Slack-to-CRM Activity Logger
Auto-log sales conversations from Slack to your CRM.
Auto-log sales conversations from Slack to your CRM with AI-powered relevance scoring. Slack Event API Webhook (message.channels) triggers per-message CRM relevance analysis. 3-signal CRS taxonomy (deal_specificity, action_implication, stakeholder_involvement) with per-signal scoring (1-10), arithmetic mean composite, and evidence-based assessment. 6-category activity classification (commitment, objection, question, decision, update, non_crm). 2-way routing: CRS >= 6 triggers Pipedrive activity + note creation, CRS < 6 silently filtered. Single-model architecture: the analysis model for structured classification. Pre-LLM filters (bot filter + short message filter) eliminate 40%+ of messages at $0 cost. Channel-to-deal mapping via configurable prefix convention. No web_search dependency. Companion to FRE (#20): FRE captures feature requests from Slack for PMs, SCAL captures sales activities from Slack for sales teams. We designed this after discovering that Slack contained 3x more deal-related activity than the CRM because reps discussed deals in channels but never logged the conversations. The logger captures deal-relevant Slack activity and writes it to CRM records.
Last updated March 15, 2026
CRM data decays at 30% per year — contacts change roles, companies rebrand, phone numbers go stale. Sales teams that do not systematically audit data quality waste pipeline on unreachable contacts. Automated data validation catches decay before it costs pipeline.
Three Agents. Three CRS Signals. Auto-Logged Sales Activity.
Step 1 — Extractor
Webhook + Code
Slack Event API Webhook fires on message.channels events. Extractor parses the raw event payload, extracts message text, sender, channel, and timestamp. Pre-LLM filters run first: bot filter removes automated messages, short message filter eliminates messages under 20 characters. Only human messages with substance pass through to the Classifier. Reduces LLM cost by filtering 40%+ of messages at $0.
Step 2 — Classifier
Tier 2 Classification
What does Classifier actually decide? the analysis model performs CRM relevance scoring across 3 CRS signals: deal_specificity (references to specific deals, timelines, products, pricing), action_implication (commitment signals, next steps, decision indicators), and stakeholder_involvement (decision-maker participation, multi-party interactions). Each signal scored 1–10 with arithmetic mean composite. Classifies activity into one of 6 categories: commitment, objection, question, decision, update, or non_crm.
Step 3 — Syncer
Code + 2-way Route
Without this step, upstream analysis sits idle. Routes based on CRS composite score. CRS ≥ 6: creates Pipedrive activity (type, subject, deal link) and note (full message context, CRS breakdown, category) via channel-to-deal mapping. CRS < 6: silently filtered — no CRM write, no noise. Channel-to-deal mapping uses configurable prefix convention for automatic deal association.We tried three providers for $0.016 savings per lead. Three API keys, three billing accounts — the customer friction was not worth it.
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 monitor private/direct messages — only public channel messages via Event API
Does not create Pipedrive deals — logs activities and notes to existing deals
Does not extract feature requests — that is what Feature Request Extractor does
Does not scrape external websites — all data from Slack events and Pipedrive API
Does not diagnose deal stalls — that is what Deal Stall Diagnoser does
Does not monitor deal stage changes — that is what Deal Intelligence Agent does
With those boundaries clear, here's everything that ships when you purchase.
The Complete Customer Success Bundle
5 files.
The technical specifications below are ITP-measured, not estimated.
Tested. Measured. Documented.
Every metric is Independent Test Protocol (ITP)-measured. The Slack-to-CRM Activity Logger monitors Slack channels for sales conversations, scores CRM relevance across 3 signals with evidence-based assessment, classifies activity into 6 categories, and routes relevant messages to Pipedrive at $0.01/message.
Workflow Nodes
25
Blueprint Quality Standard
12/12 PASS
Agent Architecture
3 agents — Extractor (Code), Classifier (Tier 2 Classification, Sonnet 4.6), Syncer (Code)
Required Credentials
Slack (Event API + Bot Token), Pipedrive API, Anthropic API
Bundle Contents
9 files
Cost per Message
$0.01/message (ITP-measured, Sonnet 4.6 only)
ITP Milestones
14/14 PASS (20 records, 100% category accuracy)
n8n Compatibility
2.11.2
Tested on n8n v2.7.5, March 2026
Slack-to-CRM Activity Logger v1.0.0 — Technical Reference━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Architecture: 25 n8n nodes, 3 agents (Extractor → Classifier → Syncer)Trigger: Slack Event API Webhook (message.channels)Input: Slack Event — message text, sender, channel, timestampIntelligence: Sonnet 4.6 (CRS scoring + activity classification)Output: Pipedrive (activity + note for CRS ≥ 6)Cost: $0.01/message (ITP-measured average)ITP: 20 records, 14/14 milestones PASS, 100% category accuracyBQS: 12/12 PASSTool A: Slack (input — message events via Event API)Tool B: Pipedrive (output — activities and notes)Intelligence: 3-signal CRS + 6-category classification + 2-way routingCost Value: 0.01
What You'll Need
Platform
n8n 2.11.2+
Est. Monthly API Cost
$10/month (1,000 messages/month)
Credentials Required
- ▪Anthropic API
- ▪Slack (Event API + Bot Token)
- ▪Pipedrive API
Services
- ▪Slack workspace (with Event API enabled)
- ▪Pipedrive CRM
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
Slack-to-CRM Activity Logger v1.0.0
$199
one-time purchase
What you get:
- ✓ITP-tested 25-node n8n workflow — import and deploy
- ✓Slack Event API Webhook for message.channels event processing
- ✓3-signal CRS taxonomy: deal_specificity, action_implication, stakeholder_involvement
- ✓Per-signal scoring (1–10) with arithmetic mean composite and evidence-based assessment
- ✓6-category activity classification: commitment, objection, question, decision, update, non_crm
- ✓2-way routing: CRS ≥ 6 logs to Pipedrive (activity + note), CRS < 6 silently filtered
- ✓Pre-LLM filters: bot filter + short message filter eliminate 40%+ at $0 cost
- ✓Channel-to-deal mapping via configurable prefix convention
- ✓Single-model: the analysis model for CRS scoring + activity classification at $0.01/message
- ✓ITP test results with 20 records, 14/14 milestones, 100% category accuracy
- ✓All sales final after download
Frequently Asked Questions
How does it differ from Feature Request Extractor?+
Complementary products covering different Slack use cases. Feature Request Extractor (FRE) monitors Slack for feature requests and creates Linear issues for product teams. Slack-to-CRM Activity Logger (SCAL) monitors Slack for sales-relevant conversations and logs them to Pipedrive for sales teams.
What are the three CRS signals?+
Deal Specificity (references to specific deals, timelines, products, pricing), Action Implication (commitment signals, next steps, decision indicators), and Stakeholder Involvement (decision-maker participation, multi-party interactions). Each scored 1–10 with arithmetic mean composite.
What are the six activity categories?+
Commitment (budget approval, timeline confirmation, verbal agreement), Objection (pricing concern, competitor mention, feature gap), Question (technical question, scope clarification, process inquiry), Decision (go/no-go, vendor selection, scope change), Update (status update, milestone achievement, delivery confirmation), and Non-CRM (internal chatter, social message, off-topic).
When does it write to Pipedrive?+
Only when CRS composite score is 6 or higher. Messages below the threshold are silently filtered — no CRM noise, no alert fatigue. The threshold is configurable in the workflow.
How does channel-to-deal mapping work?+
Uses a configurable prefix convention. For example, channels named #deal-acme-corp or #sales-project-x map to their corresponding Pipedrive deals via prefix matching. The mapping guide explains how to set up your channel naming convention for automatic deal association.
Why only Sonnet instead of Opus?+
CRS scoring and activity classification are structured classification tasks that the analysis model handles with 100% category accuracy in ITP testing. Opus would add cost without measurable quality improvement for this task type. Single-model architecture keeps cost at $0.01/message — the lowest per-record LLM cost in the ForgeWorkflows catalog.
How much does each message cost to process?+
ITP-measured: $0.01/message with the analysis model only. Pre-LLM filters (bot + short message) eliminate 40%+ of messages at $0 cost before they reach the LLM. 1,000 messages/month costs approximately $10/month in LLM usage.
Does it use web scraping?+
No. All data comes from two sources: Slack Event API (message payload) and Pipedrive API (deal lookup for CRM writes). No web_search, no external data sources, no scraping.
Is there a refund policy?+
All sales are final after download. Review the Blueprint Dependency Matrix and prerequisites before purchase. Questions?