We’re experiencing significant data latency issues with our Power BI dashboards that pull data from Oracle Transactional Business Intelligence (OTBI). When we connect Power BI to OTBI using OData feeds, the data in our dashboards consistently lags by approximately 24 hours compared to what we see directly in OTBI reports.
For example, transactions posted today in Oracle Fusion Cloud appear in OTBI reports within a few hours, but the same data doesn’t show up in Power BI until the next day. This is causing problems for our executive dashboards that need near real-time visibility into financial metrics.
We’ve checked the Power BI refresh schedule and it’s set to refresh every 4 hours during business hours. The OTBI data refresh timing appears to be working correctly when we run reports directly in Fusion. I’m wondering if there’s a batch schedule or ETL process between OTBI and the OData service layer that we’re not aware of. Has anyone successfully achieved near real-time integration between OTBI and Power BI?
Let me break down the timing issues between OTBI and Power BI that cause this 24-hour data lag.
For Power BI dashboards showing real-time data, you need to optimize all three components:
- Configure incremental BI warehouse ETL jobs to run every 4 hours instead of daily
- Adjust the OData service cache refresh to 15 minutes (minimum supported)
- Set Power BI refresh schedule to align with ETL completion times
Here’s what happens with your current setup: Transaction data flows from Fusion transactional tables to the BI warehouse via ETL jobs. If these run once daily at 2 AM, any transactions after that won’t appear until the next ETL run. The OData feed that Power BI consumes reads from this warehouse, adding another 15-minute cache delay.
To fix this, modify your ETL schedule in Scheduled Processes. Search for incremental ETL jobs for your subject areas (AP, AR, GL). Change the schedule from daily to every 4 hours during business hours. This balances freshness with system performance. Then align your Power BI refresh to run 30 minutes after each ETL completion to allow for cache updates.
For truly real-time requirements (under 1 hour latency), consider using Fusion REST APIs directly instead of OTBI. The REST API queries transactional tables directly, bypassing the warehouse layer entirely. You can build custom Power BI connectors using REST endpoints for Invoice, Payment, and GL Balance APIs.
The OTBI data refresh timing works well for standard reporting, but the warehouse architecture introduces inherent latency. By optimizing the ETL batch schedule and understanding the OData integration with Power BI’s caching layers, you can reduce lag from 24 hours to 4-5 hours, which meets most dashboard requirements without the complexity of direct REST API integration.
The daily ETL schedule is the default configuration, but you can modify it. Navigate to Tools > Scheduled Processes and search for “Incremental” ETL jobs. These can be scheduled more frequently than full loads. For subject areas like GL Balances or AP Invoices, you can run incremental loads every 4-6 hours without significant performance impact. Just be aware that more frequent ETLs consume more resources during business hours.
The 24-hour lag you’re seeing is likely related to how OTBI’s underlying subject areas refresh their data warehouse tables. OTBI doesn’t query transactional tables directly - it uses a data warehouse layer that’s refreshed on a schedule. Check your BI Analytics Warehouse ETL jobs in Scheduled Processes to see when they’re running.