Data Store

A data store is a lightweight built-in database that lives inside an automation platform, letting a workflow save and read data between runs without connecting an external app like Airtable or Postgres. Make offers "Data stores"; other tools have similar key-value or table storage. Why it matters: automations are often stateless — each run starts fresh with no memory of the last one. A data store gives them memory. Typical uses: deduplication (record which order IDs you've already processed so you never send a confirmation twice), counters, caching an API token until it expires, or holding a small lookup table. Practical note: a data store is not a substitute for a real production database — capacity is limited, querying is basic, and it's tied to your automation account. Use it for automation-local state (dedupe keys, flags, cursors), not as the system of record for customer or order data. Always design idempotent flows so a re-run reads the store and skips work it already did.