I learned the business before I learned the tools.
Before I worked in data, I spent 14 years working in operations and procurement. That taught me to question numbers before trusting them, understand how processes create data, and recognise when something looks technically correct but operationally wrong. Now I combine that judgement with SQL, Python, Power BI and modern data workflows.
Four live New Zealand government datasets combined to score how well each rural parcel suits kiwifruit.
Weather, port congestion, fruit quality and regulatory load drive a simulator of risk, cost of delay and a 26-week forecast.
Ministry of Education roll data for the Bay of Plenty-Waiariki region, modelled in a SQL star schema and analysed with window functions. An interactive dashboard tracks growth and capacity pressure; a Power BI report is planned next.
Both projects run on live government and public sources, with the real data-quality problems that come with them.
Every model states what it doesn't cover, so nobody mistakes a scope decision for an oversight.
Analysis should reduce uncertainty, not just produce another chart.
I write for the person who has to act on the analysis, not just the person who built it.
I trust it after I've asked where it came from, how it was transformed, what assumptions were made, and whether the result behaves the way reality says it should.
Tools: SQL, Python, R, Power BI, Excel, Power Query, ETL, data quality checks, Git.
You don't need a new system. You need someone who can look at your Xero data and your spreadsheets and tell you, plainly, what's going on.
Example dashboard with invented data. Yours is built around your own questions.
If you're looking for someone who can bridge operational reality and analytical work, I'd be happy to talk.
Want to know what your own numbers are telling you? Start with a fixed-price Snapshot Report.