No-Show Predictor

AI scores no-show risk the moment a Calendly meeting is booked, then sends a personalized recovery email to high-risk prospects before they ghost.

AI scores no-show risk the moment a Calendly meeting is booked, then automatically sends a personalized recovery email to high-risk prospects before they ghost. 31-node swarm. $0.148/booking blended. We built this after an AE team calculated they lost 6 hours per week to meetings that never happened. The predictor scores booking risk factors so reps can send targeted confirmations to high-risk slots.

Last updated March 9, 2026

No-show rates for B2B meetings average 15-20%, with some segments hitting 30%. Each no-show wastes 30-60 minutes of prep and calendar space. Pattern analysis on historical no-show data enables proactive intervention — confirmation sequences, rescheduling prompts, priority re-ranking — before the empty calendar slot.

Real-Time PipelinewebhookCalendly01ResearcherSonnet 4.602Scorer5 signals — Opus 4.603Risk Routertier + confidence?HIGH+confEmail+Act +NoteMEDNoteLOWLog$0.148/booking — dual LLMITP-measured across 20 bookings

Four Agents. Five Risk Signals. Recovery Before the Ghost.

The Researcher

Step 1The Researcher

Tier 2 Reasoning

Fires on every Calendly booking webhook. Enriches the prospect with CRM history, booking context, and engagement signals from Pipedrive. the analysis model.

The Scorer

Step 2The Scorer

Tier 1 Reasoning

What does The Scorer actually decide? Scores no-show risk across 5 signals: Lead Quality, Engagement History, Timing Risk, Booking Context, and Cold Signal. Returns risk tier (HIGH/MEDIUM/LOW) plus confidence score (0-1). the primary reasoning model.

The Risk Router

Step 3The Risk Router

IF Logic

This step exists because raw data alone is not enough. Confidence-gated routing. HIGH risk + confidence ≥ 0.7 → recovery email + Pipedrive Activity + Note. MEDIUM → Note only (human review). LOW → log only (no noise).

The Syncer

Step 4The Syncer

HTTP

Without this step, upstream analysis sits idle. Writes to Pipedrive and Gmail. HIGH-confidence bookings get a personalized recovery email with AI-generated angle specific to the prospect — never a generic template. Activity and Note created for all HIGH-risk bookings.When 3 of 10 leads scored below threshold, the fix was the pipeline, not the criteria. Acceptance criteria is locked — no post-hoc relaxation.

That's the full pipeline. Here's what it intentionally does NOT do — and why those boundaries exist.

What It Does NOT Do

×

Does not cancel or reschedule meetings — it predicts no-show risk and sends recovery emails before the meeting

×

Does not work with calendars other than Calendly — no Google Calendar, Outlook, or Cal.com integration

×

Does not work with CRMs other than Pipedrive — no HubSpot, Salesforce, or custom CRM integration

×

Does not send emails for MEDIUM or LOW risk — only HIGH risk with confidence ≥ 0.7 triggers outreach

With those boundaries clear, here's everything that ships when you purchase.

The Complete Customer Success Bundle

7 files.

CHANGELOG.mdVersion history
README.mdSetup and configuration guide
blueprint_dependency_matrix.mdThird-party service dependencies
no_show_predictor_v1_0_0.jsonn8n workflow (main pipeline)
risk_scoring_rubric.mdRisk scoring rubric
system_prompt_researcher.txtResearcher system prompt
system_prompt_scorer.txtScorer system prompt

The technical specifications below are ITP-measured, not estimated.

Tested. Measured. Documented.

Every metric is Independent Test Protocol (ITP)-measured. The No-Show Predictor scores risk at $0.148/booking with dual LLM calls (Sonnet research + Opus scoring).

Workflow Nodes

31 specialized n8n nodes

Tested on n8n v2.7.5, March 2026

No-Show Predictor v1.0.0 — Technical Reference━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Pipeline:     Calendly Webhook → Researcher (Sonnet 4.6) → Scorer (Opus 4.6) → Risk Router (IF) → Syncer (Gmail + Pipedrive)Topology:     Real-time event-driven, dual LLM, confidence-gated recoverySignals:      LEAD_QUALITY | ENGAGEMENT_HISTORY | TIMING_RISK | BOOKING_CONTEXT | COLD_SIGNALRouting:      HIGH + conf ≥ 0.7 → Email + Activity + Note | HIGH + conf < 0.7 → Note | MEDIUM → Note | LOW → LogNode Count:   31 (Webhook 1, HTTP 6, Code 12, IF 4, LLM 2, Gmail 1, Dead Letter 1, ITP Bypass 4)LLM:          Claude Sonnet 4.6 (Researcher) + Claude Opus 4.6 (Scorer), temperature 0.2Cost:         $0.148/booking blended | dual LLM callITP:          20/20 (100%) — NSP-01 through NSP-14BQS:          12/12 PASSConsistency:  NSP-03 variance=0 across 3 runs (threshold ≤ 2.0)

What You'll Need

⚠ Data Source Dependency

This blueprint uses Anthropic's managed web search tool to research contacts and companies. Search results depend on live web data and may vary.

Platform

n8n 2.11.2+

Est. Monthly API Cost

$5–15/month

Credentials Required

  • Anthropic API
  • Calendly API
  • Pipedrive API
  • Gmail OAuth2

Services

  • Calendly account
  • Pipedrive CRM
  • Gmail account

Setup Track

Quick Start

~15 min

All credentials live, n8n running

Full Setup

1–2 hrs

Needs API config + tables

From Scratch

2–4 hrs

No n8n, no credentials

No-Show Predictor v1.0.0

$199

one-time purchase

What you get:

  • ITP-tested 31-node n8n workflow — import and deploy
  • Real-time Calendly webhook trigger — fires on every booking
  • 5-signal no-show risk scoring with confidence calibration
  • Confidence-gated recovery: email only when HIGH + confidence ≥ 0.7
  • Personalized recovery emails — AI-generated angle per prospect, not templates
  • Pipedrive Activity + Note for every HIGH-risk booking
  • Full ITP test results with 20 fixtures and cost analysis
  • BQS-certified (12/12 PASS)

Frequently Asked Questions

How does the Calendly webhook trigger work?+

The workflow listens for Calendly invitee.created events. Every time a prospect books a meeting, the pipeline fires automatically — no polling, no batching, no manual trigger.

What are the 5 risk signals?+

Lead Quality (CRM history, deal stage), Engagement History (email opens, meeting history), Timing Risk (day/time patterns, timezone gaps), Booking Context (short notice, rescheduled), and Cold Signal (no prior contact, unknown source).

When does a recovery email actually get sent?+

Only when risk is HIGH and confidence is ≥ 0.7. This dual gate prevents false positives — the system must be both concerned and confident before emailing a prospect. MEDIUM risk gets a Pipedrive Note for human review instead.

Are the recovery emails generic templates?+

No. Each recovery email uses an AI-generated angle specific to the prospect based on their CRM history, booking context, and risk signals. The Scorer produces the angle, and the Syncer drafts a personalized email — no two are alike.

How much does each booking cost to process?+

ITP-measured: $0.148/booking blended average. This includes two LLM calls — the analysis model for research and the primary reasoning model for scoring. A team processing 100 bookings/month spends ~$14.80/month on API costs.

Why two different LLM models?+

the analysis model handles research (context gathering, enrichment) where speed matters. the primary reasoning model handles scoring where reasoning depth matters — calibrating confidence across 5 signals requires stronger inference. This dual-model architecture balances cost and accuracy.

What CRM and calendar does this work with?+

Calendly for booking triggers and Pipedrive for CRM sync. Recovery emails go through Gmail OAuth2. It requires API credentials for all four services: Calendly, Pipedrive, Gmail, and Anthropic.

Read the full guide →

Related Blueprints