Design an n8n workflow for any sales process
Produces a node-by-node n8n build plan with trigger choice, data shapes between nodes, error workflows, edge cases, and a five-record test plan. It replaces the usual approach — dragging nodes until something works — with a design you can build in one sitting and debug six months later without the original builder.
You are an n8n architect who has shipped over 200 revenue workflows and inherited plenty of spaghetti builds where one dead node silently killed a pipeline for weeks. You design for observability first: every workflow you spec can fail loudly and be debugged by someone who didn't build it. Design an n8n workflow for the sales process I describe. Deliver this exact output: 1. WORKFLOW SUMMARY — one paragraph, end to end. 2. TRIGGER NODE — type (webhook, schedule, app trigger) and why. 3. NODE-BY-NODE PLAN — a numbered list; for each node: node type, what it does, key configuration notes, and the shape of the data it passes on. 4. ERROR HANDLING — error workflow design, retry logic, and where failure notifications go (Slack or email node). 5. EDGE CASES — at least 3 inputs that would break the naive build, and how the design absorbs each. 6. TEST PLAN — how to validate with 5 records before switching on volume. Rules: never chain two AI nodes without a validation step between them. No 'one Code node that does everything' — if a Code node has more than one clear job, split it. Flag every step where a native integration beats a raw HTTP Request node, and vice versa. Before you design, interview me. Ask these questions ONE AT A TIME, waiting for my answer each time: 1. Describe the process this workflow should run, start to finish. 2. What triggers it, and how many times per day will it fire? 3. Which tools does it need to touch, and do you have API access to each? 4. What should happen when a step fails at 2am? 5. Who maintains this after it ships — you, or someone less technical? Then produce the design. If any part of my process is the wrong job for n8n, say so and propose where that step should live instead.
How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Answer the interview; be precise on question 3 — API access gaps are the number one reason builds stall halfway.
- 3
Build the workflow in n8n following the node plan, running the test plan before enabling the trigger.
- 4
When something breaks later, paste the failing node's error back into the same chat for a targeted fix.
Best practices
Tell it your real daily volume — a workflow firing 20 times a day gets designed differently than one firing 5,000 times.
If the plan includes an AI node, ask for the exact prompt and output format for that node too.
Insist on the error workflow even for 'simple' builds; silent failures cost more than the hour it takes to add alerts.
Screenshot-free maintenance matters: paste the final node list into your ops docs next to the workflow name.
Example: what this looks like in practice
A founder at a 12-person agency-adjacent consultancy wants positive email replies pushed from Smartlead into their CRM with an enrichment stop along the way. The interview establishes a webhook trigger from Smartlead, about 30 fires a day, HubSpot API access, and a non-technical VA as maintainer. The design comes back with 9 nodes: webhook, reply-sentiment filter, Clay enrichment call, a validation IF node, HubSpot upsert, Slack notification, plus a connected error workflow that posts failures to an #ops-alerts channel. The test plan catches that out-of-office replies were being marked positive before launch, and the fix is one added condition rather than a week of polluted CRM data.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Yes — it sits in the sweet spot between Zapier's simplicity and custom code. n8n handles webhooks, branching, AI nodes, and self-hosting, which makes it popular for outbound teams wiring Smartlead, Clay, and CRMs together. The tradeoff is a steeper learning curve, which is exactly why designing the workflow on paper first, as this prompt does, pays off.
More automation & workflow design prompts
Build a system prompt for an AI cold email reply agent
Interviews you about your offer, booking flow, and voice, then generates a complete production system prompt for an AI reply agent — classification taxonomy, response rules, escalation conditions, and a strict JSON output contract an n8n or Zapier workflow can parse. You get the artifact teams pay thousands to have written, tuned to how you actually talk.
Turn a manual sales process into an automation blueprint
Converts a fuzzy 'we do this by hand every week' description into a structured blueprint with a detectable trigger, labeled steps, edge cases, human checkpoints, and failure alerts. It forces the design decisions people usually discover mid-build, so what you hand to n8n, Zapier, or an ops hire is assembly work, not archaeology.
Architect a Clay table before you burn the credits
Designs the Clay table before you build it: sources, column order, run conditions, waterfalls, outputs, and a credit estimate. Column ordering and gating are where most teams silently overspend, so the prompt enforces filter-first architecture and makes every paid column defend its run condition — typically cutting credit burn 40-70% versus enrich-everything builds.