Workday Financial Management and Snowflake integration

What moves between them

The main flow is Workday Financial Management into Snowflake. Suppliers, invoices, purchase orders, payments, GL accounts, and journal entries are read from Workday on a schedule you define, typically every 30 to 60 minutes for transactions and daily for reference data. ml-connector writes these records into Snowflake tables, preserving source IDs and timestamps for lineage and reconciliation. Invoice status updates and payment allocations flow the same direction, giving your analytics team a complete history of every transaction from creation through posting.

How ml-connector handles it

ml-connector stores the Workday tenant hostname, ISU credentials or OAuth2 refresh token, and Snowflake account ID and key pair encrypted. It polls Workday Financial Management on a schedule tied to your close calendar, requesting suppliers and invoices first to establish reference dimensions, then requesting GL accounts and journal entries. Since Workday has no delta endpoint, each polling interval requests the full set but filters by date range to find new and changed records. Snowflake credentials are refreshed on every session; the JWT is regenerated when nearing expiry. ml-connector handles Snowflake rate limits by backing off and retrying, uses timestamp watermarks to skip records already loaded, and logs the source record ID in every Snowflake row so failed GL postings can be replayed from the source.

A real-world example

A mid-market financial services firm uses Workday Financial Management to manage vendor invoices, purchase orders, and cash flow across three regional offices. Previously, the finance team exported invoice and payment reports from Workday daily and loaded them into spreadsheets for reporting and reconciliation, a manual process that took two hours and was error-prone. The firm also needed to reconcile Workday GL accounts against external auditor requests, requiring hand-written lookups. By connecting Workday to Snowflake, every invoice, purchase order, payment, and GL posting flows into Snowflake tables automatically on a three-times-daily schedule. The finance team now builds BI reports directly in their analytics tool, reconciles GL accounts in seconds via SQL, and the daily export step is eliminated.

What you can do

• •Sync suppliers, invoices, purchase orders, payments, and GL accounts from Workday Financial Management into Snowflake on a schedule you control.
• •Preserve source record IDs and timestamps in every row so failed GL postings can be replayed from the source without re-keying.
• •Handle Workday OAuth2 refresh tokens or Integration System User credentials, and Snowflake key pair authentication with JWT regeneration.
• •Filter Workday records by date range to reduce polling overhead and load only new and changed data since the last sync.
• •Resolve Snowflake rate limits and network policies so connector egress IPs are registered and retries succeed on the first attempt.

Questions

Which direction does data move between Workday Financial Management and Snowflake?

The main flow is Workday Financial Management into Snowflake. Suppliers, invoices, purchase orders, payments, GL accounts, and journal entries flow from Workday into Snowflake tables for reporting and reconciliation. Snowflake is a data warehouse, not an ERP, so financial transactions are not written back to Workday.

How does ml-connector handle Workday's lack of webhooks and requirement for polling?

ml-connector polls Workday Financial Management on a schedule you define, typically every 30 to 60 minutes for transactional records. It requests records by date range to find new and changed data, then loads only the new records into Snowflake. Since Workday has no delta API, ml-connector tracks the timestamp of each load and requests only the date range since the last sync.

What authentication is required for Workday Financial Management and Snowflake?

Workday Financial Management requires either an Integration System User account with WS-Security credentials or an OAuth2 refresh token generated in your Workday tenant. Snowflake requires a key pair (RSA private key and public certificate) or a programmatic access token. ml-connector stores both credential sets encrypted and refreshes Snowflake's JWT on every session or when expiry approaches.