Relay.app Is Shutting Down: Deadlines and Next Steps

Relay.app shutdown is happening now: free workspaces lose access Aug. 15, 2026 and paid workspaces remain available until Sept. 14, 2026, so export your workflows, tables, and run history before those dates and plan to re-authorize any connected apps after you rebuild automations locally or on another service [S1].

Credit: Photo by konkapo on Pixabay

Why This Matters

The shutdown breaks running automations and removes stored data and credentials unless you act during the wind-down windows. That means business processes that quietly relied on Relay for reporting, notifications, or task automation can stop without another look; the deadline turns a soft migration task into a calendar item you cannot ignore.

How the Shutdown Timeline Works

The timeline is simple and timeboxed: free accounts lost access on Aug. 15, 2026, and paid accounts retain access through Sept. 14, 2026, after which workspaces and associated data are deleted [S1].

What that looks like in practice: during each account’s wind-down window your workflows keep running, export tools remain available, and new signups or plan upgrades are turned off. Use the remaining days to extract everything you need and to test rebuilds on a new platform. Treat the dates as strict operational cutoffs for data retention and credentials.

Free vs Paid: Deadlines and What They Mean

Free and paid workspaces have different windows for taking action: free users had a shorter window that already closed on Aug. 15, 2026, while paid customers have a longer window that ends on Sept. 14, 2026 [S1].

Operationally, that means paid customers have more breathing room to export large workspaces, to request help from vendors, or to negotiate pro-rated refunds if you have an annual plan. Free users need to prioritize exports first because they had fewer days to complete the work. If you still have access, start with exports of tables and workflow definitions; if you no longer have access, check any email Relay sent for links to a downloadable export or support options.

What You Can Export

Your main assets to extract are workflow definitions (the sequences, triggers, actions), tables and data used by automations, run history and logs you need for audits, and any agent or AI prompt templates you rely on.

• Workflows: export the workflow JSON/YAML or the provided bundle so you can read the trigger/action structure.

• Tables and stored data: download CSV or database exports for any tables that feed automations.

• Run history and logs: export runs you need for debugging, billing reconciliation, or compliance.

• Agent prompts and prompt templates: copy any custom prompts or templates you use for text generation.

Note the practical limits: some export tools generate archives that can take time to prepare and the download links may be available only for a short window after generation. Start exports early and verify the files open correctly before you delete anything from the old workspace.

Credentials, Tokens, and Connected Apps

Relay has stated that connected-app credentials and tokens are deleted when the account is deleted. That means you should expect to re-authorize every integration when you move to a different service, and treat any saved API keys or OAuth tokens in Relay as ephemeral backups you cannot rely on after the wind-down.

Practically, that means these steps:

• Inventory connected apps: list every external service your automations touch (email providers, CRMs, cloud storage, APIs).

• Note auth method: mark which use OAuth, which use API keys, and which rely on service accounts.

• Rotate credentials where appropriate: if a token was created inside Relay and cannot be exported, plan to revoke and recreate a token on the target platform to avoid a security gap.

One important operational note: because Relay will delete tokens at account deletion, you cannot count on those tokens surviving the shutdown. If you have tokens stored only in Relay and nowhere else, plan for immediate re-authentication on the replacement platform.

Migration Checklist

Migration is a project, but it does not have to be heroic. Treat it as a short four-stage process: inventory, export, rebuild, validate.

• Inventory (1-2 days): list workflows, tables, connected apps, schedules, and run histories you need.

• Export (1-3 days): use Relay’s export tools for workflows, CSV exports for tables, and download run logs. Verify files open and include the fields you expect.

• Choose replacements and rebuild (2-7 days): pick an alternative automation platform or a combination of lightweight scripts and cron jobs. Recreate triggers and actions, re-authorize integrations, and migrate any prompt templates into the new environment.

• Validate and cut over (1-3 days): run test incidents, compare run outputs to historical logs, and schedule the cutover so you don’t lose data during the switch.

When choosing replacements, balance how much you want to re-architect versus lift-and-shift. If you need to stay simple, look for tools that accept JSON workflow imports or that provide a straightforward API for recreating automations from a script. If you want more control, rebuild key automations as small services or serverless functions that you manage directly.

Common Mistakes to Avoid

People rush the export but forget validation, or they export only the visible parts of a workflow and miss the tables or secrets that make it work. Another frequent error is assuming tokens will survive the transition and delaying re-authorization until after the cutover.

• Don’t treat export as a backup obligation alone. Open the exported files and run a dry rebuild on the new platform before Relay access ends.

• Don’t ignore run history. It helps you understand edge cases and timing requirements you will need to preserve.

• Avoid single-point dependency on a Relay-created token. If a credential cannot be reissued outside Relay, plan to create a service account or an alternative auth method on the target.

Frequently Asked Questions

When is Relay.app shutting down?

Free workspaces were closed on Aug. 15, 2026, and paid workspaces remain available through Sept. 14, 2026; those are the dates to calendar for exports and re-authentication tasks [S1].

What happens to Relay.app data after shutdown?

After a workspace’s wind-down window ends, Relay deletes the workspace and associated data. That includes stored tables, run history, and related workspace artifacts. Because deletions include connected-app tokens tied to the workspace, you cannot rely on Relay to hold credentials for you after the cutoff.

How do I export workflows from Relay.app?

Use the workspace export tool to download workflow bundles, CSV exports for tables, and run logs. Generate exports early; some export jobs can take time to complete, and you should verify the exported archives open and contain the fields your downstream systems expect.

What are the best Relay.app alternatives after the shutdown?

The best replacement depends on what you automate: if you need a no-code editor and many connectors pick an established integration platform; if you value control and repeatability consider building serverless functions or small services that call APIs directly. Evaluate replacements on connector availability, ability to reauthorize integrations, logging, and export/import capabilities. If you need to compare roles and companies in the AI automation space while you pick a platform, explore meta data engineer salary at Meta and a recent guide on how agentic AI is changing hiring workflows to understand demand for these skills and where teams are hiring.

Do Relay.app paying customers get refunds?

Relay’s wind-down messaging mentioned pro-rated refunds for annual plans and extra credits to ease migration, but check your billing email or the account dashboard for the details that apply to your workspace. If you do not see a refund posted, reach out to the support contact Relay provided before your window closes.

Final Thoughts

Most teams treat automation tooling like plumbing until it stops working; the Relay.app shutdown makes that invisible plumbing visible fast. The single action that changes outcomes is this: inventory and export everything you depend on now, then run a test rebuild on your chosen replacement. It will be work, and it will feel tedious, but a small, validated migration beats a frantic recovery after the cutoff. Take the calendar note seriously and plan a short, concrete migration sprint this week.

Explore AI automation companies and related roles