Conjoint analysis is an advanced survey-based statistical technique used in market research to determine how people value specific attributes (features, function, and benefits) in a product or service.
Why conjoint analysis?
Conjoint analysis allows you to determine the ideal combination of a
limited number of attributes. A good conjoint analysis solution will help you
answer
What if
questions about which attributes consumers consider the
most influential.
Alida's conjoint analysis solution uses established mathematical and statistical methodologies to deliver data you can rely on. At the same time, the intuitive interface allows everyone, from less experienced survey authors to veteran market researchers, to focus on their project without having to worry about the intricacies of experimental design and analysis.
As a market research technique, conjoint analysis can help you:
- Identify consumer preferences.
- Analyze the trade-offs consumers would make.
- Understand the perceived importance and preference for each feature within the product or service.
- Determine the ideal combination of attributes for a new product or service.
Use cases for conjoint analysis include:
- Enhancing product development and prioritizing features.
- Estimating price sensitivity and the influence of cost on purchasing decisions.
- Gauging which employee benefits to offer.
- Determining software packaging.
- Determining marketing focus.
Example
Marin + Rose is trying to find out more about what consumers want from a new moisturizer they're launching. Instead of asking separate questions about price, packaging, active ingredients, and skincare benefits, Marin + Rose can ask if consumers want a $25 SPF moisturizer in a pump bottle, or a $20 hyaluronic acid moisturizer in a jar. Marin + Rose can ask about other combinations as well to see what is most appealing.
Why Choice-Based Conjoint?
Choice-Based Conjoint presents participants with several profiles in a choice set. Each profile represents a bundle of product traits (or attributes) such as brand, color, size, price, and so on. Each attribute can have multiple values (or levels). Participants choose the most appealing profile from each set and repeat this task multiple times. Each time they choose a profile, they are making trade-offs about which levels they value over others.
Based on participants' selections, the application calculates the implicit valuation of all the levels. You can use these implicit valuations (or part-worth utilities) to estimate market share, revenue and even profitability of new designs. Part-worth utilities are also useful for determining the optimal bundle of traits that your offering should have.
Choice-Based Conjoint is the most popular conjoint methodology because it comes closest to how consumers approach purchasing decisions in the real world. Therefore, a Choice-Based Conjoint question feels more natural and presents less decision fatigue for participants.
For example, let's say participants are evaluating concepts for new cereals. It's much easier to ask participants to pick one cereal they want out of a set of 4-5 cereals, and to repeat that task over several rounds. This question structure resembles what participants would do if they were shopping in a supermarket: comparing one box of cereal to others in the same aisle. The mental load on the participant is minimized. At the same time, you get a wealth of comparative and inferential data, from which you can extrapolate what participants value most.
Other conjoint methodologies exist and have their advantages, but they also present a greater mental load for participants. What's more, they are either older methods, or methods that present complex experimental design challenges.
Terms and definitions
| Term | Definition |
|---|---|
| Conjoint analysis | An advanced survey-based statistical technique used in market research that helps determine how people value different attributes (feature, function, benefits) that make up an individual product or service. |
| Attribute | A characteristic of a product or service.
For example, brand, color, and price range are all attributes.
Attributes should be independent of each other. Attributes with overlapping meanings can lead to incorrect answers. |
| Level | A value for an attribute. For example, for the attribute color, the different levels could be red, blue, and yellow. |
| Exclusion set (or exclusion rule) | A combination of attribute levels that will
never be shown to participants. For example, if you know that a T-shirt will
never be made in turquoise, you can create an exclusion set that prevents
T-shirt and turquoise from being shown together in the same concept.
It is best to be extremely conservative about using exclusion rule as it impacts precision of analysis values and reduces the flexibility of market simulator - use them sparingly and avoid them where possible. |
| Profile | A combination of attribute levels displayed to participants as a concept for selection. |
| Choice set | Multiple profiles presented together, allowing participants to compare profiles and make a selection. A Choice-Based Conjoint question contains multiple choice sets to be presented to a participant. |
| Part-worth (or part-worth utility) |
A statistically deduced value indicating the impact an attribute level has on consumer preference. Part-worth is calculated based on the trade-offs participants perform when they choose one profile over others in the same choice set. |
| Zero-Centered Differences (ZCD) | ZCD is a scaled version of the part-worth that makes values easier to compare across studies, attributes, and respondents. These values are only used in modern reports. |