Automating stock photo processing and metadata for 500+ photos a day
The problem
A company that shoots photos and sells them on photo stocks produces 500-600 new photos every day. Before each photo can be uploaded, it needs post-processing, a title, a description and a set of keywords. Doing this by hand took a huge amount of time and delayed uploads, and the later a photo goes live, the later it starts earning.
What I built
A scheduled workflow on n8n that runs every night, without anyone involved, and prepares the whole batch for upload.
How it works:
- Drop the photos in a folder. The team leaves the day’s 500-600 photos in a shared folder. That is the only manual step.
- Runs on a schedule at night. The workflow starts by itself and picks up every new photo from the folder.
- AI post-processing. Each photo is processed automatically, so it is ready for upload.
- AI analysis of every photo. The AI looks at each image and understands what is in it: subject, scene, mood and context.
- Metadata for each photo. A title, a description and keywords are written for every photo.
- One ready CSV file. All the metadata is collected into a single CSV file, ready to use for the upload to the stock.
Result
- In the morning everything is ready: processed photos and a finished CSV. The team only uploads them to the stock.
- 500-600 photos a day are handled with no manual work.
- Saves the team more than 20 hours a week.
- Titles, descriptions and keywords stay consistent across the whole catalog.
Tools
n8n, AI image analysis, CSV export.
Details are anonymized. The client has not been named and no figures are shown beyond what was approved.