Have an idea?

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

Pricing study: fixed + variable component levels

I'm doing a study where one of the attributes is pricing and the levels are various pricing models. Some of them are flat rates, while others are two-part tariffs. That means, people pay a (monthly) access fee and marginal fee for usage (e.g 10 dollar fixed + 0,80 cents per time the service is used).

I would like to measure the relative preference for various combinations of fixed and variable fees in the two-part tariff, for example comparing the preference between a high fixed rate in combination with a low variable rate versus a low fixed rate with a high variable rate. However, this preference of respondents will be dependent on their expected usage and not on their preference for the configuration of the tariff per se. Any suggestions how to account/solve for this?

Help would be appreciated, thanks a lot!
asked Oct 6, 2021 by gakker (140 points)

1 Answer

0 votes
You are assuming that people always behave "price rationally", while evidence suggests they often don't. Many are risk averse and are willing to pay for unlimited even though it doesn't make rational sense given their normal usage, while others prefer variable plans because they think it will limit consumption, etc. So I don't think there's damage in asking this as a conjoint and measuring people's eccentric, sometimes irrational looking preferences.
answered Oct 12, 2021 by Aaron Hill Gold Sawtooth Software, Inc. (12,245 points)
Thank you Aaron!

What I now was planning to do is to give the usage of an average household and ask the participant to imagine this usage to be his/hers. I have set up the levels that the total annual costs would be identical for each level given the average household consumption, so the effect of opting for the cheaper option should be cancelled out if the participant would engage in mental accounting. In this case, participants choose their prefered combination of fixed and variable costs. Do you see any issues with this methodology?