Zoho Sentiment × Dental Intel

No filters active — click anything to drill down

Tickets analyzed

High escalation

Negative sentiment

Avg sentiment

Tickets / $100K TTM

Practice leaderboard real SGA practices · click to filter

Practice	TTM $	Tickets	T/$100K	Neg	Avg	% neg

Category heatmap click to filter

Category	Tickets	Neg	% neg

Root cause click a bar

Requester state click a bar

Top emotions click a bar

Tickets vs TTM production real DI TTM through · click a bubble to filter

Bubbles above the trendline generate more tickets than production predicts (friction tax). Far below with low production = possible silent suffering.

Sentiment click a slice

Escalation risk click a slice

Focus here first

Practices ranked by severity × size × friction. Click any row to drill in — filter the dashboard and see the full context. How is this scored?

Algorithm + weights (click for details)

ticket_severity (0-11 per ticket):

Escalation: high=5, medium=2, low=0
Sentiment: very_negative=3, negative=2, neutral=0, positive=-1, very_positive=-2
Requester state: escalating=3, resigned=3, frustrated_but_trying=1, else=0

size_multiplier = clamp(TTM / median_TTM, 0.5, 2.0) — bigger practice = bigger revenue at risk

friction_flag = 1.5× if tickets/$100K exceeds peer median, else 1.0×

These weights are judgment calls. When real data lands we'll see which practices "feel wrong" in the top 5 and tune. The numbers are exposed here so nothing is a black box.

All tickets

Ticket	Practice	Category	Pri	Score	Emotions	Esc	Cause	State	Conf	Model

Methodology + column legend

Score -1 to +1 (sentiment); Conf model self-reported (0-1). Esc model-judged escalation risk — hover the row to see the reason.

Routing: short/low-priority → Haiku 4.5; long/urgent/keyword-flagged → Sonnet 4.6.

Data: 50 English rows from Tobi-Bueck/customer-support-tickets, reshaped into Zoho schema with real SGA practice metadata from Dental Intelligence. For dental-shaped signal, feed a real Zoho Desk export.