Gadget is shutting down: moving Shopify-to-ERP integrations

Gadget has announced that it is winding down. A lot of teams built real businesses on it, and the team there has given everyone a clear schedule to plan around. This post is for one group of those teams: the ones whose Shopify store is connected to an ERP or accounting system by something that runs on Gadget.

The dates

These come from Gadget’s announcement, which is the source to rely on:

Now: new applications can no longer be started.
End of November 2026: the last day for a first production deploy.
End of 2026: support ends. After that the platform is kept running and critical issues are handled.
31 March 2027: the platform shuts down.

Existing apps keep running as normal until then. Gadget has said it will publish migration guides and tools, so watch for those before you decide anything.

Is this post for you?

Runtara automates operations work between business systems. That covers part of what people built on Gadget, not all of it.

A good fit:

Shopify orders that have to become sales orders in an ERP or accounting system.
Products, prices and inventory kept consistent between the ERP and the store.
Fulfilment, tags or order status written back to Shopify from the ERP.
Anything above where some records need a person to check them before they are written.

Not a fit:

Public Shopify apps listed on the App Store.
Storefront and admin extensions, or any app with its own customer-facing interface.
SaaS products built on Gadget as their backend.

If you are in the second list, Gadget’s own migration guides will serve you better than we can.

What to inventory before you move

Whatever you move to, write these down first. Most of the time in a migration goes into rediscovering what the old integration quietly did.

What starts each job. Shopify webhooks, schedules, a button someone presses.
What data lives in Gadget. Which tables are a copy of Shopify or the ERP, and which hold something that exists nowhere else, such as mappings, overrides or a history.
Which systems are written to. Every write to Shopify, the ERP or anything else, and what it changes.
The rules in the code. How a SKU is matched to an item, which price wins, what happens to an order with an unknown product.
What happens on failure. Retries, alerts, and the cases someone fixes by hand today.
Credentials and access. API keys and tokens, who owns them, and what they are allowed to do.

Item 2 matters most. Export anything that exists only in Gadget well before the shutdown date.

How the same work runs on Runtara

Each process becomes a workflow. An order from Shopify starts a run; the customer, items, quantities and prices are checked against your ERP; orders that match go straight in, and only the exceptions you define wait for a person, with the order and the ERP data side by side. Product data works the same way in the other direction: each field follows the rule you agreed for it, and unclear differences are held for review.

The Shopify connection covers products, variants, prices, metafields, inventory and locations, orders, fulfilments, draft orders and customers. On the ERP side, our team has integrated Business Central, Odoo, QuickBooks, Epicor Prophet 21, Epicor Eclipse and Infor CSD and SX.e.

You keep control of where it runs

Moving off one platform is a good moment to ask how hard the next move would be. Our answer:

The engine is open source. The runtime that executes Runtara workflows is published under the AGPL-3.0 license on GitHub. You can read how runs are executed and recorded, and run it on your own infrastructure.
Or use our Cloud. Runtara-managed Cloud runs in a US or EU region of your choice, and your data stays in that region.
Your database or ours. Business records can live in the database Runtara hosts for your instance, or in your own PostgreSQL database through a connection. Either way they are ordinary tables your team can query and export.
Your model provider. AI steps call the provider you choose, with your credentials, billed to your account.

The security model describes each deployment option in detail.

How a migration goes

Pick one process. Usually the one that hurts most if it stops, such as orders into the ERP.
Run a paid pilot of that process on Cloud. Our team builds and runs the workflow, live in two weeks once access is in place.
Run both side by side. Compare what the new workflow would write with what the existing integration does write, before anything is switched.
Switch, then move the next process.

Pricing covers the plans and how a pilot is quoted.

A timeline that leaves room

Working back from 31 March 2027:

By the end of December: the inventory above is written, and data that exists only in Gadget is exported.
January: the first process is running in parallel.
February: the remaining processes are moved, one at a time.
March: kept free for whatever the inventory missed.

If your Shopify-to-ERP integration is on Gadget and you would like a second pair of eyes on the inventory, book a call and bring the list.