Website accessibility
Show or hide the menu bar
Main home
Section home
Log in

Menus, configurators, mock-ups and Simalto

The benefit of any form of research into choices and trade-offs is that it forces people to make decisions and so reveal what they most want or most prefer. Although conjoint analysis is by far the best known of what can be characterised as 'trade-off analysis' it is not the only option.

There are many cases where conjoint analysis is not applicable and other trade off techniques such as MaxDiff, menu-building or configurators, brand-price trade-offs (BPTO) or techniques like SIMALTO (simultaneous alternative level trade-off) or trade-off grids are more applicable, or observing real choices can be combined with questions and reasoning behind the choices being made.

Alternatives and adjuncts to conjoint analysis

Conjoint is a well proven trade-off technique, but it still has limitations for certain applications. The use of multiple choice tasks can seem repetitive and slightly nonsensical for respondents and alternatives like Adaptive Choice Based Conjoint can be more time-consuming. Conjoint is also limited in the number of attributes that it can be comfortably used for.

We have a number of options for making conjoint 'more intelligent' for respondents.

Menu-driven choices and configurators

The first is to use 'configurator' or menu-selector type questions (of the same type that you might use if you're ordering a Dell computer).

For these types of questions, respondents are asked to choose their preferred option for each of a set of attributes. This can be designed for use with conjoint-type analysis in mind using options such as menu-based conjoint analysis.

However, it can also be used as a way of structuring customer choices. For instance the first choice on a configurator gives a best-preference view. To make trade-offs, price and a budget constraint can be added to see how the respondent changes their decisions. This process can be repeated or combined with  traditional conjoint questions.


An older technique which is similar to menu-selector/configurator is Simalto. In this approach a large number of attributes (up to 60 in some cases) are laid out in a trade-off grid. The attribute is still broken down into levels. The respondent is then asked to complete a number of tasks on the grid. This can include selecting their best or first class performance, measuring current performance, prioritising improvements and adjusting preferences based on costs.

Layered decision making

In addition, conjoint itself can be broken down into more detail. The use of intermediate questions enables direct probing as to why certain items are chosen - for example marking the main drivers directly, rating a profile on likelihood to buy, asking for next best improvements. In a qualitative setting forcing a choice and then discussing the choice enables decision making processes to be researched. Alternatively choices can act as part of a stimulus-response exercise to gauge emotional content and reactions and so gather implicit learnings and drivers.

E-commerce mock ups

A more complex option is to build a full e-commerce mock up, but based on conjoint style considerations of attributes and levels. In a typical e-commerce system, customers are making choices from a much larger set of options than would be considered traditionally for conjoint analysis. These 'super-large choice' sets can be mocked up with e-commerce systems and combined with search and filter tools for use by respondents. We find that the relatively fixed text descriptions of attributes and levels makes the choices too unrealistic in this scenario, so we recommend using 'textual noise' versions of conjoint (as described in our IJMR paper).

By being creative, the conjoint task can be made much more engaging with respondents and gathering more data from the information

For help and advice on alternatives to conjoint analysis or realism in research contact

Previous article: Sensory-emotional research Next article: Usability and customer experience research
More details

Go to Notanant menuWebsite accessibility

Access level: public

This site uses cookies. By continuing to use this site you agree to our use of cookies: OK