StoryCareer Editorial
Published Aug 3, 2026 · 1 min read
I joined a company where every team had its own spreadsheet and nobody trusted the numbers. My job became less about analysis and more about building one source of truth.
My first analytics role was at a company with fourteen dashboards, all slightly different, all contradicting each other. I spent my first month just trying to understand which number was correct for a single metric.
The data itself was a mess: no consistent definitions, no documentation, and team members hard-coding numbers into presentations. Any report I produced was immediately questioned because past reports had been wrong.
I wrote down a single definition for each core metric and got it approved by the stakeholders who argued about it. Then I built one query per metric, documented every transformation, and published a short weekly summary using only those queries.
I maintain the core data model in the morning, answer ad-hoc questions over lunch, and spend the afternoon building one new dashboard or refreshing the weekly numbers.
Analysts fail when they answer questions nobody asked. Before you build anything, write the question down and ask people what decision the answer will change. Trust is built by documenting your assumptions, not by having the fastest query.
After eight years in corporate jobs, I left to build a small software product. It failed twice before I found something people were willing to pay for.
I started a weekly newsletter about marketing experiments. Two years later it pays a meaningful part of my income — and it taught me more than any job.
I moved from a retail job into security analysis through free labs, homelab practice, and a first role that paid less than my old job but taught me everything.