‹ all projects
Procter & Gamble · 2020–2022

Automating the finance work nobody wanted

An ETL + web-scraping automation that reproduced a manual finance task end to end, saving thousands of hours a year.

automationetl

Problem

A finance colleague spent ~4 hours a week on a manual data task — and 20+ colleagues did similar work. Nobody wanted to touch it, and nobody wanted it disrupted.

Approach

I built an ETL + web-scraping automation that reproduced the task end to end, then validated it against their manual output before anyone had to trust it.

Result

~4,000 work-hours per year saved across the team, and the approach was reused in other departments.

Notes

The hard part wasn’t the code — it was earning trust before replacing a process people depended on.

Screenshots coming soon.