verticals — the first, BIKE, has been on the market since 2021
Pipeline platforms share one product model and are bound to a vertical by design; standalone ones sit outside that pipeline and are deployed on their own. Products are a different thing: stateless tools, published as a single instance whoever arrives uses directly.
KOSMOS ingests and normalises raw price lists → EIDOS is the data-model authority every other platform reads from → AGORA, EMPORION, HERMES, KYKLOS and LOGOS all consume that normalised identity. Inside the pipeline, no product exists unless it exists first in EIDOS. The standalone ones — TAMEION, ABAX, METIS, STOA and PRAXIS — have no such dependency: each is deployed on its own, one instance per client.
Ingests raw supplier price lists in any shape and resolves brand, product type and attributes against the central catalogue — proposing only values the model already contains, never inventing them, and learning from every correction an operator makes.
The authority on what a product is. Holds the model, the taxonomy and the digital assets that every other platform reads from. This is the agreement the whole system rests on.
The public-facing marketplace and discovery layer of each vertical: structured product pages generated from the model, plus new and used sales listings from businesses and private sellers.
The transactional layer, retail and dealer. It lives inside the pipeline: it consumes the normalised product identity, and a product created here is registered in the central model automatically.
An immediate start and fast updates. It is the right fit for catalogues that do not need continuous updates. It can host configurators, sales journeys built around the product, specific pricing and booking rules…
The data-communication layer between brands, distributors and retailers: controlled inventory visibility, price lists, purchasing rights, territorial exclusivities and dealer segmentation, with published and uniform fees.
Warehouse and inventory operations on the same model the commercial platforms use, so stock is counted against the identity that is actually sold, listed and ordered.
The content layer: builds and manages the websites of the network on top of model data, so public pages stay aligned with the products a business actually carries.
Accounting registration driven by AI. Standalone: it does not depend on the product taxonomy and can be deployed on its own.
Reads across every platform plus external sources to support decisions. Standalone, and already running on a client outside the bike vertical.
Simple, transparent condominium management. A working tool for the managing agent and the board, and a newspaper always open to every resident.
A tool for organising training projects: from planning to assessment, by way of teaching materials, attendance tracking and much more.
A product is not a platform deployed small: the difference is one of nature, not of how it is installed. A platform has state — a database, users, one instance per client; a product has none, and lives as a single public instance the end user works with directly. AGORA and HERMES sit on a public address, but they remain platforms in full: behind them there is a data model, clients and permissions.
What a car really costs, month by month, with every cost in: the total cost of ownership comparator, in five languages, online since 1 October 2026. There is nothing to install and no client data lives inside it.
A pipeline platform is only as good as the shared model behind it, which is why it cannot be dropped into a company in isolation: it needs the vertical. A standalone platform has no such dependency and can start on day one. Knowing which kind you are looking at is the difference between a two-week deployment and a programme.
A product, on the other hand, is not a smaller platform: with no state to keep, there is no instance to assign to anyone and nothing to configure before using it. You open it and use it. For a platform, even one with a public address, the word is always deployment — which is why AGORA and HERMES are platforms and TELOS is not.
That is the problem we work on. Tell us what you have and what you want to get out of it: we will tell you honestly whether it makes sense — and you will probably be surprised, quite soon, by how much a well-built instrument can surface from data you have been sitting on for years. You will see and measure patterns you would never have guessed were there.
Write to us →