Skip to main content

AI Insights overview

Genuics automatically analyzes your data and surfaces insights - friction points, drivers, trends, and themes.

AI Insights are automated analyses that Genuics generates from your data. Unlike Geni (which you ask questions), Insights run automatically and surface what's important without you having to look for it.

How insights are generated

When you upload data or receive new rows through a connector, Genuics runs a multi-pass analysis engine:

  1. Metric scoring - calculates your key metrics (NPS, CSAT, etc.) and compares them to previous periods
  2. Variance detection - identifies unexpected movements: did something jump or drop significantly?
  3. Dimension analysis - breaks down metrics by every dimension to find which segments are over- or under-performing
  4. Driver detection - figures out which factors are driving changes in your scores
  5. Theme extraction - clusters free-text responses into themes and measures sentiment
  6. Comment mining - extracts specific phrases and language patterns from open-text fields
  7. Scorecard building - assembles summary cards with trends and breakdowns

Where to find insights

Insights appear in your dashboard alongside your widgets. They're also accessible through Geni - ask "What are the key insights?" and Geni will pull them up.

Each insight is classified by type:

  • Friction - something that needs attention (low scores, negative trends)
  • Positive - strengths and good news (high scores, improving trends)
  • Variance - unexpected changes worth investigating
  • Driver - what's causing a metric to change
  • Theme - patterns in free-text responses

Insight priority

Every insight gets a priority: low, medium, high, or urgent. Priority is calculated based on the magnitude of the finding and its potential impact. Urgent insights typically indicate significant negative changes that need immediate attention.

Acting on insights

When you see an insight worth following up on, you can:

  • Create an action directly from the insight card
  • Link the insight as evidence to an existing action
  • Dismiss the insight if it's not relevant

Incremental processing

Genuics doesn't reprocess your entire dataset every time. It tracks a high-water mark and only analyzes new rows, so insights stay fresh without redundant processing.

Next steps

Was this helpful?