Multi-brand industrial manufacturer
≈ $200M revenue across five brands (approx.) · 5 brands · 4 ERPs · 1 CRM
Consolidation, forecast and pricing intelligence across four ERPs
One governed lakehouse behind an AI agent: audit-grade consolidated financials that trace to the source transaction, forecast and pipeline by business unit, large-job margin watch, and plain-English inquiry over all of it, inside the company's own Azure tenant.
- Organization
- Privately held industrial manufacturer, five brands built through acquisition
- Systems
- Four ERPs across the brands, one CRM, spreadsheet-driven consolidation
- Runs on
- Microsoft Azure, in the customer's own tenant, on Microsoft Fabric
- Scope
- Governed lakehouse, financial core, forecast and pipeline, pricing intelligence, AI inquiry
Where they started.
Each acquired brand kept its own ERP, four different systems from four eras, and the group's consolidated view lived in spreadsheets stitched together every month by hand. Leadership could not ask a simple question across brands, the controller wanted a ledger tape back to the source transaction before trusting a consolidated number, forecast accuracy by business unit was never tracked, and every new acquisition restarted the whole exercise. The group runs on Microsoft, so a second data cloud was never on the table.
What was built.
- A governed lakehouse on Microsoft Fabric in the customer's own Azure subscription, fed from all four ERPs and the CRM over the company's WAN, with entity matching across brands and lineage back to every source transaction.
- The financial core on that store: consolidated financials, business-unit scorecards, forecast and pipeline, large-job margin watch, an AR and collections cockpit, and data-quality dashboards, all reading the same governed numbers.
- FlatClaw on top, through MCP connectors: finance asks why a margin moved and gets a cited answer; sales asks about a strategic account and gets revenue, margin, pipeline and whitespace across every brand it touches, with updates written back to the CRM.
- Pricing intelligence and aftermarket analytics as the next phase on the same foundation, because the estimate and the actual finally live in one place.
- An onboarding pattern for the next acquisition: map the new ERP once, in weeks, and the agent, the scorecards and the consolidation inherit it.
What changed.
- One set of numbers, traceable to source, instead of a monthly reconciliation exercise.
- Forecast versus actual tracked over time by business unit, so forecast accuracy becomes a measured thing.
- Plain-English inquiry across brands for people who never opened an ERP.
- Acquisitions land as a repeatable, separately priced integration instead of a project each time.
- Nothing about the data or the model leaves the company's Azure tenant; other brands' staff join as guests with their home-tenant sign-in.
Consolidated financials, margins and pricing for a private group are exactly the data that cannot be sent to a third-party model on every question. The agent answers from the lakehouse inside the tenant, and the audit trail makes every answer defensible.
Microsoft Azure
Components involved.
More like this.
European logistics group
≈ $2B revenue · 6,000+ employees · 50+ countries
- Use Case:
- Quote control tower over forwarding systems
- Results:
- Quotes that track fuel prices, routing and political risk across the group's forwarding systems, assembled by agents over the company's own data lake and handed to the desk with the reasoning attached.
Sleep-therapy provider
≈ $60M revenue (approx.) · 40 clinics · one clinical reporting app
- Use Case:
- Natural-language clinical reports
- Results:
- Clinicians ask for patient cohorts in plain English and a constrained pipeline turns the question into validated filters, never free-form database queries, on a private GPU that costs a fraction of the hosted model it replaced.
Global law firm
≈ $3B+ revenue · Thousands of lawyers · dozens of offices
- Use Case:
- Walled-matter research assistant
- Results:
- Document search and drafting inside the firm's own tenancy: every answer cited to its source, every matter isolated by role, and nothing ever sent to a third-party model.
Have one like it?
Every spotlight started as a conversation about a process nobody liked doing, under a data-locality constraint.