Connect n8n to GoHighLevel: The Complete Guide (3 Methods + Free Workflows)

I connect n8n to GoHighLevel almost every week — for restaurant ordering systems, lead pipelines and client automations. There are exactly three ways to do it, and most guides only cover one (often incorrectly). This page covers all three, when to use each, and gives you three importable n8n workflows to start from.

Method 1 — the native HighLevel node

n8n ships a built-in HighLevel node. Add it to your canvas, create a HighLevel credential (OAuth2), and you get ready-made operations for contacts, opportunities and tasks without touching an endpoint.

Use it when: you need standard CRM operations — create/update contacts, move opportunities — and want them working in minutes.

Its limit: the node covers a subset of the API. Invoices, calendars, conversations, custom objects, payments — for those you’ll graduate to Method 3. Most of my production workflows end up on Method 3 for exactly this reason.

Method 2 — webhooks (GHL ⇄ n8n)

GHL → n8n

1. In n8n, add a Webhook trigger node and copy its production URL.
2. In your GHL workflow, add the Webhook action and paste the URL (method POST).
3. Every contact that hits that workflow step is instantly POSTed to n8n with its fields — n8n takes over from there: enrich, score, sync to another system, anything.

A typical payload arrives under body in n8n, so a contact’s email is {{ $json.body.email }}. Fire a test contact through the GHL workflow once and pin the data in n8n — building against real payloads beats guessing every time.

n8n → GHL

GHL workflows can also start from an Inbound Webhook trigger: GHL gives you a URL, and your n8n HTTP Request node posts to it. Use this when you want the automation logic to live in GHL (SMS sequences, appointment bookings) but the trigger to come from your own systems — a new order, a form on a custom site, a payment event.

Use webhooks when: events should flow between the two systems in real time. This pair is the backbone of most of my client builds.

Method 3 — the full API v2 with a Private Integration Token

This is the method that does everything. GHL’s API v2 covers contacts, conversations, calendars, invoices, opportunities, payments and more — and the simplest way to authenticate server-to-server is a Private Integration Token (PIT).

1. In the sub-account: Settings → Private Integrations → Create. Pick only the scopes you need (e.g. contacts.write, conversations/message.write).
2. In n8n, use an HTTP Request node with two headers on every call:

Authorization: Bearer pit-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
Version: 2021-07-28

That Version header is the step every broken tutorial misses — without it the API rejects the call. A contact upsert looks like this:

POST https://services.leadconnectorhq.com/contacts/upsert
{
  "locationId": "YOUR_LOCATION_ID",
  "email": "jane@example.com",
  "phone": "+447700900123",
  "firstName": "Jane",
  "tags": ["ordered", "source:website"]
}

Use it when: you need anything the native node can’t do, or you want one consistent pattern across all your workflows. Store the token in an n8n credential (Header Auth), never inline, and prefer one PIT per sub-account so you can revoke cleanly.

Bonus — GHL’s MCP server for AI agents

If you’re building AI agents in n8n, GoHighLevel exposes an MCP (Model Context Protocol) server, which lets an n8n AI Agent node call GHL as a tool — look up a contact mid-conversation, book an appointment, add a tag — without you hand-wiring each endpoint. Connect it through the agent’s MCP client tool with your PIT, and treat it as Method 3 with the agent deciding when to call. For deterministic automations, stick to Methods 1–3; MCP shines when an LLM is in the loop.

Three free workflows to import

These are simplified versions of patterns running in my production client systems. Import into n8n (Workflow menu → Import from file), add your PIT and location ID, and they run on any plan including self-hosted:

Which method should you pick?

If you’d rather have this built for you — including the parts that never make it into tutorials, like retries, deduplication and error alerting — that’s what I do. See my GoHighLevel services or, for agencies, white-label fulfilment.