Menu Based Choice (MBC)

The concept of Menu Based Choice

In many purchasing situations the decider only has a choice between finished products or complete offerings (e.g. shampoos, televisions or package tours). But in many other cases the decider has the possibility of putting together a package of individual components, for example …

• Choosing from a menu in a restaurant
• The configuration of a new car
• Internet and telephone connections, TV program offerings and mobile phone rate plans
• The combination of insurance plans with additional options

Since conjoint approaches are based on the principle that complete products are rated and selected, they are inappropriate for such situations. Therefore a procedure has been developed that allows for the possibility of being able to wholly or partly configure one’s own product.

Survey method

Here the respondents are presented with individual components from which they can put together their own preferred product, which may also be supplemented by complete or partial packages with reduced prices. In MBC, in contrast to conjoint procedures, prices and product features are directly linked. Each feature is allocated a certain number of price steps, which vary from task to task in a systematic way.

In principle the design of a MBC procedure consists of features that vary in price and which can be combined by the respondents in accordance with certain rules, whereas the design of a conjoint procedure consists of attributes with their various levels from which complete products are constructed and made available for selection. In conjoint, the price represents a separate attribute that varies independently of the remaining attributes.

For MBC, fewer tasks are necessary than in a conjoint procedure, resulting in a shorter interview time. On the other hand, larger sample sizes are required. Every price level of a feature should be shown at least 1,000 times across the complete sample. Thus, a survey with 8 MBC questions and 5 price steps per feature requires a sample size of 625.

MBC tasks must be designed individually in accordance with the objectives of the survey. This demands a high degree of flexibility and a large amount of programming work.

Results

The results are, to a degree, comparable with those of classic conjoint analyses. The optimization of product/service offerings, price elasticity analyses and estimation of potential can be carried out on the basis of individual utility values. However it is not a holistic model that is defined and estimated, but a separate one for each choice option. The relevant features with their price steps flow into the individual models.

Due to the high level of individuality of a MBC design it is necessary, to a much greater degree than for classic conjoint procedures, to anticipate all the sequential steps (data collection, analysis, implementation of the results) during its development.