See also: Closed-Ended Surveys, Data Mining, Multivariate Testing
Difficulty: Hard
Requires existing audience or product
Evidence strength
Relevant metrics: Price elasticity, Attribute importance
Validates: Desirability
How: Determine how people value features, functions, and benefits of a product and statistically determine which combination influences decision-making more. Conjoint analysis is most often used in existing markets where product attributes are generally known by the customer as it's hard to determine values of unknowns.
Why: Conjoint analysis helps estimate psychological trade-offs consumers make and can reveal real or hidden drivers not apparent to consumers themselves. Use that knowledge to test specific mock-ups, prototypes, or products.
This experiment is part of the Validation Patterns printed card deck
A collection of 60 product experiments that will validate your idea in a matter of days, not months. They are regularly used by product builders at companies like Google, Facebook, Dropbox, and Amazon.
Get your deck!Before the experiment
The first thing to do when planning any kind of test or experiment, is to figure out what you want to test. To make critical assumptions explicit, fill out an experiment sheet as you prepare your test. We created a sample sheet for you to get started. Download the Experiment Sheet.
Preparing the experiment
To perform a conjoint analysis, first narrow down the top 3-5 product attributes you want to test. Choosing which attributes to test and in what priority should be qualified by generative research such as Customer Discovery Interviews as a preparation to the experiment.
The more attributes you choose to test at the same time, the more responses are needed. As a general rule of thumb, the appropriate sample size can be calculated by:
Sample size = (Total count of tested attribute values – Attribute count) · 10
By mixing all possible variations of the various attributes and its values, you will generate all product attribute variations. There are several software tools that will help you do this.
Example case
To illustrate conjoint analysis in action, consider the following example examining what kind of phone plan would sell the best (source):
| Attributes | Possible attribute values |
|---|---|
| Brand | Brand A Brand B Brand C Brand D |
| Price | $60/month $75/month $100/month |
| Minutes included | 800 1,000 1,400 2,000 |
| Rollover options | No rollover of unused minutes Unused minutes rollover for 1 month Unused minutes rollover for 1 year |
| Call options | No free calling based on contacts Free calling to top 5 contacts Free calling to top 10 contacts |
A limited number of available product variations are chosen to be presented to participants side by side. One combination of product variations for participants to choose from could look like this:
| Brand A | Brand B | Brand C |
|---|---|---|
| 1,400 minutes | 1,000 minutes | 800 minutes |
| Unused minutes rollover for 1 month | No rollover of unused minutes | Unused minutes rollover for 1 year |
| No free calling based on contacts | Free calling to top 5 contacts | Free calling to top 10 contacts |
| Costs $100/month | Costs $75/month | Costs $60/month |
To dertermine which attribute values are more powerful choice indicators, how often a product value was included in the product selected are counted and presented as the percentage of times the attribute value was included in the selected product.
For the Call options attribute, results could look like this:
| Call Options | Value |
|---|---|
| Free calling to top 10 contacts | 50% |
| Free calling to top 5 contacts | 20% |
| No free calling based on contacts | 0% |
For the Rollover options, results could look like this:
| Rollover Options | Value |
|---|---|
| Unused minutes rollover for 1 year | 100% |
| Unused minutes rollover for 1 year | 30% |
| No rollover of unused minutes | 0% |
In the example above, the Call Options had values ranging from 0 to 50 and the Rollover Options had values ranging from 0 to 100. This would indicate the Rollover Options was more important to participants than the Call Options.
Primarily used for existing markets
Conjoint analysis concerns itself about uncovering customer preferences (not actions). This is why using this methods primarily makes sense in existing markets, where product attributes are generally known by participants.
In the early stages where totally new attributes are introduced to participants, it can be hard for participants to comprehend and understand what they are choosing between. This is when results become inaccurate with the risk of producing false negatives.
This is why it makes sense to conduct significant explorative and generative research before starting a Conjoint Analysis experiment. This method is rarely used in early stage innovation products, but after product/market fit has been reached.
After the experiment
To make sure you move forward, it is a good idea to systematically record your the insights you learned and what actions or decisions follow. We created a sample Learning Sheet, that will help you capture insights in the process of turning your product ideas successful. Download the Learning Sheet.
Popular tools
The tools below will help you with the Conjoint Analysis play.
-
Conjoint.ly
Can help narrow down feature selection, marginal willingness to pay, price elasticity of demand, pricing your products
-
Conjoint Survey Design Tool
A free tool created in 2014, supplied by Harvard
-
SurveyGizmo
Survey tool that includes a conjoint analysis tool
-
1000 minds
Scientifically used Conjoint analysis tool.
Examples
IMS Health
IMS Health regularly uses conjoint analysis to evaluate the importance of a product’s attributes to consumers. For a pharmaceutical product such as a drug, attributes may include price, dosing, efficacy, and side effects, among others.
Source: Conjoint analysis to understand preferences of patients...
Related plays
- Conjoint analysis by
- Things You Need To Know About Conjoint Analysis by Pestle Analysis
- Conjoint Analysis 101 by Brett Jarvis
- Conjoint analysis in pharmaceutical marketing research by Gang Xu, Yilian Yuan
- Conjoin Analysis by Tristan Kromer