Anonymized engagement summary: Power BI semantic model and four tax-operations reports on a MySQL source for a tax services firm (2024)
Anonymized engagement summaryTax services firm · 2024
Opening answer. In 2024 EPC Group was engaged by a tax services firm to make Power BI the firm's organization-wide BI tool, starting with a semantic model on its MySQL database and up to four reports chosen from the firm's own list:
- fixed asset reconciliation
- tax bill check-off
- tax accrual summary
- tax savings
The work runs from source assessment through deployment in the Power BI service, with a user guide and on-the-job knowledge transfer.
This is an anonymized engagement summary published under EPC Group's anonymized-engagement rule: the client is not named, no commercial terms are published, and every statement is EPC-reported. Named, documented engagements are in the Evidence Center.
Problem and scope
| Item | Scope |
|---|---|
| Organization | A tax services firm |
| Situation | A growing firm wanting to work more efficiently and extend Power BI as its organization-wide business intelligence solution |
| Data source | A MySQL database |
| Reports | Up to four Power BI reports from the firm's numbered report list: fixed asset reconciliation, tax bill check-off, tax accrual summary and tax savings, styled after reference dashboards the firm supplied |
| Delivery | Assessment and discovery, on-the-job knowledge transfer, short delivery cycles that each ship a working piece, led by EPC Group's Power BI architects and developers |
Not reported: data volumes, user counts, the go-live date and any measured change in tax-operations effort.
Approach and artifacts
The engagement breaks the build into named steps.
| Step | What it covers |
|---|---|
| Initial assessment and planning | Assessment of the MySQL source |
| Semantic model | Identify each report's metrics; define the logic, filter criteria and expected values for each; build source queries and model relationships; complete transformations; create measures and calculated columns; validate every component against its expected value |
| Report design and security | Assemble the validated components into reports with drill-down, drill-through, bookmarks, buttons, slicers and other customizations; create end-user groups |
| Deployment and documentation | Provision Power BI workspaces, configure gateways, and write an end-user guide |
Validate the pieces before the page. Every metric gets an agreed expected value before it is built, and each measure is checked against that value before it is placed on a report. That is what makes a reconciliation or accrual report trustworthy to the people who sign off tax positions.
What EPC Group delivered and what is not claimed
| Claim | Basis | Status |
|---|---|---|
| The firm contracted EPC Group for a semantic model and up to four tax-operations reports | EPC engagement record | Contracted (EPC-reported) |
| Reports delivered, accepted or in use; any measured result | Not reported | Not claimed |
Related services
Power BI consulting · All case studies · Talk to a senior architect
Last updated by Errin O'Connor, Founder & Chief AI Architect, EPC Group
Related service: Power BI Accelerators: CoE, Health, Migration
Author: Errin O'Connor, Founder & Chief AI Architect, EPC Group · Published October 25, 2026 · Project alias EPC-2024-ANON-TAX-PBI · Published under the anonymized-engagement rule.