Long and Short-Term Forecasting
Long-term Forecasts
Usually, these involve big decisions. What country will you operate in? What size and type of facilities will you have? What type and how much equipment will you need? What markets will you compete in? How much financing will you need? The answers to all of these questions are influenced by your long-term forecasts. Your long-term forecasts will provide you with the ability to make long term plans.
Make a list of the main items in your business that would benefit from long term forecasting.
Short and Medium-term Forecasts
Depending on your business, these are usually more adaptable forecasts that are easy to analyze and run on short cycles. They provide you with the means to answer questions like: How much labor will be required? How much inventory will you need? How many units should be produced? How much advertising for a particular campaign is feasible? How many supplies should be purchased? The goal is to make the best forecasts possible, knowing that it is not possible to have them be 100% accurate, and then to monitor your results and use what you learn to make better and better forecasts.
While specific departments are usually responsible for compiling forecasts, all departments within a business participate in the process from beginning to end. In the beginning stages, different departments, such as sales and operations, provide the forecaster with information. In the final stages, the forecasts are translated into action plans and communicated to the various departments. For example, a sales forecast may be translated as a target for the sales force, but it must also be converted into production levels for the production department, forecasted collection efforts for the accounting department and forecasted demand for the purchasing department.
Forecasting Caveats
The past informs the future, but variables must be considered. Accurate forecasts include all the variables. Past sales of a dying media format, for instance, may have no bearing on future sales. It's a manager’s job to use his or her own judgment when applying data. Contingency plans need to be put in place in case the forecast becomes invalid. An earthquake, drought, or recession, for instance, can completely change the buying habits of a population and invalidate even the best forecast.
Forecasts are never 100% accurate. Your forecast is a tool that gives you some idea about the future, but unless you have a crystal ball or a time machine, you need to think of your forecasts as educated guesses. You can refine them over time, but you should always remember that you are talking about the future.
A forecast for multiples is usually more accurate than a forecast for a single item. This stems from the fact that forecasting errors among items in a group tend to cancel each other out. For example, a sales forecast for a particular model of car will be more accurate than forecasts for each color.
Forecasts are more accurate over shorter time frames. The longer the duration of the forecast, the more variables there are, and the more room there is for error.
A good forecast gives your business usable and accurate information about the future. When you're setting up your forecasts, consider the following;
Is your forecast going to save you money? Most forecasts cost something to implement and monitor. Ensure your forecast is going add value to your business. If the items are trivial, and your forecast are costly to implement, save your money instead.
Does your forecast give you usable information? Does it tell your sales people how many units they're expected to sell? Does it tell your purchaser how much of a particular supply he or she needs to purchase? Is it grounded in the real world and can everyone involved understand it clearly and easily?
Does it give you information in time to use it? A day late and a dollar short won’t do you any good. Your forecast has got to be designed to give you timely information so that you can make plans and changes as the need arises.
Does everyone who will be using the information understand its margin of error? We know that forecasts deal with the future and have inherent uncertainties. Make sure that everyone involved in using the information supplied by the forecast knows the appropriate weight to attribute to it.
Does your forecasting method provide consistent results? Forecasts often run on cycles. A forecast is made, acted on, results are monitored, the forecast is adjusted and acted upon again, etc. If your forecast is fairly accurate for one term, and not for the next, its value as a predictor for the future is radically diminished. In essence, what this means is your forecasts are really only useful to you if they consistently give you an understanding of the future.
Forecast Accuracy
The point behind any forecast is obtaining the ability to plan. Forecasts give answers to questions like: How many kitchens will you sell next month? How long will it take to obtain all the necessary permits for building a franchise? How much inventory do you need to order? Answers are one thing, and accurate, reliable answers are another. Forecast accuracy is measured by looking backward to see how close your forecasts have been to your actual numbers over time. Depending on your needs, there are three main ways to assess the accuracy of your forecasts.