All Collections
Analysis
Segmentation
Filter by Segment: Recalculate your survey results and focus on one specific group of people
Filter by Segment: Recalculate your survey results and focus on one specific group of people
Updated over a week ago

πŸ‘‰ This premium feature is available to customers on Analyze tier or higher.

Stack ranking will tell you what matters most to all of your participants, but sometimes what's even more interesting is how participant's priorities differ when you compare subgroups of them. That's where segmentation comes in...

If segmentation is new to you as a research concept, I recommend checking out our guide to Needs-Based Segmentation. If you're looking to learn how to use the segmentation feature in your OpinionX survey, continue reading πŸ‘‡


Why should I care about segmentation?

Instead of creating 10 different stack ranking surveys to research your users' priorities, you can pool them all together in one survey and use the segmentation feature to look at specific groups of participants in isolation.

For example: imagine you're running a survey where your customers are ranking all the frustrations they have with your product today. Using the segmentation filter, you could see how the results change when you look at only your free users or your best paying customers. That's 10x more valuable data than just the overall ranked results!

How do I use segmentation in my OpinionX survey?

  1. Use a Multiple Choice question in your survey to gather data on your participants, such as what pricing plan they're on (eg. Free, Basic, Team, Enterprise).

  2. Go to your results page and click the "Segment" button.

  3. Select which Multiple Choice question(s) and answer(s) you want to filter by.

  4. The full results page is recalculated to show only data from your chosen segment.

Where can I find the segmentation filter?

1. Floating "Segment" Button

2. Results Table "Segment" Button

3. Click Any Bar Chart

How do I set up a segment?

Once you open the "Segment" window, you can set as many segmentation criteria as you'd like. Here's an example:

^ Multi-variable segmentation follows OR logic within questions and AND logic between questions. The Boolean logic in the above screenshot is: (Basic OR Team OR Enterprise) AND (Sales/Marketing).

Once you hit "SEGMENT", the entire results page will recalculate to show only the data collected from participants that match your segmentation conditions. When segmentation is active, the colored tile beside ranked data shows how many ranking positions that opinion has gone up/down compared to the original results:

To remove a segmentation filter, click any "Segment" button and hit Reset, as below:

How do I compare segments on OpinionX?

Once you've activated a segmentation filter, the "Segment" button in the bottom-left corner of your screen turns into three Segmentation Options: (1) Clear, (2) Compare, (3) Configure. Click compare to open the "Compare Segments" window.

How do I make sure I can use segmentation?

To use the segmentation filter, all you need is at least one Multiple Choice question and a Scale tier or higher survey. Think ahead about what ways you'll want to segment your ranked results and make sure you gather that data from your participants by including a Multiple Choice block in your survey.

Will segmentation still work if I allow participants to select multiple answers in my Multiple Choice blocks?

Yes β€” segmentation works the exact same whether your Multiple Choice block is single-select or multi-select.

Can I prefill certain questions if I already have that data for my participants?

No, however we're planning to add this capability to OpinionX in Q4 2023.

Can I manually add tags to each participant on the results page with the data I already have on them?

No. While participant tagging is also planned future feature, it has not been scheduled for development yet.


For further information regarding needs-based segmentation and segmentation strategies, I recommend you check out our ultimate guide to needs-based segmentation.

Did this answer your question?