Infor CloudSuite and Tableau integration

What moves between them

Infor CloudSuite data flows into Tableau as published data sources. ml-connector polls CloudSuite on a daily or weekly schedule (configurable), extracts supplier master data, purchase orders, invoices, and GL account balances, and publishes the data as a Tableau extract data source. Tableau dashboards and reports then query that extract. Data moves in one direction: CloudSuite to Tableau. Tableau extracts are refreshed on the same schedule as the CloudSuite polling so dashboards stay in sync with the ERP.

How ml-connector handles it

ml-connector stores CloudSuite OAuth credentials encrypted and proactively refreshes the bearer token before expiry, since token lifetime is tenant-configured and can be as short as 1 hour. It accepts the full CloudSuite base URL per customer, extracted from the .ionapi credentials file, and validates API paths against that instance. ml-connector polls CloudSuite's list endpoints (LstSuppliers, LstBySupplier for invoices, GetHead for purchase order details) on a schedule you define, typically aligned with your month-end reporting cycle or daily reconciliation. It transforms the XML or REST response data into Tableau's expected CSV or columnar format and publishes it as a new data source via Tableau's REST API, authenticating with a personal access token. Tableau's extract refresh is triggered immediately after publishing so dashboards see current data. If a CloudSuite API call returns a 429 rate-limit response, ml-connector backs off and retries within CloudSuite's per-tenant rate limits (typically 100-500 requests per minute). Every record published carries an audit trail so if a downstream dashboard query fails or a data issue is detected, the source record can be traced and replayed.

A real-world example

A mid-market industrial manufacturer runs Infor CloudSuite Industrial for procurement and finance across three plants. Before the integration, the procurement team exported supplier invoices and purchase order status from CloudSuite each morning, manually reformatted the data, and uploaded it to Tableau as a new extract so managers could see cash flow forecasts and supplier aging. The process took 30 minutes daily and broke whenever the export format changed. With CloudSuite and Tableau connected, ml-connector polls CloudSuite every morning before the daily standup, publishes the latest invoice and PO data, and refreshes the Tableau extract automatically. Procurement managers and finance leaders see current supplier metrics and cash projections without the manual step, and the process recovers automatically if a single poll fails.

What you can do

• •Poll Infor CloudSuite suppliers, purchase orders, invoices, and GL accounts on a schedule tied to your reporting cycle.
• •Publish extracted financial and procurement data as Tableau data sources and trigger extract refreshes automatically.
• •Handle CloudSuite's tenant-specific OAuth credentials, token refresh, and per-customer base URL configuration.
• •Bridge CloudSuite's REST API rate limits with retries and backoff so extracts complete reliably.
• •Maintain a full audit trail on every record published so data discrepancies can be traced back to the source.

Questions

Which direction does data move between Infor CloudSuite and Tableau?

Data flows from Infor CloudSuite into Tableau only. ml-connector polls CloudSuite for suppliers, invoices, purchase orders, and GL account data, then publishes that data as Tableau extracts. Tableau is read-only for reporting and dashboards, so no data flows back into CloudSuite.

How does ml-connector handle CloudSuite's lack of webhooks and token expiry?

CloudSuite publishes no webhooks, so ml-connector polls REST endpoints on a schedule you control, typically daily or weekly. ml-connector also tracks CloudSuite's OAuth token lifetime (configurable 1-24 hours per tenant) and refreshes the token proactively before expiry so polling never stalls waiting for a 401 challenge.

What happens if a CloudSuite API call fails or Tableau extract refresh is slow?

ml-connector backs off and retries on CloudSuite rate-limit responses (HTTP 429) within the typical 100-500 requests-per-minute limit per tenant. Every record extracted carries an audit trail with source details and timestamps, so failed or slow refreshes can be traced and re-run if needed.