Is Maximum Difference Scaling (MaxDiff) Right For You?

Is Maximum Difference Scaling (MaxDiff) Right For You?

Is Maximum Difference Scaling (MaxDiff) right for you?

Have you ever needed to know what your customers prefer? Whether you have a list of ice cream flavors, a set of logos, car features, or advertising claims…how can you measure which options will have the most impact on customers?

Maximum Difference Scaling (MaxDiff) is a survey research technique for obtaining preference scores for a list of items. It allows you to prioritize a set of product features or test concepts. Data from MaxDiff indicates which items are most desired and by how much.

In a MaxDiff exercise, survey participants are shown a subset of items from the test list and asked to indicate which is their most and least preferred item. (The exact wording can be modified as needed e.g. most / least likely to buy; most / least important). Participants typically see several such questions, each time with a new subset of items from the test list. This exercise is carefully designed to enable the researcher to analyze and derive relative preference scores for each item.

Is Maximum Difference Scaling (MaxDiff) right for you?

Figure 1. In a MaxDiff exercise, survey participants are asked to select their most and least favorite type of candy.

Is Maximum Difference Scaling (MaxDiff) right for you?

Figure 2. Analysis of MaxDiff data results in preference scores for each item. Preference scores are on a 0-100 scale; the higher the score, the better. For example, Skittles with a score of 25, is roughly twice as preferred over Milk Duds.

Benefits of MaxDiff

How does MaxDiff compare to other alternatives for understanding customer preferences? Rating scale and ranking questions are the traditional methods researchers use to prioritize a test list. While each technique has its pros and cons, MaxDiff overcomes the short-comings of rating scale and rank order tasks.

Is Maximum Difference Scaling (MaxDiff) right for you?

Figure 3. A rating scale (left) and ranking (right) question

With traditional rating scales, survey participants usually show a strong bias for the positive end of the scale. Since they aren’t forced to prioritize one candy choice over another, many items are often rated as ‘liked.’ On the other hand, MaxDiff forces respondents to choose the best/worst from the test set, allowing you to get a better understanding of how customers discriminate one item over another.

In a ranking exercise, only the order of preference is indicated, not how strongly an item is preferred over another. Ranking exercises also become difficult for the participant to complete (e.g. how do you distinguish between the 11 and 12th most preferred ice cream flavor) as the list of items becomes longer. With a MaxDiff study, up to 20 or 30 items can be investigated with relative ease since only a small subset of items is evaluated at a time.

MaxDiff is an invaluable technique for measuring preference for a list of items or test concepts. With it, you can gain an understanding of how customers think and feel about specific product features or claims you are considering. That way, you can go to market with the optimal and most impactful offer for your customers.

Share this article:

Want to talk? Have a question? We’d love to hear from you. E-mail bloggers@focusvision.com

See for yourself: Begin your journey to better customer insights

Request a demo

Sign up to receive news from FocusVision