Live SyteLine Dashboards Without a BI Project
The standard path to dashboards over SyteLine runs through a BI project: stand up a warehouse, build ETL, model the data, license a BI tool, train users, and somewhere around month six, look at a chart. The result is often good, and the journey kills most attempts. Meanwhile the questions the dashboards were meant to answer, what is open, what is late, where does the month stand, get answered the old way: someone exports to Excel every morning. SyteRay takes a different path: it ships 13 curated dashboard views that read the live database directly, through the same semantic layer as its chat answers. This post covers what those views are, the interaction model, and the design decision that makes the numbers trustworthy.
Thirteen Views That Cover the Daily Questions
The shipped set covers the operational core of a manufacturing business: Sales Overview and an Open Orders Board for the order book; AR/Collections and AP/Payables for finance; Inventory Health and Item Watch for materials; Production/Jobs for the shop; Purchasing; Shipping; and Quality. On top of those sit three cross-module views: an Executive Pulse of headline KPIs, a My Day view, and a Site Comparison for multi-site operations.
These are curated, not generated: each view was designed around the questions its audience actually asks daily, which is why a plant manager's 7am scan and a controller's collections review both land on a screen that already has their numbers arranged sensibly.
Live Data, Working Interactions
Every dashboard renders fresh from the database when opened, stamped with an as-of time and a refresh control, so there is no overnight-batch ambiguity about what you are looking at. Tables sort by clicking a header. Ranked data flips between table and bar-chart form with a toggle. Time series render as line charts where a trend is the real question.
And because dashboards feed meetings, the unglamorous features matter most: every table exports to CSV in a click, and every view prints cleanly for the production meeting binder. A dashboard you cannot get out of the screen is a demo; these are built as working tools.
- 13 curated views spanning orders, finance, inventory, production, purchasing, shipping, quality, and cross-module summaries.
- Fresh on every open, with a visible as-of timestamp and refresh control.
- KPI tiles, click-to-sort tables, chart/table toggles, and site comparison.
- CSV export and print-ready layouts on every view.
The Design Decision That Matters: Same Views as Chat
The most important property of these dashboards is invisible: they read the same semantic layer as SyteRay's chat answers. Open orders on the dashboard is the SyteRay_ open-orders view; ask the question in chat and the generated SQL reads the identical view. The dashboard number and the conversational number cannot diverge, because they are the same query target by construction.
This is the difference between a dashboard product and a truth layer with dashboard surfaces. In the traditional stack, the BI semantic model and the operational reports drift apart, and reconciling them becomes a standing chore. Here there is one set of definitions, visible as SQL views in your own database, and every surface renders from it.
When a Dashboard Raises a Question, the Answer Is One Click Away
Dashboards are where questions start, not where they end. A spiking tile on the Executive Pulse raises why, and in a conventional stack, why means opening a different tool. Because SyteRay's dashboards live beside its conversational layer, the follow-up happens in place: ask about the anomaly and drill into the same data the tile rendered, with the evidence attached.
That loop, notice on the dashboard, interrogate in chat, export the answer, is the actual workflow these tools exist to serve. The curated views are also defined as data rather than code, so extending the set for a site is configuration work, not a development project.
Key Takeaways
- 1SyteRay ships 13 curated dashboards over live SyteLine data, covering orders, finance, materials, production, and cross-module summaries.
- 2Every view is fresh on open, sortable, chartable, exportable to CSV, and print-ready.
- 3Dashboards and chat read the same semantic views, so their numbers agree by construction.
- 4Follow-up questions happen in place through chat, turning dashboards from endpoints into starting points.
See your own SyteLine data on these dashboards, live, during an evaluation. Contact ajay@netray.co.
More Insights
From Question to Governed SQL: How SyteRay's Evidence-First Answers Work
Generic AI chatbots fail on ERP data because they guess at schemas and invent numbers. SyteRay takes a different contract: every answer over your SyteLine database ships with the SQL that produced it and the tables it read. Here is the pipeline that makes that possible.
SyteRayHow SyteRay Classifies Every SyteLine Stored Procedure's Side Effects
A mature SyteLine site runs on thousands of stored procedures that nobody fully remembers. SyteRay crawls the whole catalog, 8,636 procedures on the reference deployment, and classifies what each one reads, writes, and calls, including transitive effects. Here is how and why.
SyteRayWhy We Built Semantic Views for SyteLine: One Source of Truth for Open Orders
When two SyteLine reports disagree about how many orders are open, the root cause is almost never a bug. It is two different definitions. SyteRay's semantic layer encodes each business definition exactly once, as reviewable SQL views, and everything reads from them.