Aakash GuptaSeriously, Please Watch This Before You Declare n8n Dead
CHAPTERS
- 0:00 – 3:21
Is n8n actually “dead”? Hype cycles, Google Trends, and why Jan says you still need it
Aakash opens with viral skepticism about n8n’s relevance, then Jan reframes the narrative: n8n has been “declared dead” many times and hype cycles distort trend lines. Jan positions n8n as complementary to agentic coding tools rather than replaced by them.
- •Viral tweet and Google Trends spark the “n8n is irrelevant” claim
- •Jan argues hype cycles create predictable rise-and-fall attention curves
- •n8n vs Claude Code: different categories (orchestration layer vs terminal agent tool)
- •Core thesis: business-critical workflows need inspectability and stability
- 3:21 – 7:09
n8n vs Claude Code: orchestration, collaboration, and moving from prototype to production
Jan explains why teams often prototype quickly with Claude Code, then migrate to n8n for durable, shareable systems. The differentiators center on auditability, deterministic logic, self-hosting, and collaboration between technical and non-technical teammates.
- •Claude Code is fast for prototypes; n8n is for operational workflows
- •Visual workflows make it easier to inspect and discuss what’s running
- •Self-hosting and privacy/security needs are increasingly important
- •n8n supports complex pipelines and more precise control than “black-box” code generation
- 7:09 – 10:45
Ad break: personalized lifecycle messaging and prompt-based experimentation
Aakash reads sponsor messages focused on lifecycle marketing automation and experimentation tooling. The emphasis is on behavior-based messaging, AI-assisted journey creation, and faster experiment iteration without heavy engineering lift.
- •Customer.io: behavioral triggers, branching journeys, and AI agent assistance
- •MCP server positioning: AI tools can access workspace context directly
- •Kameleoon: prompt-based experimentation to reduce developer dependency
- •Theme: operational tooling that makes growth loops faster
- 10:45 – 11:41
Did Claude Cowork / Codex hurt n8n? Shifting from personal automations to business-critical use cases
Aakash presses on whether new agent products impacted n8n’s business. Jan argues these launches often increase interest rather than cannibalize, while n8n’s center of gravity has moved toward higher-stakes workflows where reliability matters most.
- •“We got killed by OpenAI” week was one of n8n’s fastest growth weeks
- •Market hype boosts awareness; real shift is in use-case mix
- •n8n seeing growth in business-critical vs personal “email summary” automations
- •Reliability, compliance, and team workflows become the main driver
- 11:41 – 15:22
n8n’s scale: revenue, users, enterprise adoption, and SAP distribution
Jan shares updated traction: beyond $100M ARR, 1.5M active users, and 1,200+ enterprise customers on the enterprise plan. He highlights SAP’s integration/distribution and examples like Mercedes to underscore large-company adoption and ROI focus.
- •Crossed $100M+ ARR and continuing strong growth (no further public breakdown)
- •1.5M active users; 1,200+ enterprise plan customers
- •SAP investment + “out-of-the-box” availability for SAP customers
- •Jan emphasizes real ROI vs hype-driven metrics
- 15:22 – 22:46
Live demo: building an email + calendar agent in n8n with models, tools, approvals, and fallbacks
Jan walks through an n8n workflow: trigger, agent, model selection, tool calls (Gmail/Calendar), and human approval gates. The demo highlights model routing (including fallback) and step-by-step execution visibility while performing real actions.
- •AI assistant can generate workflows from plain-language prompts
- •Default model (Claude Sonnet) via n8n gateway; optional own accounts
- •Fallback models to handle provider unreliability
- •Human-in-the-loop approval before sensitive actions (e.g., sending email/creating events)
- •Execution trace shows each node’s inputs/outputs for debugging
- 22:46 – 24:41
Reliability and auditability explained: self-hosting, maintained integrations, retries, and step-level provenance
Aakash asks for concrete definitions of “reliability” and “auditability.” Jan breaks down operational guarantees: infrastructure control via self-hosting, maintained connector code, retries/observability, and the ability to replay/debug from specific workflow points with full run history.
- •Self-hosting reduces dependency on external network conditions and supports privacy
- •Connectors/tools are maintained and updated centrally when APIs change
- •Operational features: retries, execution logs, data handling, debugging flows
- •Auditability = seeing what ran, why it branched, and what data moved between steps
- •Deterministic nodes + AI nodes enable controllable, inspectable systems
- 24:41 – 29:40
Ad break: Bolt.new, Jira Product Discovery, and AI PM Certification
Aakash reads sponsor segments about shipping real code from AI tools, managing discovery in Jira’s ecosystem, and training PMs to build and evaluate AI products. The common theme is taking prototypes to production and building rigor around AI systems.
- •Bolt.new: exportable source code enables engineer review and iteration
- •Jira Product Discovery: connect discovery/prioritization/roadmaps to delivery
- •Product Faculty AI PM Certification: agents, evals, routing, RAG, guardrails, scaling
- •Reinforces “prototype becomes product” and operational AI skill-building
- 29:40 – 30:58
Handing workflows to teammates: sharing, versioning, exporting, and review/approval flows
Jan shows how n8n supports collaboration: inviting teammates, version history, export/import, and review steps before publishing. This section frames n8n as a shared operational asset rather than individual scripts living in one person’s terminal.
- •Share workflows with teammates via access/invites
- •Version history with publishable versions (Git-like)
- •Export workflows to send via email or move between environments
- •Review steps (approval gates) before publishing changes
- •Supports handoff and governance in teams/enterprises
- 30:58 – 32:42
Where n8n shines in enterprises: security orchestration, HR lifecycle flows, DevOps and monitoring
Aakash challenges whether n8n’s demo is unique; Jan answers with higher-stakes enterprise examples. He emphasizes workflows where partial failure is unacceptable—security triage, attachment scanning, onboarding/offboarding, and DevOps automation/monitoring.
- •Security orchestration: scan attachments, archive, alert on issues
- •Sensitive data workflows (employment data) need strong controls
- •Onboarding/offboarding must be completed fully and correctly
- •DevOps use cases and operational automations
- •Core test: choose n8n when you need 100% completion and inspectability
- 32:42 – 35:42
What PMs should build next: start small, focus on repetitive cross-app work, and use templates for leverage
Jan advises PMs to pick high-frequency pain points rather than “transform everything” projects. He recommends simple workflows (sometimes without AI) that save real time, then expanding through internal sharing and template-driven inspiration.
- •Identify daily repetitive tasks involving copy/paste across tools
- •Start with small workflows; avoid weeks-long ‘big bang’ builds early
- •Deterministic automation can deliver huge ROI (e.g., password reset workflows)
- •Show quick wins to spark org-wide adoption
- •Template library offers 10,000+ workflows for reuse and inspiration
- 35:42 – 39:45
Template library tour: scraping, enrichment, alerts, SSL monitoring, and AI-assisted incident response patterns
Jan tours the template library and calls out popular patterns like web scraping, enrichment to Sheets, and sending results to Slack. He also highlights monitoring workflows (SSL expiry checks) and alerting patterns that combine deterministic steps with optional AI auto-fixes.
- •Filter templates by tools (e.g., Google Sheets) and functions (sales/marketing/IT ops)
- •Common use cases: web scraping, enrichment, and routing results to Slack
- •Monitoring templates: SSL certificate tracking and multi-channel alerts
- •Pattern: always-notify deterministically, then attempt AI-based auto-remediation
- •Templates can be copied directly and activated by adding credentials
- 39:45 – 41:42
Zapier vs n8n: power, flexibility, self-hosting, and deeper AI agent capabilities
With Zapier top-of-mind, Jan positions n8n around extensibility and complexity handling. He emphasizes features like code nodes, configurable agent memory/tools, strict output controls, fallback models, human approvals, and self-hosting—especially as workflows evolve beyond simple automations.
- •n8n optimized for power/flexibility; build simple then scale complexity
- •Code nodes enable escape hatches for custom logic
- •Advanced agent design: memory, tool selection, output shaping/constraints
- •Fallback models + human-in-the-loop for safer operations
- •Self-hosting is a major differentiator vs SaaS-only alternatives
- 41:42 – 1:09:05
Product + growth philosophy: ‘AI as core’ (not sprinkled), pricing choices, community-led distribution, and hiring technical AI PMs
Jan explains why “sprinkling AI” yields limited gains versus embedding AI in the core value chain—reflected in n8n’s agent-heavy workflows and rapid growth. He covers long-term metrics choices (de-emphasizing lead gen/per-seat), an efficiency-first org goal, community-driven enterprise adoption, product team structure, and what n8n looks for in technical PMs—including eval maturity and deep understanding over buzzwords.
- •‘Sprinkling AI’ = superficial features; ‘AI as core’ = enabling users to build agents/workflows
- •80% of workflows use AI agents; users evolved by adopting AI, not being replaced
- •Rejecting per-seat + killing lead-gen targets to optimize long-term adoption
- •Org efficiency goal: massive scale with constrained headcount; profitability enables patience
- •Community flywheel drives enterprise: users bring n8n into large orgs
- •Product org: small squads (3–5 eng), PM(s), designer; CEO stays close to design quality
- •Hiring: technical depth, builder mindset, live problem-solving to detect real competence; evals important and owned by AI Trust team
