Synthetic demo · Evidence published

Synthetic multi-entity finance analytics

A public-safe reference for group-style management reporting: synthetic operating-company data → dimensional model → finance semantic model → management views for P&L, variance and period comparisons.

Flow diagram from synthetic OpCo sources through dimensional and semantic models to Power BI — Architecture flow for the finance topic. Report stills and demo GIF appear under Result.

Business context

Multi-entity environments often need comparable management reporting across operating companies — actual vs budget, forecast, MTD/YTD and like-for-like views — without every team maintaining its own spreadsheet logic.

The problem

Fragmented spreadsheet reporting makes definitions drift, reconciliation harder, and executive views slower to trust when entities, currencies and periods differ.

My role

Designing and building a synthetic end-to-end reference that mirrors those patterns so the modelling, DAX logic and controls can be shown publicly.

Approach

Define a small set of synthetic OpCos and a shared account / period / scenario structure.
Build a dimensional model suitable for management reporting.
Encode P&L, variance and period logic against that model.
Publish labelled management-style stills (native Power BI .pbix optional).
Document controls: definitions, lineage and synthetic disclosure.

Architecture / model

See the flow above. Optional Fabric path lives under Data Platforms — not required for this demo.

Business logic (in the stills)

P&L structure by entity and account
Actual vs budget (and forecast in the dataset)
Group operating-profit trend across periods
Variance bridge from budget operating profit to actual
Entity hierarchy for group vs OpCo views

Controls

Shared synthetic metric definitions (Revenue, COGS, OpEx, operating profit)
Single currency (GBP) and labelled scenarios (Actual / Budget / Forecast)
Clear lineage: CSVs → measures → labelled stills
Every visual watermarked as synthetic / public-safe

Result

Published evidence from the synthetic model — management-style report stills (not employer dashboards). Native Power BI .pbix remains optional.

Short looping demo cycling through synthetic finance report stills — Short demo motion (GIF) cycling labelled synthetic stills — not a live Power BI embed.

Operating profit by entity Actual vs Budget for June 2026 synthetic data — OpCo operating profit — Actual vs Budget

Group operating profit trend Actual vs Budget over synthetic periods — Group operating profit trend

P and L variance bridge from budget to actual operating profit — P&L variance bridge

Composed synthetic management report page with KPIs and charts — Composed management overview

Source tables (CSV):

What I am learning

How to make finance reporting logic explicit and reviewable — so a semantic model is not just a dashboard, but a controlled analytical product.

Public-disclosure note

This case uses synthetic companies and figures only. It does not represent discoverIE data, systems or internal architecture.