Data analysis · Power BI · Excel

From scattered spreadsheets to numbers you can act on

I build dashboards and automate reporting for companies that have plenty of data and very little insight. No six-month implementation programme — a working report, usually within weeks.

What usually brings me in

  • Monthly reporting eats three days of manual work
  • Three departments, three versions of the same revenue figure
  • A dashboard that was built and never opened again
  • A spreadsheet only one colleague still understands
  • Data living in four systems and meeting in none

What I do

Why a former maths teacher builds your dashboard

I spent six years teaching secondary school mathematics, part of it in bilingual education. That is not trivia on a CV — it is the skill most BI projects are missing.

Dashboards rarely fail on technology. They fail because nobody agrees on what a KPI measures, because two departments define "revenue" differently, or because the intended user was never taught what to do with it. Explaining things until they land was literally my profession.

Alongside that I work day to day as a data analyst in a regulated, KPI-driven environment. It keeps me honest about the things that actually matter: source data that does not add up, numbers that need to survive an audit, and stakeholders who are right to be sceptical.

How it works

  1. Intro call (free, 30 minutes)

    You show me what happens today. I tell you honestly whether I add anything — sometimes the answer is that you need a better export, not a dashboard.

  2. Fixed price on a defined outcome

    No open-ended hourly rate against a vague goal. We agree what exists when it is done, and what it costs.

  3. Build with interim deliveries

    You see the report while it is taking shape, so it surfaces early if I am answering the wrong question.

  4. Handover and documentation

    Training for the users, documentation for your administrator. The goal is that you do not need me to keep it running.

Want to talk through your reporting?

Half an hour, free, no sales pitch. You will at least know whether the problem sits in the data, the tooling or the process.