When to Use Max Diff vs. Conjoint Analysis in Market Research

When to Use Max Diff vs. Conjoint Analysis in Market Research

During the process of collecting data for product development research, there are several different methodologies and techniques you can employ. Among the most popular is Conjoint Analysis. Another is Max Diff Analysis (otherwise known as Best Worst Analysis).

Proactive market research can save your company heartache, time and money during the development and launch of a new product by helping you understand what features customers most value or prefer.

During the process of collecting data for product development research, there are several different methodologies and techniques you can employ. Among the most popular is Conjoint Analysis. Another is Max Diff Analysis (otherwise known as Best Worst Analysis).

Both are especially useful for researching consumer preferences while developing a new product or making enhancements to an existing one—before you make a substantial financial investment into a final version.

What is Conjoint Analysis in Market Research?

As a type of quantitative research, conjoint analysis simulates a real-world buying scenario and generates realistic data so that you can:

  • Modify product features based on preference
  • Assess sensitivity to price
  • Forecast how likely customers are to purchase a potential product 

The term conjoint is defined as “pertaining to or formed by two or more in combination,” and that provides context for this statistical technique and how it’s applied. To begin with, you break a product down into its individual components, also called attributes or levels. Then, when surveying stakeholders for product purchase decisions, you present various combinations of the components to test different potential product concepts and gauge consumer preferences. 

During the survey, respondents are asked to select which combinations they most prefer. For example, if you’re developing a new window, some attributes you might want to test include different sizes, styles, shapes, materials, opening/closing mechanisms and colors.

The goal is to measure how much each feature or attribute influences the responses you receive from potential and existing customers. This numerical value is referred to as a preference score, and it gives you a simple way to assess the relative importance of all the individual elements that comprise a particular product. Based on the answers you receive, some features will jump to the front of the line, demonstrating themselves to be high priorities, while others will reveal themselves to be less influential to customers’ purchase decisions. After all, what we may consider to be most important may not be most important to all buyers.

Not only does this data offer valuable insight into potential market shares for a particular product concept, but it also enables your company to determine how to balance pricing with particular features in order to appeal to the right customer base.

As an example, for your cordless circular saw or rubber slate roofing material, you can test pricing along with features. Certain functions or installation considerations may be appealing to customers, but only at a certain price point, while others are a high priority, regardless of cost. With conjoint analysis, you can measure what customers are willing to pay for certain elements and features and then determine if it’s worth your while to add them to your product concept.

What is a Max Diff Analysis for Product Research?

A Max Diff analysis follows a similar methodology to conjoint analysis, but it is less nuanced and decreases the number of decisions your respondents have to make. A helpful way to think of it is as “conjoint-lite.” Unlike Conjoint, Max Diff is more effective in getting discrete preference among a large set of attributes or features.  And unlike Conjoint, Max Diff is not an ideal tool to determine pricing elasticity.

Rather than leading survey-takers through each iteration of tradeoffs for a product, you ask them to choose which attributes are most important and least important among various sets of attributes.

For example, if you’re identifying how to improve your lawn mower, you might ask respondents to pick which of the following attributes would make them most likely to select a particular lawn mower, and which would make them least likely, and then list items such as:

  • Price
  • Brand
  • Easy-start features
  • Quality of Cut
  • Cutting Deck Size
  • Engine Size / Motor Power
  • Bag Inclusion
  • Availability for purchase online
  • Availability for same day in store purchase

Try to limit the number of features to three to eight for your respondents to choose from to generate the best data. The total number of attributes you want to test, along with your total sample size taking the survey, will determine how many iterations a survey respondent will see.  It’s best to increase your sample rather than the number of iterations to prevent respondent fatigue..

How Do You Read Max Diff Results?

At the end of a Max Diff survey, the results will show how many times each item was selected as a “most preferred” feature and how many times it was selected as a “least preferred” feature. From there, you can rank each element as to its level of importance to customers, also known as the preference magnitude.

A simple formula for calculating the preference magnitude of a given feature is to subtract the number of times it was selected as “least important” from the number of times it was selected as “most important,” and then divide the sum by the total number of responses. The more data, the better.

Since the maximum value for preference magnitudes is 1, the closer to 1 that your quotient is, the stronger your customers prefer the corresponding feature. On the flip side, the closer the quotient is to -1, the less they prefer a feature.

One benefit of Max Diff analysis is that it forces your customers to make a tradeoff, thereby resulting in significantly more useful information to guide your product development process. You should use this technique when you want to definitively measure the relative importance of each feature compared to others and use that data to determine where to invest your resources.

Testing a Product with Conjoint or Max Diff Analysis

As a company leader, you may have a good sense of what attributes and features most appeal to the customers in your target market. However, intuition isn’t a strong basis for investing in product development or enhancement. You need hard data to quantify which items to prioritize, and you can collect that through Conjoint or Max Diff Analysis. This is one of several research methods we use at The Farnworth Group to assist you in creating a successful product concept and determining what modifications can further increase adoption rates.