Insufficient information isn't a sufficient reason for not making an educated guess. You have no choice. You're in business; the only thing worse than guessing is not guessing at all. Many people think that they aren't qualified to forecast because they don't understand statistics or advanced mathematics. Relax, you can do it. More than mathematics and statistics, you need patience and a little bit of confidence mixed together with common sense.
With that in mind, let's look at some concerns about forecasting. These have come up in our Ask the Experts feature on Bplans.com.
How can I forecast a new product without past history?
There may not be history on your new widget, but there is a lot of history on introduction of new products in general. There is the standard product life cycle, the sociological research on idea adoption, and you could even use a diffusion model that compares the spread of new products to the spread of disease.
You can also look at descriptions of existing products, and make reasonable guesses. It turns out that there is a lot of history on new products, even if there isn't history on your particular new product. Most new businesses fall into the pattern of the product life cycle. Growth is slow at first as the product is accepted by the innovators, then accelerates as it penetrates the larger market, and then slows when it becomes an older product selling replacements only.
Product life cycle, idea adoption, and diffusion model theories are discussed in our online book, On Target: the Book on Marketing Plans.
How can I forecast when I can't get good demographic information?
This is a common problem. Lots of business forecasters don't have the luxury of starting with good data. Take a look at our section on Web links for getting market data. There may be some data that you just haven't found yet. There are some excellent sources for basic demographic data refered to there.
Even so, you can still be looking at markets that don't have basic demographic data. In this case look for a way to estimate what you don't know from what you do. Here are some examples:
- You can't find data on private schools for the Mexican state of Sinaloa. But you can find private school attendance for the entire country, and you can find student age population for Sinaloa, and student age population for the whole country. Then you can estimate that the private school pupils in Sinaloa will be the same percentage of the total population as they are for the country.
