Pulls new Etsy listings, generates Pinterest-ready captions and descriptions with Claude, and queues them for posting. Automates the daily grind of cross-promoting a print-on-demand shop.
The Etsy to Pinterest Bot is an internal Python system that pulls new listings from an Etsy shop, generates Pinterest-ready captions, descriptions, hashtags, and pin titles using Claude, and queues them for posting via Make.com. It replaces the daily ritual of manually cross-promoting every new product, while preserving the brand voice and avoiding the generic Pinterest-spam patterns.
Running a print-on-demand Etsy shop means uploading new designs regularly, and every new design needs to be cross-promoted on Pinterest. Pinterest is where print-on-demand lives or dies: it is the discovery engine, and a listing without pins is a listing that does not sell.
The manual version of this workflow was brutal. For every new Etsy listing: download the cover image, log into Pinterest, pick a board, write a pin title, write a description, pick hashtags, schedule it. Multiply by 5-10 listings uploaded per week and you have 30-60 minutes of menial work every day that directly impacts revenue. Skip a week and sales dip.
The Pinterest-spam pattern was the other problem. Most automation tools produce the same formulaic pin descriptions: "Beautiful [product]! Perfect gift! Click to buy!" That style gets throttled by Pinterest and ignored by users. Whatever replaced the manual version had to write pins that actually read like a designer wrote them.
The system is a three-stage pipeline: Etsy listing fetcher, Claude-powered content generator, and Make.com-based Pinterest publisher. Every new listing gets pulled from the Etsy API on a schedule, Claude writes the pin title, description, and hashtag set in the brand voice, and Make.com handles the actual Pinterest posting with the image attached.
The Claude prompt is the heart of the system. It includes a voice specification, an anti-pattern list (no "perfect gift", no "click to buy", no excessive exclamation marks), and platform-specific guidance for Pinterest SEO: keywords in the first 80 characters, long-tail phrases in the description, hashtag discipline. Every pin that comes out feels written by the shop owner, not a bot.
A review queue sits between generation and publishing. Nothing goes live until the human says yes. Most days that is a 5-minute review instead of an hour of manual work.
The lesson from this build is that content automation only works when the voice layer is tight. A fast bot that produces generic pins is worse than posting manually, because Pinterest throttles generic content and the owner has to clean up the mess. A slower pipeline with strict voice constraints is the version that actually saves time.
Book a free 30-minute call. We map your situation, scope a similar build, and agree on a fixed quote before anything starts.
Book Free 30-min Call