How to Do Segmentation (subgroup analysis)?

Learn how to do segmentation on the Conjoint.ly platform with this visual guide.

Conjoint.ly allows you to apply segmentation to any kind of report. The report can be split by different types of variables:

  • Respondent information recorded by Conjoint.ly (such as location and device type).
  • Additional questions: Questions that you ask under the “Additional questions” (e.g., multiple choice or open ends).
  • GET variables
  • Externally supplied variables

In order to set up segments in your report, you need to:

  1. Under “Segmentation” tab, click on Add segment:

    Adding a segment to a study on Conjoint.ly's platform
  2. Click on the default name to rename your segment appropriately:

    Naming a segment on Conjoint.ly's platform
  3. Click on Add a new condition and you will be prompted to select which type of information to segment the report by:

    Selecting segmentation type
  4. In this example we will use an “Additional question” asking the gender of respondents and “Select values that match” with “Female” as the value:
    Selecting what information determines segments in the survey report
  5. If your segment definition is more complicated and involves condition from multiple questions with AND/OR logic, you can select Add a new condition after creating the first logical condition. You can then specify if these conditions should be combined through either an AND or OR statement. For example, this is how you can create the Segment: Female Age 25 and above.
    Adding logic to segmentation
  6. You can also use nested logic when creating segments. When adding a new condition, you can select the option Add a group of conditions. This allows you to create more complex logical statements. For example, the below segment comprises all Female Age 25 and above, as well as Female who have purchased a car for herself, family or business.
    Adding nested logic to segmentation

You can add another segment simply by repeating the above steps. Once you are done, click on Save and apply these segments in green and the system will apply your segmentation. Please allow for a couple of minutes for the segmentation tab to appear.

Interpreting segmentation results

Each segment can be viewed through changing the options in the drop-down box in the top right-hand corner.

Viewing segmentation results in the survey report

For each of your segments you are able to investigate the different preferences for each segment, as well as responses to the additional questions.

How do you check the conditions for a segment?

You can check the conditions for your segment under the “Segmentation” tab. You can select the segment that you are interested in on the left and review the conditions set for that segment will be displayed on the right:

How to check conditions used in segmentation

How to do cluster analysis in Conjoint.ly?

Conjoint.ly segmentation allows you to do data cuts / subgroup analysis based on existing variables.

If you are looking for cluster analysis, you can export individual level data into Excel and run your cluster analysis externally (e.g. in R).

Then, you can upload the cluster membership back into Conjoint.ly and do the subgroup analysis on the new variable.


Written on 7 July 2020 by:
Harrigan Davenport image
Harrigan Davenport
Market Researcher

Read these articles next:

Methodology Behind Claims Test

Methodology Behind Claims Test

Research tools 24 September 2018

The methodology behind our Claims Test tool is based on a proven choice-based technique & was refined through multiple projects for FMCG brands. Learn more here! View article

Do Russians really trust Putin? And can you trust WCIOM?

Do Russians really trust Putin? And can you trust WCIOM?

Target audience 10 April 2020

We conducted our own online poll amongst 372 panellists to validate WCIOM’s claims about the levels of support for Russian politicians. View article

Kano Model of Feature Selection

Kano Model of Feature Selection (with Free Excel Template)

Research tools 22 October 2020

The Kano Model is used to analyse consumer preferences for different features and group them into multiple categories. This allows firms to identify which features they should focus on when developing a new product. View article