Our previous web developer, who originally set up and managed our n8n account, has now left the business. Unfortunately, we do not currently have access to the workspace or administrative account associated with it.We believe this instance is being used to manage important workflows related to our website, including contact forms and automated emails. As a result, regaining access is critical for maintaining our operations.We would like to request assistance with:
Regaining access to the existing workspace, or
Transferring ownership/admin rights to a new company email address
If this is n8n Cloud, you’ll need to contact n8n support with your workspace URL, proof of ownership, and the new admin email for a transfer.
If it’s self-hosted, access will need to be recovered via your server/hosting provider or existing deployment (e.g., VPS, Docker, AWS).
Once access is restored, consider adding multiple admin users to avoid this situation in future.
Hi @SelbraeH Welcome!
Please read this for how to read out for support:
Also for self hosted instance, contact the person who set that up and request credentials if you dont have like user email and password as those are the only one you require to login.
Beyond recovering access, I’d also treat this as a governance gap.
For any business-critical n8n setup, I’d strongly recommend having company-owned admin accounts, at least two internal admins, and documented ownership for workflows, credentials, and integrations. That way, when a developer or vendor leaves, access to critical automations doesn’t leave with them. In cases like this, the access issue is usually just the symptom , the bigger risk is lack of operational continuity.
If you’ve lost admin access to n8n, there’s no direct way to recover it from inside the app. You’ll need to contact n8n support and provide proof of ownership (billing details, company email, workspace info, etc.) to request access or transfer ownership.
Ran into something similar on my end a few weeks back and ended up stepping through it the same way you’re describing. The part that tripped me up initially was how sensitive the defaults were to the environment — once I adjusted for that the rest fell into place. Curious what you landed on in terms of the testing loop, because iteration speed seemed to matter more than the specific choice of tool. Happy to trade notes if anyone has a cleaner workflow.