Analytics & insights
Sentiment
per turn.
See the exact moment a call went sideways. Per-turn sentiment scoring surfaces the trigger word, the dropped tone, the recovery — coaching-ready.
30-minute free trial. No card. No procurement.
How it works
Score.
Visualize. Search.
Drag, wire, deploy. Each step is reversible — every save creates a versioned snapshot.
Score per turn
Each caller + agent turn scored on emotion + valence. Sub-30s from call end to full timeline.
Outcome
Booked
Sentiment
+0.71
Duration
2:14
▸ captured · intent: booking
▸ captured · slot: Tue 14:00
Visual timeline
Call transcript overlaid with sentiment curve. Inflection points highlighted with the trigger turn.
Evals passing
4 / 5
Avg score
88.8
+1.2
Calls evaluated
12,840
Filter + cohort
Search calls by sentiment shift ('calls that started positive + ended negative'). Build cohorts. Compare across reps + workflows.
Calls today
12,840
+8.1%
Pickup rate
94.2%
+1.4%
Avg sentiment
+0.71
+0.08
Talk minutes
48,212
+12.0%
Call volume · last 30 days
Inflection-point analysis
See the moment.
Fix the prompt.
No bolt-ons. No second platform. Every primitive ships with the editor.
Per-turn granularity
Not a single call-level score — every turn scored so you see what specifically moved the needle.
Multi-axis emotion
Not just positive / negative — emotion families (anger, confusion, satisfaction, urgency) tracked independently.
Inflection detection
Auto-surfaces the turn where sentiment shifted. Coaching surface for reps + prompt-tuning surface for workflow owners.
Searchable + cohortable
'Show me calls that went negative in the first 30 seconds.' Build cohorts. Compare across reps, workflows, regions.
Shipped in production
Real teams, real outcomes.
“TOFA launched a national campaign with Finn — 10K+ calls daily and seamless human escalations.”
Alice Smith
Senior Engineer, Gofts · TOFA
Trust + ecosystem
Plugs in. Audits clean.
Wires into your existing stack via 100+ connectors. Compliant from day one across the workloads that demand it.
Salesforce
HubSpot
Slack
Notion
Linear
ZendeskDon't see yours? Finn plugs into any system via REST + webhooks.
FAQ
Questions and answers.
How granular is the sentiment scoring?+
Per-turn — each caller + agent utterance gets an independent sentiment score on multiple axes (valence + emotion family).
What signals drive sentiment scoring?+
Acoustic (pitch, pace, volume), lexical (word choice), conversational (turn-taking dynamics). Multi-modal model tuned on call-center data.
Can I search for specific sentiment patterns?+
Yes — search filters like 'started positive, ended negative' or 'sentiment dropped after escalation request.' Build cohorts from these queries.
Is sentiment scored in real time during the call?+
Yes — available via webhook during the call for real-time escalation triggers ('caller anger detected → escalate to human').
How does this differ from post-call analysis?+
PCA tags outcomes + structured fields per call. Sentiment timeline is the moment-to-moment emotional curve within each call. Complementary.
Coach the moment
Stop reading summaries.
Watch the inflection.
Sentiment timeline shows where calls go right + wrong. Coach reps on what worked. Drop prompts that didn't.