A segment is a combination of two or more conditions, such as demographics, answers to certain questions, waves or countries. Examples of segments could be: female millennials vs male millennials, Londoners who prefer online shopping vs. in-store shopping, your brand buyers 6 months ago vs. now, etc.
Here we’ll explain in detail how you can create segments in your results dashboard.
If you want to know more about what types of segments are interesting to analyse or when segments can be useful, we recommend this article.
Where do you create segments?
You will see “Segments” at the top of your filter panel (on the Overview & Trends page) or at the top of your variables list on the Analysis page. If you or someone in your organisation has already created segments created for the survey, you’ll see them here; otherwise you’ll see the option to “+ Create”.
How to build segments
Step 1: Name your segment
Start by choosing a name for your segment. You might already have predefined segment names in your company or you can just write a short description such as “Gen Z - Non TikTok users”. You can edit your segment name at any time.
Step 2: Adding conditions
Now it’s time to start adding your conditions. You can add the following types of conditions:
Demographics (e.g. gender, age, income)
Country
Answers (to single choice, multiple choice, grid & open text questions)
Waves (if you’ve sent your survey multiple times)
When adding a condition, you can choose between “is” and “is not”. When choosing “IS NOT” you are adding a condition that will show respondents that don’t fit a certain demographic criteria or haven’t selected a specific answer. This can be particularly helpful if you want to look at consumers that haven’t purchased your brand. For open text questions you can choose between "contains" and "does not contain".
You can add as many conditions as you like, but keep in mind that you don't want to create segments that have a low number of respondents (we advise a minimum of 50 respondents per segment).
Step 3: Combining multiple conditions
When you add more than one condition, they will automatically be connected with “AND” logic. When you select multiple options within a condition it uses “OR” logic (e.g. Answer to “Q1. Which social media app do you use is TikTok or Snapchat).
You can change the logic used between conditions from “AND” to “OR”, this will change the logic between all your conditions. If you want to use a combination of “AND” and “OR” you will have to group conditions together first.
How to group conditions
How to group conditions
When you have three or more conditions, you will see the option to “Create groups” in the top right corner.
Click “Create groups” and then select the conditions that you want to have grouped (these are conditions that you want to have validated first) and click “Group conditions”.
You will now see your conditions grouped together and you are able to change the logic within the group from “AND” to “OR”, independently from the logic outside of the group. There is a summary at the top, that will clearly show how your segment is built, in this case A and (B or C).
Click “Save groups” to return to your segment builder. You can edit your groups at any time.
Step 4: Create your segment
You can save your segment just for the survey you’re looking at, or choose to “Save to library”, which will make this segment available for all surveys in your organisation. This can be useful for segments that you often use for your reporting, such as a certain age ranges or income categories. Saving to the library is only supported for segments created with demographic conditions.
When you are happy with your segment, you can click on “Create segment” and your segment will be added to your list of segments.
Step 5: Managing segments
You can manage your segments at any time to edit, delete or duplicate one of your segments or create a new one.
Start analysing!
When you’ve created your segments, you can use them as filters on Overview & Trends and see how the respondents in this segments have answered to your questions. You can also compare them against each other on the Analysis page in either crosstab or charts and see if there are any statistically significant differences between them.
