# We build the technology too

- Language: en
- Canonical: [https://www.yaneken.cl/en/technology](https://www.yaneken.cl/en/technology)
- Other languages: [Español](https://www.yaneken.cl/tecnologia) · [简体中文](https://www.yaneken.cl/zh/technology) · [Deutsch](https://www.yaneken.cl/de/technology) · [Italiano](https://www.yaneken.cl/it/technology)

We develop our own tools, with artificial intelligence, to improve how we buy, run stores and execute across our channels. We use them every day in our 136 stores.

## What a brand can expect

### Store standards and operational follow-through

*Will my brand be executed consistently?*

Store teams check every campaign against the brand standard with a photo and report issues with a voice note that becomes a routed ticket. (Iris)

### Demand planning and buying

*Can you make informed buying decisions?*

Before the season, we size footwear buys by store cluster, with an uncertainty band for every buy or pass decision. (Delphine)

### Marketplace and e-commerce operations

*Can you manage multiple channels?*

We automate marketplace operations and check every day that the price in SAP matches the storefront, with evidence of competitor prices. (Octavius, Cerebro)

### Store economics and portfolio management

*Can you scale responsibly?*

Every store has its own P&L and every lease its financial scenario, so openings, closures and renegotiations are decided with traceable information. (Meridian, Atlas)

## The systems behind the operation

These products were born inside Yáneken, and each one generates data that feeds the others.

### Iris · Retail AI

The retail operating layer that runs on WhatsApp: store teams send a photo or a voice note and Iris checks campaigns against the brand standard, opens and routes maintenance tickets and answers customer reviews. No apps, no training.

- 136 stores covered
- <10s from photo to verdict
- 196 listings managed

### Delphine · Demand Forecasting

Pre-season footwear buying decisions: from a single photo of a model that has not launched yet, it predicts weekly sell-through by store cluster, sizes the order and returns a buy/pass verdict with a calibrated uncertainty band.

- 134,690 training rows
- 0.723 out-of-sample AUC
- 105 model variables

### Octavius · Marketplace Ops

AI automation platform for marketplace operations: it orchestrates multi-channel workflows, removes manual work and keeps e-commerce running at scale, in real time.

- Multi channel
- AI workflow automation
- 24/7 real-time operation

### Cerebro · Pricing Intelligence

Competitive price control for multi-unit retail: it centralizes official and competitor prices with evidence per source, checks SAP against the storefront every day and flags exposed products before they cost margin.

- Daily price checks
- 3 business units
- 100% backed by evidence

### Atlas · Retail Real Estate

The platform that connects real-estate management with each store's profitability: it centralizes leases, rent terms and expiries; organizes openings, closures and relocations; and simulates the impact of new commercial terms on results. A shared view of the portfolio to negotiate rents and assess investments with traceable information.

- 136 stores managed
- 25,477 m² floor area managed
- 3 financial scenarios

### Meridian · Management P&L

The platform that connects the income statement with operations: monthly P&L by brand, store and GL account, budget vs. actual and four-wall profitability for every point of sale, with traceable allocation of central costs. A single source of truth for openings, closures and renegotiations.

- 136 stores with their own P&L
- 4-wall EBITDA per store
- Monthly budget vs. actual

And the stack keeps growing: Andrea (marketing), Chispeza (voice AI), Cifra and more products built in-house by the Yáneken team.

## How we built them

- [Yáneken runs on its own artificial intelligence across all its stores (ES)](https://www.yaneken.cl/noticias/yaneken-retail-ai-first)
- [How we built Iris: the AI that runs our stores over WhatsApp (ES)](https://www.yaneken.cl/noticias/asi-construimos-iris)
