Discover the true drivers behind the choices your customers make, using Gradient Metrics' state-of-the-art statistical conjoint models.
We are constantly making choices, some conscious, but most unconscious. Standard methods with traditional question scales are unable to reveal what component of a product or concept drives its preference.
Most of these scales don’t simulate real-life conditions, which always include some type of situational trade-offs. To simulate realistic consumer decision making, the most precise approach is through either a conjoint analysis or MaxDiff analysis. Gradient has designed countless experiments within a wide range of product types and sectors.
Conjoint analysis is a technique for quantifying how the attributes of products and services affect preference. It is typically used to help identify the optimal design of products, messaging, and pricing.
A conjoint compares combinations of attributes. Each respondent is shown multiple lists with varying combinations of the attributes.
Across multiple lists of choices from hundreds of respondents, a statistical model identifies precisely how much each attribute and level contributes to making the decision.
The attribute importance is the proportion that a larger category or attribute (e.g. genre), contributes to driving preference.
The level importance is the proportion that a specific level within an attribute (e.g. within the genre attribute, drama and action) contributes to preference.
Conjoint analysis is a technique for quantifying how the attributes of products and services affect preference. It is typically used to help decision makers identify the optimal design of products and pricing.
An ideal conjoint module will have roughly 5 attributes (rows), 4 levels per attribute (columns), and roughly 5 - 10 choice sets. A detailed breakdown is below:
Attributes - Roughly 5 attributes with no more than 10 total levels per attribute.
Profiles - Roughly 4 profiles to show each set. Too many profiles per set, and you risk respondents not making effective choices.
Total Tasks - No more than 15 to avoid respondent fatigue.A showcase of the potential and versatility of discrete choice experiments.
Product Guide
Take a deep dive into conjoint experiments by going through our guide. Understand how to design a conjoint and get an appetizer of what the output looks like.
Download the product guide →Client Case Study
Gradient Metrics pushed a conjoint experiment to its limits, and designed an innovative approach to measuring success.
After an initial prototyping phase, our new methodology was rolled out on a national scale and has transformed how our client thinks about measuring Americans’ priorities across many domains.
Download the case study →Client Case Study
A creative education platform approached Gradient to help them assess their class offerings and customer willingness-to-pay for those offerings. We used a conjoint analysis to determine the class offering features
most attractive to customers, what impact the offerings have on market share and revenue, who the platform should target, and more.
The world's most forward-looking organizations trust Gradient