There is no universal set of forecasting techniques that can be used for all types of businesses. Forecasting can fall into a fairly comprehensive range of techniques with respect to level of sophistication. Some forecasting can be done on an intuitive basis (e.g., back-of-the-envelope calculations); others can be done with standard computer programs (e.g., Excel) or programs that are specifically dedicated to forecasting in a variety of environments.
A brief review of basic forecasting techniques shows that they can be divided into two broad classes: qualitative forecasting methods and quantitative forecasting methods. Actually, these terms can be somewhat misleading because qualitative forecasting methods do not imply that no numbers will be involved. The two techniques are separated by the following concept: qualitative forecasting methods assume that one either does not have historical data or that one cannot rely on past historical data.
A start-up business has no past sales that can be used to project future sales. Likewise, if there is a significant change in the environment, one may feel uncomfortable using past data to project into the future. A restaurant operates in a small town that contains a large automobile factory. After the factory closes, the restaurant owner should anticipate that past sales will no longer be a useful guideline for projecting what sales might be in the next year or two because the owner has lost a number of customers who worked at the factory. Quantitative forecasting, on the other hand, consists of techniques and methods that assume you can use past data to make projections into the future.
You will find examples below of both qualitative forecasting methods and quantitative forecasting methods for sales forecasting. Each method is described, and their strengths and weaknesses are given.
Qualitative Forecasting Methods
Qualitative forecasting techniques are subjective, based on the opinion and judgment of consumers and experts; they are appropriate when past data are not available. They are usually applied to intermediate- or long-range decisions.
- Simple Extrapolation
- Sales Force
- Expert Opinion
- Historical Analogy
- Market Research
Simple extrapolation is an that approach uses some data and simply makes a projection based on these data. The data might indicate that a particular section of town has many people walk through the section each day. Knowing that number, a store might make a simple estimate of what sales might be.
- Strength: An extremely simple technique that requires only the most basic analytical capabilities.
- Weakness: Its success depends on the “correctness” of the assumptions and the ability to carry them over to reality. You might have the correct number of people passing your store, but that does not mean that they will buy anything.
In firms with dedicated sales forces, you would ask them to estimate what future sales might be. These values would be pieced together with a forecast for next year.
- Strength: The sales force should have the pulse of your customers and a solid idea of their intentions to buy. Its greatest strength is in the B2B environment.
- Weakness: Difficult to use in some business-to-customer (B2C) environments. Sales force members are compensated when they meet their quotas, but this might be an incentive to “low-ball” their estimates.
Similar to sales force approach, the expert opinion technique ask experts within the company to produce estimates of future sales. These experts may come from marketing, R&D, or top-level management.
- Strength: Coalescing sales forecasts of experts should lead to better forecasts.
- Weakness: Teams can produce biased estimates and can be influenced by particular members of the team (i.e., the CEO).
In the Delphi technique, a panel of outside experts would be asked to estimate sales for a particular product or service. The results would be summarized in a report and given to the same panel of experts. They would then be asked to read their forecast. This might go through several iterations. Best used for entirely new product service categories. One has to be able to identify and recruit “experts” from outside the organization.
- Strength: Best used for entirely new product service categories.
- Weakness: One has to be able to identify and recruit “experts” from outside the organization.
With a historical analogy, one finds a similar product’s or service’s past sales (life cycle) and extrapolates to your product or service. A new start-up has developed an innovative home entertainment product, but nothing like it has been seen in the market. You might examine past sales of CD players to get a sense of what future sales of the new product might be like.
- Strength: One can acquire a sense of what factors might affect future sales. It is relatively easy and quick to develop.
- Weakness: One can select the wrong past industry to compare, and the future may not unfold in a similar manner.
Market research is making use of questionnaires and surveys to evaluate customer attitudes toward a product or a service.
- Strength: One gains very useful insights into the stated desires and interests of consumers. Can be highly accurate in the short term.
- Weakness: Experienced individuals should do these. They can take time to conduct and are relatively expensive.
Quantitative Forecasting Methods
Quantitative forecasting models are used to forecast future data as a function of past data. They are appropriate to use when past numerical data is available and when it is reasonable to assume that some of the patterns in the data are expected to continue into the future. These methods are usually applied to short- or intermediate-range decisions.
- Trend Analysis
- Moving Average
- Seasonality Analysis
- Exponential Smoothing
- Causal models—regression analysis
A trend analysis is a technique that assumes that sales will follow some form of pattern. For example, sales are projected to increase at 15 percent a year for the next five years.
- Strength: Extremely simple to calculate.
- Weakness: Sales seldom follow the same growth rate over any length of time.
Moving average is a technique takes recent class data for [latex]N[/latex] number of periods, adds them together, and divides by the number [latex]N[/latex] to produce a forecast.
- Strength: Easy to calculate.
- Weakness: The basic use of this type of model fails to consider the existence of trends or seasonality in the data.
Many products and services do not have uniform sales throughout the year. They exhibit seasonality. Seasonality analysis attempts to identify the proportion of annual sales sold for any given time. The sales of swimming pool supplies in the Northeast, for example, would be much higher in the spring and summer than in the fall and winter.
- Strength: Many products and services have seasonal demand patterns. By considering such patterns, forecasts can be improved.
- Weakness: Requires several years of past data and careful analysis. Useful for quarterly or monthly forecasts.
Exponential smoothing is an analytical technique that attempts to correct forecasts by some proportion of the past forecast error.
- Strength: Incorporates and weighs most recent data. Attempts to factor in recent fluctuations.
- Weakness: Several types of this model exist, and users must be familiar with their strengths and weaknesses. Requires extensive data, computer software, and a degree of expertise to use and interpret results.
Causal models—regression analysis
Causal models, of which there are many, attempt to identify why sales are increasing or decreasing. Regression is a specific statistical technique that relates the value of the dependent variable to one or more independent variables. The dependent variable sales might be affected by price and advertising expenditures, which are independent variables.
- Strength: Can be used to forecast and examine the possible validity of relationships, such as the impact on sales by advertising or price.
- Weakness: Requires extensive data, computer software, and a high degree of expertise to use and interpret results.
Forecasting key items such as sales is crucial in developing a good business plan. However, forecasting is a very challenging activity. The further out the forecast, the less likely it will be accurate. Everyone recognizes this fact. Therefore, it is useful to draw on a variety of forecasting techniques to develop your final forecast for the business plan. To do that, you should have a fairly solid understanding of the strengths and weaknesses of the various approaches. There are many books, websites, and articles that could assist you in understanding these techniques and when they should or should not be used. In addition, one should be open to gathering additional information to assist in building a forecast. Some possible sources of such information would be associations, trade publications, and business groups. Regardless of what technique is used or the data source employed in building a forecast for business plan, one should be prepared to justify why you are employing these forecasting models.
Other Forecasting Resources
- The Balance Website provides three simplified approaches to sales forecasting.
- Time-critical decision making for business administration This site has an e-book format with several chapters devoted to analytical forecasting techniques.
“5.3 Building a Plan” from Small Business Management by Anonymous is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.
“3. Forecasting” from Introduction to Operations Management by Mary Drane and Hamid Faramarzi is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.