Post-Call Deal Updater

Transform sales call transcripts into structured deal intelligence, CRM updates, and follow-up tasks — automatically.

The Post-Call Deal Updater is a 17-node n8n workflow that receives a call transcript via webhook, extracts key information (commitments, objections, buying signals, next steps), scores deal health on a 1-10 scale, generates CRM-ready outputs, and writes results directly to Pipedrive. Three specialized AI agents work in sequence: Extractor identifies structured data from raw conversation, Analyst scores deal health across 4 weighted criteria, and Scribe generates actionable CRM content with at-risk flagging. Independent Test Protocol (ITP)-certified: 20/20 PASS (100%), $0.085/call average. We designed this after sitting through a pipeline review where the AE admitted the CRM notes were three calls behind. The pipeline listens for call summaries and pushes structured updates to the deal record within minutes.

Last updated March 6, 2026

Sales teams send and receive thousands of emails per week, but fewer than 10% receive systematic follow-up analysis. Response time, sentiment shifts, and intent signals buried in email threads go undetected. Automated communication intelligence catches patterns that human review at scale cannot.

3-Agent PipelineinputCall Transcript01ExtractorCommitments + Objections02AnalystDHS 1–10 + at_risk03ScribeCRM Content + ActivityoutputPipedrive + ActivityDHS < 5 triggers at_risk flag$0.09/call — Cheapest in Lineup

Three Agents. One Transcript. Complete Deal Intelligence.

The Extractor

Step 1The Extractor

Tier 1 Reasoning

Parses raw call transcript to identify commitments (with party, specificity, timeline), objections (severity, status, context), buying signals, and next steps. Structured extraction — no interpretation yet.

The Analyst

Step 2The Analyst

Tier 1 Reasoning

What does The Analyst actually decide? Scores Deal Health (DHS 1–10) across 4 weighted criteria: Commitment Quality (30%), Objection Severity (25%), Momentum Signals (25%), and Next Step Clarity (20%). Sets at_risk flag when DHS < 5.

The Scribe

Step 3The Scribe

Tier 1 Reasoning

Without this step, upstream analysis sits idle. Generates CRM-ready content: call summary, deal note, and follow-up task with due date. When at_risk is true, prefixes tasks with URGENT and frames output around risk signals. Writes 5 custom Pipedrive deal fields + creates a follow-up activity.Same prompt, same input, different scores across runs. That is why ITP tests run each fixture multiple times and document variance ranges.

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 transcribe calls — you provide the transcript text via webhook

×

Does not access external data sources — all intelligence extracted from the transcript only

×

Does not change Pipedrive deal stages — output is custom field updates and a follow-up activity

×

Does not use web_search — cheapest product in the lineup at $0.09/call

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

The Complete Customer Success Bundle

9 files.

CHANGELOG.mdVersion history
LICENSE.mdLicense terms
README.mdSetup and configuration guide
dependency_matrix.mdThird-party service dependencies
error_handling_matrix.mdError handling reference
post_call_deal_updater_v1.jsonn8n workflow (main pipeline)
prompts/analyst.txtAnalyst system prompt
prompts/extractor.txtExtractor system prompt
prompts/scribe.txtScribe system prompt

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

Tested. Measured. Documented.

All values below are from Independent Test Protocol (ITP) testing — not estimates, not projections. Measured across 20 test transcripts spanning strong calls, weak calls, stalled deals, competitive mentions, and edge cases.

Workflow Nodes

17 specialized n8n nodes

Blueprint Quality Standard

BQS v2 — 12/12 PASS

Agent Roles

3 (Extractor, Analyst, Scribe)

Credentials Required

Anthropic API + Pipedrive API

Bundle Files

9 files

Cost Advantage

No web_search costs — cheapest product in lineup

Cost per Call (ITP-Measured)

$0.085 (ITP-measured average)

Inspection & Test Plan

PCDU-01 through PCDU-06 — ALL PASS

DHS Consistency (3 Runs)

[9, 9, 9] variance=0

n8n Versions Tested

Self-hosted

Tested on n8n v2.7.5, March 2026

Post-Call Deal Updater v1.0.0 — Technical Reference━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Pipeline:     Webhook → Extractor → Analyst (DHS) → Scribe → Pipedrive Deal Update + ActivityTopology:     Modular Agentic Swarm (3 agents, linear pipeline)Routing:      No IF node — pipeline always runs to completion. at_risk flag modifies Scribe behavior.Node Count:   17LLM:          Claude Opus 4.6 (Extractor, Analyst, Scribe — Tier 1 Reasoning)Cost:         $0.085/call (ITP-measured avg). No web_search costs. Cheapest product in lineup.ITP:          20/20 PASS (100%)BQS:          12/12 PASSTool A:       Webhook (call transcript input)Tool B:       Pipedrive (deal update + activity creation)

What You'll Need

Platform

n8n 2.11.2+

Est. Monthly API Cost

$5–20/month

Credentials Required

  • Anthropic API
  • Pipedrive API

Services

  • Pipedrive CRM (Professional plan for custom fields)

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

Post-Call Deal Updater v1.0.0

$199

one-time purchase

What you get:

  • 17-node n8n workflow (.json) — you own it
  • 3 tested agent system prompts (ITP-validated)
  • Error handling matrix (33 failure modes documented)
  • 5 Pipedrive custom field setup instructions (fw_ prefix)
  • Dependency matrix with ITP-measured costs
  • README setup guide (10 minutes)
  • All sales final after download

Frequently Asked Questions

What is DHS (Deal Health Score)?+

DHS is a 1–10 composite score across four weighted criteria: Commitment Quality (30%), Objection Severity (25%), Momentum Signals (25%), and Next Step Clarity (20%). Higher scores indicate stronger deal health. The Analyst agent evaluates each criterion independently with explicit reasoning.

What happens when a deal scores below 5?+

The Analyst sets at_risk: true in its output. The Scribe detects this flag and adapts: follow-up task subject is prefixed with URGENT, the deal note highlights risk factors, and the call summary frames the conversation around warning signals. No separate routing node — the behavior change is reasoning-layer conditioning.

What Pipedrive custom fields does this write?+

Five custom deal fields with the fw_ prefix: fw_dhs (Deal Health Score), fw_key_commitments (extracted commitments), fw_objections_raised (objections with severity), fw_last_call_summary (Scribe-generated summary), and fw_call_analyzed_at (ISO 8601 timestamp). The README includes step-by-step creation instructions.

Does this product use web_search?+

No. The Post-Call Deal Updater processes transcript text only — no web searches, no external data lookups. This makes it the cheapest product in the ForgeWorkflows lineup at $0.085 per call.

What triggers the workflow?+

A webhook POST with two required fields: deal_id (Pipedrive deal ID) and transcript (call transcript text, minimum 50 characters). Optional fields: call_duration_minutes, participants, and call_date. Connect your call recording platform to the webhook URL after activation.

What credentials are required?+

Two credentials: Anthropic API key (used by all 3 agents) and Pipedrive API token (for deal field updates and activity creation). Pipedrive Professional plan or higher required for custom deal fields.

How consistent is the DHS scoring?+

ITP consistency test: the same transcript scored [9, 9, 9] across 3 consecutive runs. Variance = 0. The Analyst uses Tier 1 Reasoning for maximum scoring reliability.

What does this blueprint NOT do?+

It does not transcribe calls — you provide the transcript text. It does not access external data sources — all intelligence comes from the transcript. It does not route deals automatically — output is Pipedrive field updates and a follow-up activity, not pipeline stage changes.

How much does each call analysis cost?+

Approximately $0.085 per call. No web_search costs. At 200 calls per month, total API cost is approximately $17.

Is there a refund policy?+

All sales are final after download. Review the Blueprint Dependency Matrix and prerequisites before purchase. Questions?

Read the full guide →

Related Blueprints