The most consequential innovations in e-commerce —Amazon’s logistics, Apple Pay, and Meta Ads — only marginally increased the share of ecommerce vs retail. The core consumer experience – browse, search, checkout – remains fundamentally staid.

Did ecommerce reach peak evolution?

I’d argue one structural problem has made the sector impervious to broad technological innovation: the ecommerce data problem.

Every ecommerce site is a snowflake (yes, even on Shopify). There is no common structure or ontology for products and categories. Checkout funnels vary wildly. There are infinitely many permutations of technology stacks, with no dominant patterns. So any change or integration of new technology requires armies of systems integrators and service agencies, and scales linearly at best. I know this firsthand because I spent 10 years as cofounder and CTO at Bluecore, building and scaling a unicorn marketing automation platform for ecommerce.

This is why innovations like virtual try-on never caught on. This is why product recommendation algorithms feel random at best. This is why finding the right-sized garment is still a guessing game. This incongruous divergence has defied automation. And even the mighty ChatGPT struggled with product feeds.

AI is a platform shift that will once again redefine how people shop online, but to realize the AI Storefront, the ecommerce data problem must be solved.

A Different Approach

Octogen is building infrastructure for AI-enabled commerce.

First, we’re eating the frog.

We’re building infrastructure to ingest, structure, and enrich ecommerce data automatically into a common structure and ontology that is both deep and wide. If you’ve built ecommerce tech before, this is the kind of work that requires teams of integration engineers scaling linearly with the number of sites and customers. Our approach is to scale this with fully automated agents.

We’re also starting with the apparel category because the data model is notoriously complex: discerning visual variants, understanding sizes and measurements, understanding fabrics, silhouettes, and closures.

The first manifestation of our technology is Cosimo. Give it a spin.

What’s next

We’re just getting started. Next are tools for brands, retailers, and chatbots.

For now: try Cosimo and let us know what you think.

We’re adding brands every week, and if yours should be one of them, get in touch.

— Mahmoud

You can subscribe here to stay tuned for updates. Our team is mostly engineers who don’t love writing, so we promise not to spam.