240 days of copy and paste, now 20 minutes
- BRP
- Global Digital Transformation Specialist
- 2026
Staff role, not a client engagement
Translations came back from the translation system as XML. The templates that build the pages are spreadsheets. Moving one into the other by hand cost about 240 days across the brands. It now takes twenty minutes.
At a glance
20 min
for what used to take about 240 days
99%
less effort on every translation round
Zero
copy-paste errors, by construction
The situation
Every page that ships in more than one language makes the same round trip: strings out to the translation system, translations back as XML, and then somebody puts them into the spreadsheet templates the pages are built from. That last step was done by hand, string by string, for every brand and every market. Added up across the brands, it was roughly 240 days of work.
What I did
Two steps, one round trip
A Python step reads the XML the translation system exports and normalizes it. The shape of that export is not the shape the templates want, and reconciling the two is most of the work. An Apps Script step then writes the result into the template spreadsheets in place, so nothing downstream has to be rebuilt or re-linked.
The error mattered more than the time
The time is the number people repeat, and it is the less interesting one. Manual import meant a string could land one row off, and nobody would find out until it was live on a page in a language they don't read.
It also meant the templates went stale. Revising a translation was cheap, but re-importing it was expensive enough that people put it off, so the spreadsheet and the truth drifted apart between rounds. Twenty minutes makes re-importing free, and that is what actually keeps the two together.
Results
- About 240 days of manual import across the brands, down to 20 minutes
- 99% less effort on every translation round
- Copy-paste errors out of the process, not reduced
- Templates stop going stale when a translation is revised
Skills used
- Data Pipelines & Automation
- Governance & Standards
Systems: Python, Google Apps Script, Google Sheets
