Small business owners and entrepreneurs use demand forecasting to figure out what their prospective and current customers will want in the future. This information helps you plan and develop new products and services, as well as expand into new markets. Simply put, demand forecasting helps you catch and ride the wave just as it begins to build and crest. There are several different types of methods used in demand forecasting, including buyer’s intentions surveys and other forms of quantitative research. Demand forecasting also estimates how much of a particular product your customers will want to buy, allowing you to create reliable sales forecasts for your business.
One type of demand forecasting uses price data from real-world markets to create a virtual market. Then, the experts analyze the data and compare it against other key economic factors, such as employment, inflation and productivity rates.
Part of the process of creating and evaluating this virtual market is incorporating foreseeable developments in the economy and market. For example, experts can use current and historic data to chart trends. This gives the analyst a crystal ball, of sorts – one that can predict future trends such as employment policies, public financing plans and predicted economic growth.
Extrapolation uses mathematical principles to predict future behavior based on current and historic data. It’s a data-driven perspective on consumer behavior, using quantitative research to access data about how your customers have behaved in the past towards your products and brand.
Let’s say your company sells artisanal cheeses, and for the past 15 months, you’ve experienced a steady uptick in sales of goat cheese. You can reasonably extrapolate from that 15-month data sample that the trend will continue and your sales will continue to increase in month 16.
The drawback of extrapolation is that it’s limited to currently available data when in reality future unforeseen events impact markets all the time. Still, it’s a helpful and simple method of demand forecasting that most small businesses can use.
Your customers can’t buy the perfect product. It doesn’t exist. They have to make a trade-off somewhere. Either they’ll pay more than they planned to for a particular feature or higher quality, or they’ll give up on a specific feature for a lower price. Trade-offs in product features happen all the time and in many different scenarios. A conjoint analysis starts from that simple premise: The customer can’t buy a product that meets all their preferences. Instead, customers find and buy the products that possess the features and attributes they most want and need, meeting as many of their preferences as possible. Conjoint analysis, therefore, is a way to figure out what those most-preferred features are, and what the customer is willing to trade in exchange.
For example, a car manufacturer might find customers value lower prices and better fuel economy over larger interior space and more color choices. A conjoint analysis will use customer input to discover exactly which feature combinations shoppers really value and prefer by having them rank major features in order of preference_._ Then the analyst will use statistical models to evaluate those responses. The final product is a written report on the conjoint analysis that can help your company refine and improve sales, marketing and production plans to better meet your customers’ needs and preferences.
Buyer’s Intentions Survey
A small business can also survey its potential customers about their intentions in order to forecast future demand. Intentions surveys ask respondents about what they intend to buy and when they intend to buy in the future.
You’ve probably seen these surveys on the web. On a media outlet site, for example, you might be prompted to fill out a short survey in order to access the content. That survey may then pose two or three questions about your intent to buy a particular product in the next six months, for instance, a new car or a hot tub.
The survey answers then give the analyst a specific probability that the person answering the questions will act in a certain way. For example, if the question asks how likely you are to buy a new car in the next six months and gives a range of answers from zero (not at all likely) to 10 (a certainty), a response of eight might translate to an 80-percent probability. The aggregate probability might then suggest a path forward on a new product that your business is contemplating.
There’s another survey-based method of demand forecasting called the Delphi method or Delphi technique. However, instead of surveying customers, in this method the business surveys experts.
Another major difference from the buyer’s intentions survey is that Delphi surveys are anonymously conducted in a series of rounds, punctuated by an analyst summarizing the opinions expressed in the prior round, then using that analysis to create the next set of questions.
The experts who are surveyed get access to the statistical summary as well as the new questions. Each round asks the expert to either stick with his prior answer or gives him the opportunity to modify his assessment, based on the way other experts responded.
The aim of the Delphi method, therefore, is to help a group of experts in your field reach a consensus. When the group of experts reaches that consensus about specific developments in your business’s market, you can then use that consensus to help guide future product development, sales and marketing campaigns.