S/01
Power BI dashboards
End-to-end delivery — data model, DAX measures, Power Query, publish and scheduled refresh. Optimised for the questions your team keeps asking.
Power BI dashboards that get read, not filed. Freelance developer & data analyst — turning messy Excel, ERP exports and SQL data into one place where the numbers finally agree.
End-to-end delivery — data model, DAX measures, Power Query, publish and scheduled refresh. Optimised for the questions your team keeps asking.
Star-schema modelling, calculated tables, DAX measures for KPIs, variance, trend and forecasting. Built to scale as your data grows.
Power Query (M), SQL, Python (pandas) for cleaning inconsistent inputs from Excel, ERPs and databases into one reliable reporting layer.
Every model documented, every KPI defined, dashboards your team can maintain after I leave. No black boxes, no vendor lock-in.
The best dashboards start with a boring conversation — not with DAX.— Working principle
Building Power BI dashboards for the engineering, project and operations teams — turning scattered Excel exports across multiple sites into a single source of truth for management review.
Includes: star-schema data model, DAX measures for progress and variance tracking, Power Query pipelines from Excel and SQL, scheduled refresh, documented handover.
Executive overviews and drill-down reports across sales, product margin, customer segmentation and store performance — the reporting stack retail teams keep opening every Monday morning.
Includes: revenue and margin models, cohort and RFM segmentation, category and store league tables, mobile-friendly Power BI service layout.
For teams whose data lives in Google Sheets, Airtable, Stripe, HubSpot and a dozen CSV exports — one Power BI layer that consolidates it, plus Python-based clean-up where the source is genuinely broken.
Includes: source audit, ownership map, KPI dictionary, incremental refresh, and a monthly review template the team can run without me.
Which decision does this dashboard help you make? Who opens it, and how often? What does good look like as a number?
One name per thing. One date format. One owner per source. One clear definition per KPI. Boring, and it prevents 80% of the bugs.
Star-schema model, measures instead of calculated columns, one date table marked as a date table, every step in Power Query named.
A written data model, a KPI dictionary, and a short walk-through session so your team can maintain and extend the report themselves.
| Tool | Where it fits | Fluency |
|---|---|---|
| Power BI (Desktop & Service) | Modelling, publish, gateway, scheduled refresh, workspace and app deployment. | Daily |
| DAX | KPI, variance, YoY, MTD/YTD, time intelligence, calculation groups. | Daily |
| Power Query (M) | Connecting, cleaning, shaping data from Excel, CSV, SQL, APIs and folders. | Daily |
| SQL | Queries, joins, window functions, view design, extracting reporting layers from OLTP. | Weekly |
| Advanced Excel | Pivot tables, Power Pivot, formulas, automation, monthly reporting templates. | Daily |
| Python (pandas) | Data cleaning, transformation and one-off analysis where SQL/Power Query hit a wall. | As needed |
| Figma / Canva | Wireframing report layouts, presenting findings, one-page executive summaries. | As needed |
Send me a short note with what your team is trying to see — the sources you have, the decisions you make on Mondays, or the report you wish existed. I'll come back with how I'd build it and a realistic timeline.