How to connect Power BI to IFS Cloud
Short answer
Connect Power BI to IFS Cloud through OData projections for small, current datasets, or through a replicated reporting database or warehouse for volume and history. Use the on-premises data gateway when the source is inside your network, and add natural language Q&A after the model is stable.
Power BI is the usual reporting layer next to IFS Cloud, but the first attempt often goes wrong. Someone points Power BI at a wide projection, refresh takes an hour, numbers disagree with Aurena, and trust erodes. The cure is a deliberate data path and a small, well-defined model.
IFS also has its own analytics and reporting capabilities, and for operational lists and standard reports they may be enough. Power BI earns its place when you need to combine IFS with other sources, share dashboards with people who do not use IFS, or build custom visuals for management.
This page covers the connection options, the dashboards that matter in manufacturing and asset-heavy operations, and how to add natural language questions on top without breaking governance.
What teams use it for
Delivery performance and OTIF
Compare promised, planned and actual ship dates from customer orders and shipments, sliced by customer, site and part family.
Inventory and supply risk
Show stock, coverage, open supply and demand by site, with aging and slow movers highlighted.
Shop floor performance
Report shop order progress, schedule adherence and scrap or rework trends by work center.
Maintenance KPIs
Track backlog, planned versus corrective ratio, mean time between failures by asset class and spare parts consumption.
Project cost control
Show budget, committed and actual by project element, with forecast to complete and drill-down to transactions.
IFS Cloud integration surface
- OData v4 projections can be consumed by the Power BI OData feed connector or Power Query. They are best for modest volumes, because paging over large entity sets is slow and loads the application tier.
- Information Sources and related reporting features in IFS expose pre-shaped datasets that may be reusable by external tools, depending on release and configuration. Confirm with your IFS team what is supported.
- The underlying Oracle database holds the data, and some estates replicate it to a reporting database, or to a warehouse, using their own tools. Direct production queries from BI are discouraged and may be restricted by agreement for cloud-hosted editions.
- Business events and scheduled extracts can feed a data lake or warehouse incrementally, which is the more scalable pattern for history and large transaction tables.
- IFS IAM issues the tokens that a gateway or dataflow uses to call projections. Use a dedicated service client with a read-only permission set.
How it works
- 1
Define the dashboards before the data
List five decisions the dashboards must support and the measures behind them, including exact definitions of OTIF, backlog and cost categories.
- 2
Pick the data path per subject area
Use projections for small reference and summary data. Use a replicated or warehouse layer for orders, inventory transactions and cost history.
- 3
Stage and transform
Load into a star schema with conformed dimensions for part, site, customer, supplier, asset and project. Do business logic once, in the warehouse or dataflows.
- 4
Install the gateway if the source is private
Set up the on-premises data gateway on a server that can reach the reporting database or the IFS endpoint, with a service account and clustered gateways for resilience.
- 5
Build the semantic model
Create relationships, measures in DAX and incremental refresh on large tables. Apply row-level security that mirrors site or company access.
- 6
Reconcile against IFS
Match totals to Aurena reports for several periods before anyone uses the dashboards. Document known differences.
- 7
Add natural language Q&A
Once the model is clean, enable Power BI Q&A or Copilot features if licensed, and curate synonyms. For private data, add a self-hosted question layer over the same warehouse.
Example flows
Nightly incremental load for orders
Trigger: Scheduled job at 02:00.
- Extract orders and lines changed since the last watermark.
- Land them in the warehouse staging tables.
- Merge into fact tables and refresh dimensions.
- Trigger the Power BI dataset refresh through the gateway.
Result: Dashboards are current each morning without heavy live queries on IFS.
Plain-language question on delivery performance
Trigger: A sales manager asks which customers had OTIF below target last quarter.
- Q&A or a private question layer maps the words to the OTIF measure.
- It queries the semantic model or warehouse with the user's row-level scope.
- It returns a table and chart with the definition shown.
Result: Quick answers without building a new report each time.
Keeping data on-prem
If IFS Cloud is customer-managed, place a replicated reporting database or warehouse inside your network and use the on-premises data gateway to let the Power BI service refresh from it, or use Power BI Report Server if reports must remain fully on-prem. For natural language questions with no cloud dependency, run an open-weight model locally that translates questions into queries over the warehouse, with the schema described to it and read-only credentials. That keeps both data and prompts inside the boundary. Cloud-hosted IFS editions may restrict direct database access, so confirm what the service terms allow.
Security checklist
- Use a read-only service identity for extracts and keep credentials in a vault, not in Power BI files.
- Replicate and expose only the tables and columns needed, and mask sensitive fields such as pricing if the audience should not see them.
- Implement row-level security that mirrors IFS company and site access, and test it with real accounts.
- Control workspace sharing and export permissions in Power BI, since exports leave your governance.
- For natural language layers, restrict generated SQL to read-only views and log each question.
Ways to get it done
Power BI directly on OData projections
Effort: Days, with refresh limits.
Small departmental reports and prototypes.
Warehouse or replicated reporting database with Power BI on top
Effort: Weeks to months depending on scope.
Cross-module reporting, history and larger volumes.
Add Netray for private natural language Q&A over the warehouse
Effort: Add-on once the model is stable.
Teams that want to ask questions in plain language without sending data to a hosted service.
Pitfalls
- !Pulling large transaction projections straight into Import mode and timing out refreshes.
- !Mixing definitions, so OTIF in Power BI differs from the one in Aurena.
- !Querying the production database directly during working hours.
- !Skipping row-level security because the pilot had a small audience.
- !Building Q&A on an unclean model and then blaming the AI for wrong answers.
Frequently asked questions
Can Power BI connect straight to IFS Cloud?
Yes for modest data, using the OData feed against projections with a service identity. For large histories it is slow and loads the application, so most teams move heavier subject areas to a replicated database or warehouse and refresh incrementally.
Do we need the on-premises data gateway?
If the source is inside your private network and not reachable by the Power BI service, yes. It lets scheduled refresh reach your database or endpoint without opening inbound ports. Run it on a stable server and consider a gateway cluster.
Why do Power BI numbers differ from IFS?
Usually because of different definitions, time zones, cancelled or closed records, currency conversion or timing of the extract. Reconcile a few periods against Aurena reports and document every rule in the model before launch.
Can we ask questions in plain language?
Yes. Power BI Q&A and Copilot features work on a well-built model if licensed and enabled. For private or on-prem data, a self-hosted model can translate questions into read-only queries over the same warehouse with logging.
Should we use IFS's own analytics instead?
If your needs are operational lists and standard IFS reports, the built-in tools may be enough and simpler. Choose Power BI when you need to blend other sources, share widely, or build management dashboards beyond what IFS offers.
Want Power BI working on your IFS Cloud?
A free 30 minute assessment: we map the integration path for your version and setup, and where it should run.
Get a free assessment