Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Practical difference between Share of Preference and Purchase Likelihood

Relatively new to CBC and still learning. What is the practical difference between Share of Preference and Purchase Likelihood?

If I run a simulation with three products, with SoP of 30/25/45 respectively, I can say "given these three products, 45% of customers are likely to choose the third option," right?

Could I say that the purchase likelihood of that third product is also 45%? Is that an oversimplification?

In this particular example we did not ask a purchase likelihood yes/no question.

Thanks for the support
asked Dec 9, 2014 by WTW_cal Bronze (1,085 points)

1 Answer

0 votes
Hello, Colin,

I think there is a practical difference.  Purchase likelihood in many surveys is measured in a vacuum - this is common in new product concept testing for example.  A close conceptual analog in conjoint analysis might be when you simulate the share of preference between a product you specify and "none."  I say this is a close conceptual analog because the numbers you get out of an exercise like this can be quite different from the numbers you would get out of a traditional stand-alone purchase likelihood question.  

Does that make sense?
answered Dec 9, 2014 by Keith Chrzan Gold Sawtooth Software, Inc. (48,200 points)
Thanks, Keith. This makes sense. Backing up a bit, was my initial question accurate --

If I run a simulation with three products, with SoP of 30/25/45 respectively, I can say "given these three products, 45% of customers are likely to choose the third option," right?
...