How is the cataloging bottleneck currently solved, and why doesn’t that work?
A patchwork of manual workflows and single-purpose tools
Today's textile resale operations use a fragmented stack:
- Photography: Smartphone cameras, basic lighting rigs, background removal apps. No automated feature extraction from images.
- Description writing: Manual and often inconsistent. Staff must identify garment type, brand, material, condition, and measurements, then write copy for each platform's audience.
- Pricing: Manual research on eBay "sold" listings and marketplace comps, or using ChatGPT. Takes 3–5 minutes per item and requires significant category expertise.
- Cross-listing tools: Automate copy-paste of existing listings but don't create them. All classification, description, and pricing work remains manual.
- POS systems: Handle in-store sales but offer zero AI classification and no online distribution.
- Inventory tracking: Spreadsheets, basic POS, or enterprise systems not designed for one-of-one SKUs.
Why this doesn't work: three structural failures
First, classification and distribution are decoupled. Resellers must use one set of tools to figure out what an item is and completely separate tools to list and sell it. This creates a bottleneck that compounds with every additional item.
Second, too much double work. Cross-listing tools push the same listing everywhere identically. They don't route a vintage Hermès scarf to Vestiaire Collective (where luxury buyers pay premium prices) while sending an H&M basics tee to Vinted (where volume buyers seek value).
Third, the tools don't scale. Existing solutions requires human effort proportional to item count. Double your inventory, double your labor. They physically cannot process more without hiring proportionally.
Fioobra is built specifically for this reality. It's an AI-native platform that turns the capture–catalogue–distribute workflow into a near-autonomous system, so resale operations can scale throughput without scaling headcount.