Power BIInfor SyteLine (CloudSuite Industrial) Agent view

How to connect Power BI to Infor SyteLine

Short answer

For on-prem SyteLine, connect Power BI to a read-only SQL Server replica or reporting database through an on-premises data gateway, and model the key tables into a star schema. For cloud tenants, use Infor's data export or API options. Add natural-language Q&A on top once the semantic model is clean.

Get this built on your Infor SyteLine (CloudSuite Industrial)

Power BI is the most common reporting tool bolted onto SyteLine because many plants already license Microsoft 365 and run SQL Server. The question is rarely whether it can be done; it is how to do it without hammering the production database or rebuilding the same business logic in every report.

SyteLine stores data in multi-site tables with the _mst suffix and many coded fields. Joining them naively produces wrong totals, especially when site_ref is missed or when currency and unit-of-measure conversions are ignored. A small, well-designed semantic model pays for itself quickly.

Once the model is solid, natural-language question answering becomes realistic, either through Power BI's own Q&A and Copilot features where your licensing allows, or through an external assistant that queries the same model. Both depend on good names and relationships more than on the AI itself.

What teams use it for

On-time delivery and backlog dashboard

Order lines with promise and ship dates give on-time-in-full by customer, product line, and plant, with backlog aging for the sales manager.

Shop floor performance

Job transactions and labor reporting feed schedule adherence, cycle time versus standard, and scrap by work center.

Inventory health

On-hand, turns, slow movers, and excess versus demand by warehouse, with drill-through to item and lot.

Purchasing and supplier scorecards

Receipts versus PO due dates, price variance, and rejection rates per vendor for supplier reviews.

Finance operating view

Margin by customer and item, WIP balances, and open AR/AP aging joined with the operational data.

Infor SyteLine (CloudSuite Industrial) integration surface

SQL Server connectivity: for on-prem SyteLine, Power BI can connect to the application database or, preferably, a read-only replica or reporting copy (log shipping, replica, or scheduled ETL). Core tables follow the _mst pattern with site_ref, for example co_mst, coitem_mst, job_mst, jobmatl_mst, po_mst, poitem_mst, item_mst, matltran_mst.
Views and reporting objects: SyteLine ships with database views and reporting-oriented objects in some versions; check what your version provides before writing your own joins, and prefer views over raw tables.
IDO and ION API: usable for smaller extracts, but row-by-row paging is slow for large history. Better for dimension data or near-real-time slices than bulk loads.
Infor OS data options: cloud tenants may have Data Lake or other Infor data services and export features. Availability and licensing vary, so confirm with Infor what your tenant includes.
On-premises data gateway: required for the Power BI service to refresh from a SQL Server inside your network; install it on a server with reliable access to the replica.

How it works

1
Create a reporting copy of the data
Set up a read-only replica or nightly copy so dashboards never compete with transaction processing. Decide the refresh frequency per subject area.
2
Define a semantic model
Build fact tables for order lines, job operations, receipts, and inventory transactions and shared dimensions for item, customer, vendor, site, and date. Keep site_ref in every key.
3
Handle conversions once
Apply currency conversion, unit of measure, and fiscal calendar logic in views or the model, not in each report.
4
Install the gateway and schedule refresh
Install the on-premises data gateway, register the data source, and set refresh schedules. Use incremental refresh for large transaction tables.
5
Build the first three dashboards
Start with on-time delivery, inventory, and purchasing. Validate each number against a SyteLine report with the owner of that process.
6
Add row-level security
Restrict by site, customer group, or department using roles in the model, matched to how people use SyteLine.
7
Add natural-language Q&A
Rename fields to business terms, add synonyms, and then enable Q&A or an external assistant over the model, testing with real questions.

Example flows

Nightly refresh into the on-time delivery report

Trigger: Scheduled refresh at 2 a.m.

The gateway connects to the reporting replica.
Incremental refresh loads order lines and shipments changed since the last run.
The model recalculates on-time-in-full by customer and week.
The sales manager opens the dashboard at 8 a.m. with yesterday's data.

Result: A consistent daily view without anyone exporting from SyteLine.

Asking a question of the model

Trigger: A planner types, which suppliers were late more than three times this quarter.

Q&A or the external assistant maps terms to the supplier and receipt tables in the model.
It generates a visual and shows the underlying filter.
The planner opens drill-through to the late PO lines.

Result: Ad hoc answers without waiting for a report developer, provided the model is well named.

Keeping data on-prem

Power BI Desktop can read from an on-prem SQL Server directly, and the Power BI service reaches it through the on-premises data gateway, so you do not have to move SyteLine into the cloud. If policy forbids even that, Power BI Report Server is a separate on-prem option, with a different feature set from the cloud service, so confirm what you need is included. Natural-language questions on-prem are best handled by an assistant that queries the same read-only replica or semantic model with an open-weight model hosted inside your network, so questions about customers and pricing never reach a public AI service. Keep the replica, gateway, and model on a segmented network with logged access.

Security checklist

Use a read-only database login scoped to the reporting views, not the SyteLine application login.
Apply row-level security by site and customer group, and test it with real user accounts.
Restrict cost, margin, and pricing columns to roles that need them.
Secure and monitor the gateway server, rotate credentials, and keep it patched.
Check where Power BI datasets are stored and who can export them, since exports bypass SyteLine permissions.

Ways to get it done

Direct SQL views plus Power BI in-house

Effort: A few weeks for three dashboards, longer for a full model.

Works for teams with a SQL developer who knows SyteLine tables and wants quick results.

Infor reporting and analytics options

Effort: Varies with licensing and skills.

Worth evaluating if your licensing includes Infor's data or BI services, especially on CloudSuite tenants.

Warehouse plus Power BI, with Netray adding natural-language Q&A

Effort: Phased over several weeks to months.

When you want governed data across several systems and private conversational access on top.

Pitfalls

!Querying production directly with heavy reports, which slows order entry and shipping.
!Forgetting site_ref in joins, producing duplicated or mixed-plant totals.
!Rebuilding business logic differently in each report, so two dashboards disagree.
!Ignoring custom user fields and local codes that finance and operations rely on.
!Turning on Q&A over an unclean model; confusing column names produce confident wrong answers.

Frequently asked questions

Can Power BI connect directly to SyteLine?

For on-prem installs, yes, through its SQL Server database, ideally a read-only replica, plus the on-premises data gateway for the cloud service. You can also use IDO or ION APIs for smaller extracts. Cloud tenants rely more on Infor's export and data service options.

Which tables do I need for an on-time delivery report?

Typically customer order and order line tables (co_mst, coitem_mst), shipment records, and the customer and item masters, joined with site_ref. Names and columns can vary by version and customization, so confirm in your own schema or with the views your version provides.

Do I need a data warehouse?

Not always. A replica with good views is enough for one site and a few dashboards. A warehouse becomes worthwhile when you combine several systems, need history beyond what SyteLine keeps, or have heavy refresh loads.

Can I ask questions of the data in plain English?

Yes, either with Power BI's own Q&A or Copilot features if your licensing includes them, or with an external assistant on the same model. Results depend heavily on clear field names, synonyms, and relationships, so invest in the model first.

Will this affect SyteLine performance?

It can if reports hit the production database. Use a replica or reporting copy, schedule refreshes outside peak shifts, and use incremental refresh. Test refresh duration and load on the source before rollout.