Case study · Operations & data
Turning a drive full of PDFs, scans and spreadsheets into clean data
The data the team needed was stuck inside files: PDFs, photos and scans, and CSV exports in different formats, all sitting in Google Drive. We built an n8n workflow that reads every file, uses AI to extract and tidy the data, stores it in Supabase and puts it on a dashboard the team can actually use.
The challenge
What wasn’t working
The data the team needed sat inside files in Google Drive: PDFs, images and scans, and CSV files, each in its own format.
Scanned documents and images can’t be searched or reported on until someone reads them, and different file formats don’t line up with each other.
They needed the information in one structured place, kept up to date as new files arrived.
The solution
What we built
Automated file intake
An n8n workflow that picks up files from Google Drive as they arrive, whether PDF, image or CSV.
AI OCR for PDFs and images
Gemini reads scanned and image-based documents that ordinary text extraction can’t handle.
Data normalization
Values from every file type are cleaned and mapped to one consistent structure, so they can be compared and searched.
Structured storage
The normalized data is stored in Supabase, ready for reporting and further automation.
Internal dashboard
A dashboard that puts the extracted data in front of the team for quick access and analysis.
One workflow, every file type
PDFs, images and CSV files all go through the same workflow, so nobody has to sort or convert files first.
How it works
From file to dashboard
- 1
File lands in Drive
A PDF, image or CSV is added to the shared Drive folder.
- 2
n8n picks it up
The workflow detects the file and sends it down the right path for its type.
- 3
AI extracts the data
Gemini handles OCR for PDFs and images; CSVs are parsed directly.
- 4
Data is normalized
Every value is mapped to the same fields and formats.
- 5
Stored and shown
Records are saved in Supabase and appear on the internal dashboard.
The result
What changed
- Data from PDFs, scans and CSV files ends up in one consistent format
- No more opening files one by one to copy values out
- New files are processed automatically as they arrive
- The team works from a dashboard instead of a folder of documents
More work
More case studies
Have a project like this?
Tell us what you need to build. In 30 minutes we’ll map what to build first and follow up with a fixed-scope estimate.
Official Retool Agency Partner5.0 on Clutch, 8 verified reviewsBuilding software since 2015Reply within one business day