Do I need a custom n8n node for LeadMind?

No. The built-in HTTP Request node calls the REST API and the Webhook node receives the job.delivered event. The n8n setup guide shows both.

Can I build the workflow without spending credits?

Yes. A test key (lp_test_…) runs the whole job flow on fixture data, so every node can be wired and tested before you use a live key.

How do I stop a workflow from creating duplicate jobs on retry?

Send an Idempotency-Key header with each POST /v1/jobs, for example the n8n execution id. A retried request returns the original job instead of starting a new one.

Five n8n lead generation workflows

Every LeadMind automation in n8n is built from the same two pieces: an HTTP Request node that starts a job from a sentence, and a Webhook node that fires when the leads are ready. The n8n setup guide wires those up; this page shows five patterns to build on top.

Guide · By LeadMind AI (how we source and check data) · Updated 25 September 2026

Start free Watch the demo

1. Scheduled prospecting

Schedule Trigger (every Monday) → HTTP Request POST /v1/jobs with a fixed prompt such as “30 new gyms in Dubai Marina with phone numbers” and maxCredits → Webhook on job.delivered → Google Sheets. The 30-day dedupe means each run only delivers, and only charges for, businesses you have not already received.

2. Form to lead list

n8n Form Trigger (or Typeform) collecting “what leads do you need?” → POST /v1/jobs with the answer as the prompt → on delivery, email or Slack the requester the CSV link from data.links.export. Sales reps get lists without logins; one API key with a daily cap controls the spend.

3. Verify before the CRM

CRM trigger on a new contact → HTTP Request POST /v1/verify/emails with up to 25 addresses → IF status is deliverable → create the contact; otherwise tag it for review. Treat risky as “catch-all or disposable”, not as bad (why). Verification uses the monthly quota of a paid plan, not lead credits.

4. Lead requests from chat

Slack or Telegram trigger → POST /v1/jobs with the message as the prompt → on job.awaiting_clarification, post the question back to the thread and answer it with POST /v1/jobs/{id}/clarify → on job.delivered, reply with the top rows and the CSV link.

5. Route by score

On job.delivered → HTTP Request GET /v1/jobs/{id}/leads (follow nextCursor) → IF score ≥ 70 → CRM as a sales-ready lead; below → a nurture sheet. The scoreReason field is worth copying into the CRM note so the rep knows why the lead was picked.

Make it robust

Verify the X-LeadMind-Signature HMAC on every webhook in a Code node before trusting the payload.
Send Idempotency-Key on job creation so n8n retries never start a second job.
Handle job.failed: nothing is charged, and the body carries a stable errorCode.
Build with an lp_test_ key, then swap in a live key.

Endpoint details are in the API reference.

Frequently asked questions

Do I need a custom n8n node for LeadMind?: No. The built-in HTTP Request node calls the REST API and the Webhook node receives the job.delivered event. The n8n setup guide shows both.
Can I build the workflow without spending credits?: Yes. A test key (lp_test_…) runs the whole job flow on fixture data, so every node can be wired and tested before you use a live key.
How do I stop a workflow from creating duplicate jobs on retry?: Send an Idempotency-Key header with each POST /v1/jobs, for example the n8n execution id. A retried request returns the original job instead of starting a new one.