What Is MaxDiff and How To Use it in Your Customer Research

In this blog post, learn what a MaxDiff analysis is, when to use it, why it's advantageous, and how quantilope can help.  

Table of Contents:

• What is MaxDiff?

• When should I use MaxDiff?

• What are the advantages of MaxDiff and what will it show me?

• How can quantilope help me run a MaxDiff analysis?

What is MaxDiff?

MaxDiff short for maximum difference scaling and sometimes called Best-Worst Scaling (BWS), is a quantitative research technique that presents respondents with a list of items and asks them to identify which they consider the most and least important. The nature of the items will depend on the research study, but they can be anything from product features and service aspects to brand messages and tangible goods or brands.

By showing respondents multiple sets of items and forcing them to make trade-offs, MaxDiff reveals customer preferences and the relative importance of each item.

Back to Table of Contents

When should I use MaxDiff?

MaxDiff is a hugely flexible and efficient methodology for understanding consumer preferences, so it has widespread application across market research studies. In a nutshell, it suits any business question that involves finding out what consumers consider to be most and least important to them within a category - and therefore which aspects a business should prioritize to drive uptake of its offer. Imagine for argument's sake that you want to create a new chocolate bar and bring it to market; there are several points along the journey at which you could utilize MaxDiff to inform your decision-making:

• Product Design and Development

  What qualities should the chocolate bar have? This could include attributes such as sweet, smooth, with nuts, without nuts, with fruit, with antioxidants, easy to break, easy to share, with a filling, dark, milk, white.... the list goes on - but the key to your product development is knowing which ones are most important to your target audience.

• Naming

  Once you have a list of possible product names, ask respondents to choose their favorites and least favorites. 

• Advertising 

  Which messages are going to grab consumers' attention? The list of items you test here could be attributes of the chocolate or experiential aspects and emotions. 
  
  

Beyond new product development, other market research studies for which MaxDiff questions are commonly used are communication testing, brand preference, and customer satisfaction.

Back to Table of Contents

What are the advantages of MaxDiff and what will it show me?

MaxDiff makes for a smooth and relatively easy respondent experience compared with some other question types, resulting in high-quality data. In addition, easy-to-read data output means MaxDiff has several advantages for analysis and business decisions: 

• Respondents are shown a number of short lists rather than one long list, making the task much less onerous than with other research techniques. Standard rating questions, for example, burden respondents with rating each item individually on a long list, risking boredom and the temptation to give similar answers for each item; furthermore, rating scales can be treated differently depending on the respondent, making them tricky to interpret. Ranking questions can also be demanding, with it being hard to rank ten or more items meaningfully, and the degree of importance between rankings is unknown. Both of these methods are also vulnerable to list order bias, as the items towards the bottom risk getting less attention. 
• The sets of items are shown to respondents multiple times, in various permutations, so that each item can be rated against every other item - this way, each item gets optimal exposure, resulting in a thorough analysis of respondents' relative preference for one aspect over another. In the chocolate bar scenario, respondents might be shown a first list that contains 'sweet', 'dark', and 'healthy' from which they must choose which is most and which is least important to them. They would then be shown subsequent lists that contain a mixture of other attributes; list 2 might be 'dark', 'with nuts' and 'smooth'; list 3 'smooth', 'sweet' and 'easy to break', and so on. While dark might rank highly in list 1, it might drop off in importance when considered in the context of list 2.
• In terms of output, MaxDiff takes the preference scores for each item and presents them in bar chart form detailing two key metrics: the relative rank of items in order of importance, and the magnitude of those items' importance relative to each other.
• The beauty of MaxDiff is that it enables greater discrimination between items and respondents than traditional rating scales. 
  • MaxDiff data might reveal that the desire for a dark chocolate bar ranks as most important and that 'smooth' and 'easy to share' rank in second and third place. But if there is a huge distance between 'smooth' and 'easy to share', you might not want to give 'easy to share' too much weight when designing your new product.
  • Market segmentation is possible when the sample is large enough, showing how customer preferences group according to demographics or background attitudes of those sampled.
    Back to Table of Contents
    
    

How can quantilope help me run a MaxDiff analysis?

Creating a MaxDiff survey with quantilope is as easy as filling one in. Simply select 'MaxDiff' from the list of methods and drag and drop it into your survey. Then all you have to do is type in your MaxDiff questions and the list of items you'd like to include for respondents to choose from. As soon as responses start coming in, data begins automatically analyzing and you can view the output on easy-to-read & easy-to-use charts. These charts instantly show how items score and how this differs by respondent type or sample criteria.

quantilope's MaxDiff is fully automated and customizable. See it in action here:

https://www.quantilope.com/maxdiff-demo-video

To learn more about quantilope's MaxDiff analysis, get in touch below: