04Automation
POD Engine
print-on-demand ecommerce system and automation
Cultural signal to listed product in hours, with one person deciding taste and legal risk.
- Status
- Live
- Period
- 2025–2026
- Relationship
- In-house venture
- Site
- Private
- Available as
- Build for you
- Retainer
The pipeline
- 01 CaptureTrends pulled from live cultural signals
- 02 ScoreSignals ranked against stored niche performance
- 03 IdeateProduct ideas generated per scored niche
- 04 ScreenLegal review pass before a person sees it
- 05 ApproveOne human approves or kills, by chat
- 06 DesignArtwork generated, then published to the printer
- 07 ScheduleSocial posts queued behind every listing
- 14production workflows, all active
- 9stages from trend to storefront
- 1human approval gate
- 51%unit margin on the flagship SKU
The problem
Print-on-demand rewards speed and punishes taste failures. Winning a trend means going from cultural signal to listed product in hours — but shipping unscreened designs against live trends is how a store collects takedown notices.
What we built
- 01
Chained the whole pipeline as discrete workflows: audience bootstrap, trend capture, scoring, idea generation with a legal review pass, an approval gate, design generation, print-on-demand publishing, then social scheduling.
- 02
Kept exactly one human checkpoint — approval — so taste and legal risk stay with a person while everything mechanical runs unattended.
- 03
Persisted niches, design styles and scored ideas in Postgres so the system learns which audience segments convert instead of restarting cold.
- 04
Drove approvals and status through a chat bot, so the operator's entire interface is a phone.
- 05
Instrumented unit economics per SKU — cost, retail and margin tracked as first-class fields, not reconstructed later from exports.
Stack
- n8n
- PostgreSQL
- Printify
- Postiz
- Telegram
- Railway
Ideal for
Merchandise, apparel and licensing operators who need daily output at trend speed without a studio payroll or a takedown notice.