Statistical testing is now built into Upsiide cross-tabs, making it easier to identify meaningful differences between subgroups in your data — without having to run external calculations.
With stat testing enabled, statistically significant results are automatically flagged so you can quickly spot differences that are more likely to be real, and not just random chance.
What is Statistical Testing?
Statistical testing helps you understand whether differences in scores are meaningful or just due to random variation in your sample.
If Product 1 has an idea score of 74 and Product 2 scores 80, it might seem like Product 2 is better- but without statistical testing, we can’t be sure that difference isn’t just due to chance.
With stat testing, Upsiide highlights results that meet your chosen confidence level (80%, 90%, or 95%), giving you more certainty in your conclusions.
How to Apply Statistical Testing To Your Storyteller Insights
Create an Insight
In Storyteller, create a new insight and open the table editor.
Add variables to the Rows and Columns sections.
Set the Display Value to %
In the table settings, choose Total % as the display value.
Note: Stat testing will not appear if you choose Count.
Enable Statistical Testing
Toggle on Statistical Testing in the table settings.
Choose your confidence level:
95% = highest certainty
90% = good certainty
80% = lowest certainty, narrowest range
Select your benchmark item
Benchmarks will automatically be moved to the front of the columns, and will be marked with an A
If no benchmark item is selected, stat testing will be applied to all items.
Understand the Rules
Compares subgroups within a single variable (e.g., Age: 18–34 vs. 35–54).
Minimum base size is 30 responses per group.
Stat Testing in Storyteller works with multi-select and grid questions.
Stat testing is not yet available for Idea Screens in Storyteller.
Total columns are not included in Stat Testing
Read the Results
Letter indicators (A, B, C, etc.) show which columns are significantly different.
Arrows (if enabled) show direction — ↑ higher significance, ↓ lower significance. If there is no significance, arrows will not be displayed.
Tooltips reveal the confidence level and comparison groups.
Add Clarity
Create a footnote in your Story to identify the stat testing applied to groups
Stat testing applied at 95% confidence internal, A/B, A/B" or “A/B/C/D/E” where each letter is the column involved and each slash "/" is stat testing, groups separated by a comma
Export Your Table
Save your table and export as CSV, PowerPoint, or image.
In the export settings, you can:
Include stat testing results
Choose a confidence level
Select a comparison type (e.g., pinned ideas vs. all ideas)
Tips for Best Results
Use higher confidence levels (95%) for maximum certainty.
Ensure your subgroups have enough responses (30+).
Use stat testing alongside your own judgment- statistical significance doesn’t always mean practical significance.