Synthetic portfolio project — not paid production experience.

Aug 2026 · Synthetic case study

Operations Reporting Control Tower

A bounded technical case showing how a reporting model can make defects, controls, reconciliation, and management review visible.

One-line outcome

Built and validated a static synthetic reporting system with 14 imported tables, 27 DAX measures, six pages, 12/12 quality controls, and 7/7 reconciliations.

View résumé
At a glance
Context
Synthetic data only · no client
Role
Independent portfolio build
Scale
14 imported tables · 27 DAX measures · six pages
Scope
Static PBIX · testing · reconciliation · documentation
Original synthetic-proof abstraction
Synthetic inputsSQL modelControlsStatic PBIX
12/12 quality controls7/7 reconciliations

Static technical evidence only. No live refresh, client, cloud, or production environment is implied.

01 / Context and ownership

Deliberate defects make the controls testable.

Operating challenge

The project supplies technical proof that confidential professional work cannot safely show: model structure, DAX measures, report-page design, quality controls, and independent reconciliation.

What Isaac owned

  • Design of the deterministic synthetic dataset and reporting scenario.
  • SQL star schema, calendar model, 14 imported tables, and 27 committed DAX measures.
  • Six report pages, automated tests, data-quality controls, and independent reconciliations.
  • Documentation, owner-QA evidence, and an explicit static-PBIX limitation.
02 / Process and decisions

Defects, controls, reconciliation, and limitations were designed together.

Deterministic synthetic inputs moved through a SQL star schema into committed measures and six static report pages. Independent tests and reconciliations verified the model rather than relying on the interface alone.

01

Seed deliberate defects so quality controls could be tested rather than merely described.

02

Validate source and model totals independently.

03

Keep the PBIX static and disclose the missing live-refresh layer.

04

Separate interface evidence, model diagrams, and validation evidence.

05

Document limitations as part of the technical result.

03 / Approved outcomes

Independent reconciliations verify the conformed result.

12/12

Quality controls

29 raw defects identified

7/7

Reconciliations

Independent checks passed

6/6

Code tests

Deterministic validation

$290,200

Net revenue

Exact synthetic reconciliation

V-403 / Validation registerRaw defects stay distinct from conformed-model results.
29 deliberate raw defects0 conformed-model failures27 committed DAX measures6 static report pages
Selected synthetic evidence

Static interface evidence

These crops are synthetic snapshots, not a live dashboard.

Synthetic Control Tower executive overview crop showing six summary measures and a management summary structure.
Executive overview · synthetic static snapshot
Synthetic Control Tower management summary crop showing what changed, what it means, and what needs attention.
Management-summary evidence · synthetic static snapshot
Technical detail where useful

Technical proof includes what the model cannot claim.

Power BI, DAX, SQL, star-schema modelling, testing, reconciliation, and documentation support the case. Seven independent checks passed, including an exact synthetic net-revenue reconciliation of $290,200.

The PBIX is a validated static snapshot with no Power Query or DataMashup live-refresh layer. It demonstrates a project model, not an enterprise architecture or operational service.

Evidence disclosure

What is safe to show

Synthetic portfolio project — not paid production experience. The data is synthetic, the PBIX is a validated static snapshot, and there is no live refresh, client deployment, or cloud or enterprise environment.

What remains outside the claim

What is deliberately absent

No client system, paid deployment, production cloud platform, enterprise warehouse, live refresh, or operational service is claimed.