I mean, besides the good idea below, how many products will we sell? It's the classic question many managers do.

In most cases we just talk about it with colleagues, sales force and few reliable customers, assuming numbers on an excel file, in order to understand how many would be enough to go break even. If it "sounds" too high we forget it, otherwise we go on.

If company is clever does much more. During product development it performs concept tests, that is a survey on a sample of potential purchasers to understand which product or service variants are more performing and to evaluate the market reaction.

It's often required to estimate the concept relating to purchase probability for them and for the company they represent, but people don't do exactly what they say! The more the purchase intent is low, the less it's reliable.

How to solve the issue?

In my market surveys experience I've developed some tricks:

1.Use a purchase probability scale with many levels (9 levels) instead of the traditional ones with 2, of the kind "I would get it", "I don't think I'll get it". It spurs people interviewed to think deeply about it and it doesn't give rise to doubts.

2. Analyze not just the concept purchase probability, but further factors (less hard) as well, such as its first impression, its avail level perceived, its reliability conveyed, its price/quality, etc. (from 1 to 9 marks).

3. Show as a sample your whole idea and its competitors. The concept, its characteristics, its positioning, its price, and the competing products or services, showing their actual market prices as well.

4. Reckon your concept potential. Ponder the average on the scores acquired by each factor at the point 2 and the purchase probability at the point 1 and compare it with your competitors' one.

If you know your product or service costs in details (and usually few managers know them when they investigate on a concept test, but this is another topic!) then you're ready to understand a concept earning power compared to its sales potential.

According to my experience this approach enhances the probability to forecast what it will actually happen and although not perfect it's better than going by intuition!