The Sprig AI Product Insights Feed centralizes all of the most relevant and interesting insights from across all of your Sprig studies into a single, easy-to-navigate feed. This means that you can easily stay up-to-date on the latest user sentiment across your Sprig studies powered by AI, without having to manually review multiple reports and dashboards.
It’s super easy to get started. Just head over to your Sprig Home Page to see your insights.
There are five types of insights and artifacts that Sprig AI generates and surfaces on your home feed:
- Opportunities: Areas for improvement, making it easy to surface and solve your product’s pain points.
- Strengths: Positive product feedback to help your team tap into your strengths and build on your success.
- Trends: Changes in feedback metrics (mean NPS score, user sentiment, etc) over time.
- Correlations: Connections between responses within a survey.
- Summaries: Paragraph summary for completed surveys highlighting key takeaways
- Create and launch a Survey
- Collect at least 30 responses
- Check back the next day, if Sprig can generate an insight, it will show up on the insights feed and the corresponding study results page
- Keep checking back, Sprig continues to generate insights throughout the lifespan of a survey based on new responses
- Click into an insight to see contributing responses or the chain of thought reasoning behind an insight
Keep in mind that AI isn’t perfect and we’re actively working to improve the insights we generate for customers.
Have feedback? Submit it here
Sprig needs a certain amount of data in order to generate insights. Additionally certain studies might not be suitable for insights generation (e.g. single question surveys, complex skip logic, etc)
- Opportunities & Strengths require at least 3 open-text responses before being insight eligible
- Trends require 50 total responses spanning multiple months.
- Correlations are dependent on survey structure and require a minimum of 50 responses and have at least two of the following question types: Open-text, MC, Likert, NPS, or rating scale
- Insights are computed nightly for insight-eligible studies. Insight eligible studies are those that have at least 30 responses or have received >10 new responses in a day.
- Summaries are generated upon survey completion.
- Any qualifying study should AI generated insights, as long as it was collecting responses prior to February 2024
Why does the study card say based on many responses but the insight only has a handful of contributing responses?
- Based on # responses refers to the number of responses that a survey had at the time the insights were generated, not necessarily the number of contributing responses
- # of contributing responses refers to responses that contribute to an insight being generated. Insights are often generated with other contextual information that might not be reflected in the insights chain-of-thought-artifact
Updated 19 days ago