Data Analysis
Built on my own journey mastering analytics and business insight since 2016
Business need
Most small and mid-sized businesses I worked with struggled to make sense of their own data. Financial information such as revenue, profit, and margin existed, but it rarely turned into actionable insight. Many owners managed operations by intuition rather than evidence, without a clear picture of what products, customers, or channels were actually driving their results.
Beyond that, basic financial control was often fragmented across different spreadsheets, with no consolidated view of performance over time. This made it hard to evaluate growth, define realistic goals, or understand how small operational decisions impacted overall profit.
One of the companies I still support today faces a more complex analytical challenge: managing and pricing a massive catalog of over 100,000 SKUs in the collectibles market. With such a large and constantly changing assortment, identifying profitable items, monitoring stock movement, and setting competitive prices across countries became an ongoing struggle.
My role and contributions
I built practical dashboards and simple financial models so small and mid-sized owners could read their numbers without complex spreadsheet manipulation. The reports pull revenue, cost, and margin from their own spreadsheets and ERPs and update automatically. Within a few weeks they were comparing product and channel profitability, adjusting prices, and planning cash flow with fewer surprises. Each view includes drill-downs to source tables and a short KPI dictionary, which improves trust in the numbers and speeds up monthly reviews.
For the collectibles retailer, I combined marketplace scraping with public and partner datasets to stitch a single catalog of 100k+ SKUs traded in Brazil and the US. I created a BR↔US product crosswalk to align items across markets, using normalized titles, set codes, editions, printings, language, condition, and package type to match products reliably. With that unified view, I flagged where it made sense to buy here and sell there, or the reverse, projected days of inventory, and suggested price ranges item by item based on live competition. Price guidance considers total landed cost, marketplace fees, and FX, meaning foreign exchange between BRL and USD. The result was fewer stockouts, less dead stock, and clearer rules for replenishment supported by a simple weekly governance cadence.
