique is a form of averaging that incorporates recent changes in demand by applying a smoothing constant to the latest data. It is useful if recent fluctuations are the result of real changes, such as seasonal trends (e.g., holiday periods), rather than random changes.

• Start with the forecast for prior periods. In the formula, this is represented as (Ft). Next, find the actual demand for that period, denoted as (At-1).
• Decide on the weight used. It is (W) in the formula, with a value between 1 and 10. Use smaller values for older data.
• Substitute the data into the formula: Ft = Ft-1 + W * (At-1 – Ft-1) or, for example: Ft = 500 + 4(W) * (590 - 500) = 504 * 90 = 45,360.

Summarize the results. Once you have gathered the data, create a graph or table that displays the demand forecast. This can be done by projecting the demand for the upcoming months. For example, when creating a line graph, place the months on the horizontal axis and the demand quantities on the vertical axis. If the forecast indicates a need for 600 units in October and 800 units in November, mark these points on the graph and connect them with a line. You can also plot historical data for comparison to see how the forecast aligns with past trends.

Analyze your results. Now that the data is organized and presented in an easy-to-read format, what does it reveal? Look for trends such as an increase or decrease in demand, as well as cyclical patterns like busy months or seasons. Compare your data with previous years to assess differences in volume and trends. Look for evidence in the data that suggests your marketing efforts have been, or were, effective in the past.

• At the same time, reflect on your initial assumptions. Were you overly optimistic with your forecast? How large is the margin of error in your predictions?

Present and discuss your forecast. Share your forecast with the appropriate stakeholders within your company and discuss it together. Gather input from departments such as marketing, sales, finance, production, and other management teams. Once you have gathered all the input, adjust the forecast as necessary. When everyone is aligned on the forecast, it will be easier to develop a more effective business strategy.

Monitor and adjust the forecast. As you collect new data, update your forecast accordingly. Make sure to incorporate all available information. If you fail to continuously track and update your forecast, you risk making costly mistakes that could affect your financial stability.