Keep the Cognito user pool, verify the same tokens in the new API, move one screen at a time behind a flag, and keep AppSync alive until old mobile app versions are gone. The order that avoids a big-bang release, and the checks that tell you it is safe to switch off the old backend.
The risky way to leave AppSync is a release that switches everything at once. The safe way takes longer and is boring on purpose: both backends run, each screen moves when it is ready, and the old one is switched off when nothing calls it.
Book a free 30-minute call. You'll leave knowing what we'd do first and roughly what it costs, whether or not you hire us.
Auth first. The new API validates the Cognito JWTs the app already has: signature against the user pool's keys, issuer, audience or client id, and expiry. Users notice nothing.
Data next, kept in sync from DynamoDB Streams, so either backend can serve reads.
Reads, screen by screen, behind a remotely controlled flag. Compare responses from both backends in staging, and in production for a sample.
Writes, one entity at a time. When an entity's writes move, publish real-time events from the new API.
Background work last: triggers, scheduled functions and anything that listens to the old tables.
Web clients update on the next page load. Mobile apps don't. Some users will run a version that calls AppSync for months. Ship the flag-driven client early, add a minimum supported version check, and keep AppSync running, with its data in sync, until that version is old enough to force an update.
AppSync request metrics have been at zero, or at a known trickle you have decided to drop, for two weeks.
Entity counts and checksums match between old and new stores.
A final export of every Amplify table is in S3 with a retention rule.
The rollback plan is written down, even if you won't need it.
Once a month: what changed in tech that month and what it means for you, from cloud and AI to security and the tools you build with. Every item links to its source. No sales pitch, unsubscribe in one click.
Book a free 30-minute call about it. Bring your setup and your bill; you'll leave knowing what we'd do first and roughly what it costs, whether or not you hire us.
Or keep reading
Leaving AWS Amplify and AppSync: When It Is Time, and How to Do It Without a Rewrite
Amplify gets a startup to launch fast. The trouble starts later: list queries that scan, resolvers that can't throw or call out, a 30-second ceiling, and a data model that was never designed. The signs it is time to leave, what to move to, what to keep, and a cutover that doesn't log anyone out.
Read
Migrating Amplify's DynamoDB Data to Postgres: Why Lists Scan, Why Pages Come Back Empty, and How to Move It
Amplify list queries use DynamoDB scans, and DynamoDB applies filters after reading up to 1 MB, so a filtered page costs a full read and can return almost nothing. When the data is relational, Postgres is the better home. How to export, transform, keep the two in sync and switch.
Read
As of November 2, 2026. Provided as is, for general information, and accurate to the best of our knowledge on the date shown. Not legal, compliance, tax, financial or other professional advice. Prices, limits and vendor services change; check current vendor documentation before relying on anything here. Terms of use.