Maximum Difference (MaxDiff) analysis, also known as “best-worst scaling”, is invented by Jordan Louviere in 1987. Research participants are repeatedly asked to indicate the best and worst item from a subset of items. After analysis, MaxDiff produces a rank between the items tested plus a metric distance between the items.
Since the ranks are established at individual level, they can vary greatly across respondents. We use cluster analysis to identify segments with similar responses, providing input into marketing strategies that recognize that not all consumers are the same.
MaxDiff: benefits & limitations
- Cultural differences as seen in rating questions are absent
- More items can be included than with regular ranking
- Greater discrimination among items and between respondents than with the use of scales
- No easy way of understanding the absolute strength of an item
- With many items, it can become boring for the respondent
MaxDiff: when to use it?
To screen among brands, image statements, product features, advertising claims, full concepts or benefit articulations. More specific:
- For consumer products: to develop a winning product claim
- In healthcare: for product messaging and communication
- In consumer health: to communicate health benefits effectively