Conjoint Rank: Overview of Conjoint Analysis Surveys on OpinionX

Conjoint Rank is OpinionX's free conjoint analysis survey format — a ranking method that is specifically for when you have multiple variables that need to be ranked together.

Conjoint surveys show 2-10 "profiles" at a time, with each profile showing a list of variable options, and asks participants to pick the profile that appeals to them most. Each profile shares the same categories (eg. size, color, price) but the options for each category vary by profile (eg. price could be $10, $25, or $50).

Whenever a participant picks their preferred profile, their vote is recorded and a new set of profiles will appear. The results of these votes are used to measure which categories and options are most important to people when choosing the profile they like most.

In this guide about OpinionX's Conjoint Rank question type, we cover:

Once you've added a Conjoint Rank block to your survey, you’ll see three different text input options: (i) question, (ii) category, and (iii) option. Here’s what each of these text fields look like on the final survey:

After setting your three text inputs, you'll have some additional ways that you can customize your Conjoint Rank block:

Here's more info on these optional next steps...

For charts like the Scenario Simulator or Marginal Willingness To Pay to work, OpinionX needs to know which category in your conjoint survey represents price.

Go to the Conjoint Rank block on your Survey Setup page, click the settings icon (⚙️) in the bottom-right corner, enable the "Price Category" checkbox, and pick which category you want to format for prices. Alternatively, on the Simulator or Marginal Willingness To Pay charts over on the Results Overview page, hit the "Configure" button and you'll be guided through a quick setup for price formatting.

You can add images as options during setup, which will appear in your survey like this:

After adding your question, categories and options, you’ll need to decide how these profiles will appear during voting. There are three things you can configure here:

A. Profiles per Set → How many profiles should show per voting set? The default and minimum is 2 profiles and the recommended maximum is 6 (the more complex the profiles, the fewer you should show per set).
​
​​B. Sets per Participant → How many times should each participant vote? Use the "Conjoint Calculator" to see our recommended number of sets per participant based on your survey design and estimated number of participants. Anyone can edit this variable, regardless of whether you’re on the free version or a paid plan.

​C. Skip Button → I personally like using forced voting for my conjoint surveys by removing the ‘Skip’ button, however you can leave the skip button there if you want to avoid forcing participants to select profiles that they don’t actually like.

The Conjoint Calculator on OpinionX uses the formula N=(O*E)/(P*S) to help researchers design their conjoint analysis surveys. To access this feature, click the "Calculator" button at the bottom of the Conjoint Rank setup card.

The calculator takes into account conjoint design variables, like how many profiles are shown per voting set and how many sets each participant will be asked to vote on, to suggest a number for how many participants you should recruit for your survey.

How the formula works:

Explained in plain English, this formula tells you how many participants to engage based on how many options are in your largest category and how many times those options should appear during voting, compared to the number of profiles in each voting set and the number of sets each participant is asked to vote on.

"Exposures" is a robustness variable, meaning the values are the same across all surveys: E = 160 for minimum, 240 for safe, and 400 for a robust sample size.

This feature allows you to decide which which options are allowed to appear together during voting so that you can prevent unrealistic profiles from being shown to participants. View our dedicated guide for this feature here.

Here's how you can share your survey and get your first participants:

A. Shareable Link → Grab the generic survey link by clicking the big “Share” button in the top-right corner of your survey dashboard. You can use this link however you’d like — share it on Twitter, Reddit, or LinkedIn, put it in an email newsletter going out to all your users, or link it in a popup or banner inside your product.
​
​B. Embedded iFrame → Add your OpinionX survey as an embedded widget on any web page by using our pre-built iframe. This is how I embedded the interactive conjoint analysis survey that you saw earlier in this blog post with the lilac-colored frame!
​
​C. Email Invites → Import a list of email addresses and send each person a unique link that can only be used once. If you’re conducting a conjoint analysis study of a sensitive nature, this is the best way to make sure that people cannot complete your survey multiple times. OpinionX charges $0.10 per email sent.
​
​D. Buy Participants → If you don’t have access to a pool of potential participants, you can always use a third-party service to recruit people to complete your survey. On these recruitment services, you pay a fixed fee per participant as an incentive for them to complete your survey. I personally recommend using Prolific, as it has a high-quality audience and is very easy to set up alongside your OpinionX survey.
​

Voting on a Conjoint Rank question on OpinionX has specifically been designed to work on any screen size, whether a large desktop, a touchscreen tablet, a small smartphone screen, or a little embedded widget on your website.

OpinionX shows each participant a set of 2-10 profiles and asks them to pick the one they prefer most. A new set of profiles is shown each time they vote until they have reached the number you set as "Sets per Participant" during survey setup.

The same categories are shown on every profile, but the options under each category vary from profile to profile. These cards are responsive based on your screen, so on smaller screens they will stack the text to make it as easy as possible to see all the profiles at once.

Conjoint results are automatically calculated once participants complete your survey. It can take up to 10 minutes for a new survey submission to be included in your results dashboard.

OpinionX uses a scoring model called Hierarchical Bayes Multinomial Logit (or HB-MNL for short), which is considered the gold standard in conjoint analysis research. HB-MNL borrows strength across respondents, so individual preference scores are estimated robustly, even in small sample sizes.

Once your conjoint results have been calculated, you can access a number of special conjoint analysis charts by clicking the Switch Charts button:

OpinionX's special charts for conjoint analysis:

Below you'll find a quick overview of each of these special chart types. Please use the links provided to access detailed in-depth guides for each chart.

The Preference Chart is the default format for Conjoint Rank results. It includes two charts: (i) Category Importance, which shows much of a respondent's decision is based on each category as a percentage that adds up to 100%, and (ii) the relative importance of each options, color-coded by category.

Survey Scores shows the same results as the Preference Chart but formatted as a plain data table, making it easier to see what score each option has and the effects of a segmentation filter.

The Scenario Simulator report lets you manually put together a set of profiles and then assumes that 1000 customers will buy one of those profiles, thereby forecasting the percentage of customers each profile would get and the revenue that each profile would therefore create.

The Ranked Concepts report creates a comprehensive ranking of every possible combination of options from your conjoint survey, helping you identify the profiles with the highest relative preference to survey participants.

The Marginal Willingness To Pay chart tells you how much a customer would be willing to pay to change one aspect of a product or offering. In the example below, we're comparing a baseline offering of a 256GB Google Pixel smartphone to see how much someone would pay for storage or brand changes — such as $150 to upgrade from 256GB to 512GB.

When your conjoint survey includes a 'none of the above' button, allowing participants to say they would not like any of the profiles shown during a voting set, then you can use the Rejection Rate chart to see which options push respondents the most to reject all profiles.

You can find examples of the Raw Data and Aggregated Exports of a Conjoint Rank survey from OpinionX here on Google Sheets (as of July 2024).

^ Aggregated Export of a survey with a Conjoint Rank block. This shows the scores for all categories and options broken down by each participant (one row = one participant).

^ Raw Data export of OpinionX's Conjoint Rank block. Each row represents one vote cast by a participant, with the full list of options that they viewed on display. For Conjoint Rank surveys with 3+ profiles, there would be more "Not Selected Profile" columns on display.

Conjoint Analysis surveys are free to create on OpinionX, just like all of our survey question formats. There are some differences to the functionality available for conjoint surveys depending on which pricing plan you're on — view our pricing page for more detailed and up-to-date information.