Fix: GoHighLevel "The token does not have access to this location" (403)
The token is valid, so GHL knows who you are. It belongs to a different sub-account, or it's an agency token, or the record lives in another location. Here's how to tell which.
Symptoms — what you're seeing
- The GoHighLevel API returns
403with{"statusCode":403,"message":"The token does not have access to this location"} - It comes from n8n's HTTP Request node, Make, Zapier, a Supabase function, or your own code
- The token is brand new, and other calls with the same token may even work
- The
locationIdyou're sending matches the one in the sub-account's URL - It started after cloning a sub-account, moving a workflow to a new client, or switching from an agency token to a location token
Why this happens
Unlike a 401, this 403 means the token is valid. GHL knows exactly who you are. It's refusing because the token belongs to a different place than the thing you're asking about. Every v2 token is tied to one sub-account (location) or to the agency (company), and each request is checked against that. Four mismatches cause almost every case.
1. The token is from a different sub-account. A Private Integration token belongs to the sub-account it was created in. Create it in Client A's sub-account, send Client B's locationId, and you get this error. It's the most common cause for agencies running one workflow template across many clients.
2. An agency token is calling a location endpoint. Agency-level tokens can manage sub-accounts, but contacts, opportunities, calendars and conversations live inside a location. With OAuth you exchange the agency token for a location token first; an agency token sent straight to /contacts/ is rejected.
3. The record belongs to another location. The locationId can be right while the contactId, calendarId or pipelineId comes from a different sub-account. That's typical after a snapshot or clone: the IDs hard-coded in your workflow still point at the original sub-account.
4. The OAuth app isn't installed on that location. For marketplace or OAuth apps, the token only covers locations where the app is installed. A new client sub-account needs its own install, and with it a new token.
The fix, step by step
- Ask GHL which location the token covers. Call the location endpoint with the token and the
locationIdyou're using:GET https://services.leadconnectorhq.com/locations/{locationId} Authorization: Bearer pit-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx Version: 2021-07-28A 200 means token and location match, so the problem is a record ID (cause 3). A 403 here means the token belongs elsewhere (cause 1, 2 or 4). - Create the token inside the right sub-account. Switch into the client's sub-account first, then go to Settings → Private Integrations → Create new integration. A token created at agency level, or in a different sub-account, won't work for this one. Store one credential per client in n8n, named after the client, so they can't get crossed.
- For OAuth, exchange the agency token for a location token. Request it with the company and location IDs, and use the returned token for every location call:
POST https://services.leadconnectorhq.com/oauth/locationToken Authorization: Bearer <agency_access_token> Version: 2021-07-28 Content-Type: application/x-www-form-urlencoded companyId=<companyId>&locationId=<locationId>
- Replace hard-coded record IDs. Search your workflow for literal
calendarId,pipelineId,pipelineStageIdand custom field IDs. Look them up per location at runtime (for exampleGET /calendars/?locationId=...) or keep a per-client lookup table. Snapshots copy the objects but give them new IDs. - Install the app on the new location. For an OAuth app, install it on each new sub-account, or use an agency-level install with location token exchange. Then save the new tokens.
- Re-test with one read call.
GET /contacts/?locationId=...&limit=1returning 200 confirms token and location line up before you run any writes.
Common variations
- "It works for one client and fails for the next" — the workflow is reusing the first client's token. One sub-account, one token.
- "Creating contacts works, booking appointments fails" — the calendar ID was copied from another sub-account, or from the snapshot source.
- "It broke right after I snapshotted the sub-account" — every object ID changed in the new location. Your workflow still holds the old ones.
- "Make or Zapier shows this error but n8n works" — the Make/Zapier connection was authorised against a different sub-account. Reconnect it while logged into the right one.
- "I'm getting 401 instead" — that's an authentication problem (legacy key, missing Version header), not location access. See the n8n 401 unauthorized on the GHL API fix.
When this isn't the issue
- A 403 with a scope message (not "this location") means the token lacks a permission. Recreate the Private Integration with the missing scope.
- If you're still on a v1 API key, migrate first — see GoHighLevel API v1 deprecated: migrate to v2.
- If the native n8n HighLevel node is giving you this, its credential may be connected to the wrong sub-account — see n8n HighLevel node not working.
Running the same automation across many sub-accounts? The GHL at scale notes cover per-client credentials and ID lookups so one template works everywhere.
Need this workflow customized or debugged? Book a 30-min scope call →
Stuck configuring this workflow?
I build and fix n8n workflows for agencies and SaaS teams. 30-minute scope call — bring the problem, leave with an estimate.