MaxDiff : Maximum Power to Act on Market Preferences

Determining preferences in product features or benefits, marketing slogans, logos, service options, or even restaurant names, allows for trial and error without the error.

CHICAGO – OCTOBER 31, 2009 – When launching a new product or service, or looking to reposition an existing one, professional marketers can draw on a wealth of advanced market research tools to enhance their chances of taking successful actions. A relative newcomer among these techniques is Maximum Difference Scaling, or simply MaxDiff, a technique that allows marketers to identify the “best” of many alternatives— preferences in product features or benefits, marketing slogans, logos, service options, or something else entirely like names for a restaurant.

The result is a clear determination of market preferences, allowing for the creation of priorities based on what the market values most. MaxDiff homes in precisely on real consumer preferences and puts a key focus into any marketing strategy.

“Typical techniques ask survey respondents to rate or rank items. Rating often results in ties as people select the same answer for each item. Especially with long lists, ranking is often too hard for respondents. With MaxDiff, we design studies that present survey respondents with a set of items and ask them to select which is most preferred and least preferred,” said Erika Bruhn, a partner in the Chicago-based market research consulting firm Sawtooth Technologies Consulting Group. “Through a series of these questions that are relatively easy for respondents, we can analyze product features or applications, benefit statements, product names, slogans or even packaging options, and determine the importance of each based on preference. This goes a long way in giving marketers a clear path to the right choice.“

Sawtooth Technologies provides its clients with a number of advanced market research techniques, by designing, conducting and analyzing such custom studies as MaxDiff, Conjoint Analysis, Perceptual Mapping, and Customer Satisfaction Modeling. Depending on the needs of the project, each type of technique can be employed alone to pinpoint certain information, or done in concert to round out an entire marketing strategy.

For example, to understand how a product or service should be configured or priced drawing from a broad list of variables, conjoint analysis is probably the best approach. To focus in on market preferences from a single list—for instance which benefits are most important, which message resonates best or which projects you should undertake first—MaxDiff is most often the technique applied.

Maximum Difference Scaling sounds like it might be complex, but for practical purposes, the technique is actually straightforward.  Also known as best-worst scaling, MaxDiff is a better way to differentiate among a list of options versus traditional techniques.  If there is a list of ten or more items and using such tools as ratings, rankings or constant sum questions is under consideration, MaxDiff instead offers distinct advantages:

1.   Better ordering and differentiation than ratings

• For example, customers often rate everything as important

2.   Better differentiation than rankings

• Rankings don’t show the difference between choices, e.g. how much better is first than second, second than third, and so on?

3.   A less complex task than constant sum, especially for larger lists of items

• After 5-10 items, sums become difficult for respondents
• Summing doesn’t guarantee a clear order or differentiation 

For example, from a list of 20 product features, a MaxDiff study could seek to determine which are most important to consumers.  The process would be to ask roughly 12 questions, each displaying a subset of about 5 of the features.  For each set of 5 features, the respondent would choose which is most important and which is least important, like this:

So, for a set of 20 cell phone features where the object is to understand which are the most important, the results might look like this, where the scores add to 100:

“In this example, MaxDiff might lead to eliminating some items from consideration, like going with a 3” screen instead of 1”, and offering a 3-megapixel camera instead of the 1-megapixel option,” noted Sawtooth Technologies’ Bruhn. “If you wanted to go further in pricing different phones, each with a smaller subset of features, a conjoint analysis study could reveal which combinations would have the greatest market impact.”

Some people might get confused between conjoint and MaxDiff, in that they both involve trade-offs to some extent; the respondent are effectively told that they can’t have everything and are forced to make choices.  However, in a MaxDiff study the respondent evaluates a single list of items, whereas in conjoint the respondent evaluates complete products made up of various features.

MaxDiff is a powerful technique used to determine which features in a product or service should be embraced or eliminated, and it is often employed early in a development process to home in on the most probable strategy for success.

“Think of it as trial and error without the error,” said Bruhn. “Advanced research techniques like MaxDiff ultimately save a lot of time and money in getting a project off the ground.” 

For more information on all of the market-changing research techniques from Sawtooth Technologies, visit www.sawtooth.com/consulting.