Connect spreadsheets, CRM, and finance data in minutesTrends, KPIs, forecasts & specialized analyticsCompetitor, customer & sentiment intelligenceBuild dashboards without a BI teamReports and board decks from your analysesAgentic AI for plain-English Q&A when you want itStart free — no credit card required
Analytics6 min

Sentiment Analysis

Sentiment Analysis — Customer Voice Intelligence — turns unstructured feedback into decisions. Analyze reviews, NPS responses, support tickets, social mentions, call transcripts, documents, and recordings, and the module scores sentiment, surfaces themes and aspect-level signals, detects emotions, urgency, and churn risk, and recommends actions for every record.

At a glance

Level
Beginner
Duration
6 min
Product
Intelligence Navigator
Prerequisites
Feedback with a text column — or a transcript, document, or recording
Audience
Owners, founders, firms, technical & non-technical teams
Subjects
Sentiment & aspectsthemesemotionsurgency & riskmulti-source importconversation journey

Analyze feedback from anywhere

You're not limited to spreadsheets. Import and analyze from a dataset, paste a call or chat transcript, upload a document (PDF, Word, or text), drop in an audio or video recording (transcribed with timestamps), connect a call platform like Gong or Zoom, or pull live Google Reviews. Everything runs through the same engine and renders in one results view.

  • Dataset — reviews, NPS responses, support tickets, social mentions, app-store reviews
  • Transcript, document, or audio/video recording — transcribed and scored automatically
  • Call platforms (Gong, Zoom) and live Google Reviews via Integrations

See Sentiment Analysis in action

Watch: import feedback, run the analysis, and read distribution, themes, emotions, and recommended actions.

Simple mapping, one required field

Only the text column is required. Add optional columns to enrich the analysis: a record ID to see who gave positive or negative feedback, a date for trend-over-time, a numeric rating (stars, NPS, CSAT) to correlate with text sentiment, and a source channel for a source-level breakdown.

Two scoring modes

Standard scoring is fast and local (VADER) — good for large datasets and quick passes. High-accuracy AI re-scores each record and handles sarcasm, negation, and mixed sentiment; it uses more credits and takes longer. Pick per run based on how much precision you need.

Read the voice of the customer

Results go far beyond positive/neutral/negative. See the distribution and trend over time, key themes and topics with mention counts, aspect-based sentiment (what features people praise or complain about), top emotions, risk signals, and a voice summary with the main negative drivers — each record with its sentiment, emotion, urgency, and a recommended action.

  • Aspect-Based Sentiment — which features or topics skew positive or negative, with mention counts and average score
  • Emotions, urgency, and risk signals flag what needs attention now
  • Conversation Sentiment Journey charts how a call or chat moved from opening to closing — recovered, deteriorated, or steady

Share and act

Present results in Presentation Mode, save charts to a Dashboard, Presentation, or Report, download the per-record table as CSV, or copy it to share. To keep a pulse on customer voice continuously, hand the analysis to the automation modules so it re-runs and alerts on its own.

Best practices

  • Map a date column so you can track sentiment trend, not just a snapshot
  • Add a rating column to validate text sentiment against stars/NPS
  • Use High-accuracy AI scoring when sarcasm or mixed sentiment matters — e.g. support tickets
  • Watch risk signals and urgency to catch churn-prone customers before they leave

Next steps

You've finished the Analytics modules. Next, OKRs help you set and track the goals these insights point to.