The AI Data Platform
built for fashion retail

See COORD on your market. Book a demo and we’ll show you where your prices, range and markdowns sit against your competitors, or send the team a question.

COORD's benchmark: 6% above market, with prices by category and markdown depth against the market

Built and advised by operators from

You already bought the AI. It can’t read your market or your business.

01The data lake.

Your data in one place.
The market’s still outside it.

Your warehouse joined up sales, stock and marketing, but competitor prices, ranges and campaigns never land in it. COORD brings your data and the market’s into one lake.

  • Snowflake
  • BigQuery
  • Databricks
  • ClickHouse

02The AI licences.

Trained on the internet.
Not fluent in fashion.

General models can’t tell a shacket from a gilet, or a markdown from a mid-season sale. COORD’s fashion-trained models and semantic layer fill that gap.

  • ChatGPT
  • Claude
  • Gemini

03The agents.

Ready to act.
Nothing current to act on.

Pricing, inventory and marketing agents are only as good as their context. Feed them last month’s export and they make last month’s call.

  • Pricing
  • Inventory
  • Marketing

A market report tells you what was true. COORD tells you what’s true this morning.

COORD is the semantic layer that joins your market and your business: competitor prices, campaigns and ranges alongside your own stock, sales and plans. Your pricing, inventory and marketing agents decide on clean, current context, not last month’s export.

Market reportStale

A report is out of date the day it lands.

Accurate the week it was compiled.

Still accurate38%
  • Camel wool coat

    Haldane

    £24512 Sept
  • Ribbed merino jumper

    Loomwell

    £12012 Sept
  • Wide-leg trousers

    Fenna

    £9512 Sept
  • Cord shacket

    Arlo & Wren

    £15012 Sept
  • Poplin shirt

    Rue Ondine

    £8512 Sept
COORD

COORD reads the market, your competitors and your store.

One semantic layer, updated as the shelves change.

Still accurate99%
  • Camel wool coat

    Haldane

    Price cut −30%
    was £245now £172Just now
  • Your camel coat

    Your store, via Shopify

    Low stock: 18 left
    £2602m ago
  • AW26 Tailoring campaign

    Maison Verre, via Meta ads

    New ads
    14 ads9m ago
  • Markdown alert posted

    #buying, via Slack

    Alert
    Camel coats11m ago
  • Ribbed merino jumper

    Loomwell

    Sold out
    £12024m ago

COORD briefs your agents in seconds, answers in the language of fashion and connects your store to the market.

Briefed by COORD in #buying

Pricing agent

AW26 outerwear, −20%

Queued

Inventory agent

Reorder camel coat ×120

Sent

Marketing agent

Pause knit ads in France

Sent

Brief your agents

Answers become actions: markdowns, reorders, paused ads. Posted to Slack or handed to your agents.

W|
Auto

Ask in fashion’s language

Ask the way your buyers talk. COORD’s semantic layer maps it to your categories and the market’s.

Connects to Shopify, Meta, Google Analytics, ClickHouse, Slack, TikTok, Google Sheets, BigQuery.

Connect your store and the market

Shopify, GA4, Meta and Google Sheets, plus COORD’s live read of 10,000+ fashion brands.

14 min
Insights series · Interview · 14 min

Why retail software fails

A decade of watching retail software land or fail to, and the failure patterns that repeat.

Kristina Mills-BarrettFormer global VP of customer success at EDITED
Watch the episode →

It drafts first, then earns the keys. You stay in control.

Step 1 · Discovery

It studies your market

Connect your store and it learns your range, prices and competitors on its own.

Watching 10,000+ brands50M+ products

Burberry
Oh Polly
Marks & Spencer
ASOS
Reiss
Club L London
AllSaints
Whistles
Mulberry
Princess Polly
Next
Barbour
Paul Smith
Nobody's Child
Primark
Jigsaw
Rixo
Boden
Topshop
With Jéan
River Island
Kurt Geiger
Stella McCartney
Damson Madder
New Look
Superdry
Phase Eight
Never Fully Dressed
Alexander McQueen
Boohoo
Vivienne Westwood
Jack Wills
Monsoon
Sister Jane
Discovery complete: studied 312 lines, your 14 closest competitors and 3 markets, with three findings and 9 ranked moves
Markdown drafted: AW26 outerwear, 38 lines at −20%, with the reason, two of the lines and Approve or Not now
Step 2 · Review

You approve, it executes

Every move arrives as a finished draft. Approve, edit or reject: ten decisions in about thirty seconds.

0/10approved as drafted

Pause ads on sold-out products · last 3 weeks

Want me to handle these? The marketing agent offers to pause ads on sold-out products on its own, or keep asking first
Step 3 · Autonomy

It earns your trust

Approve the same move ten times and it offers to take it over. Every action still has an undo.

Platform

Built for the fashion teamsthat take data seriously

Built on a modern data stack and designed around fashion: fast, accurate and always current.

Search demand, UK

Capri pantsNapoleon jackets
COORD forecast
Capri pants−72% from peak
Napoleon jackets+178% in 4 wks
Today
Apr ’26JunAugOctDec ’26

Trend forecasting so you see what's peaking, what's fading and what's next, before it hits the shop floor.

Raw listings

Wmns Oversized Camel Wool-Blend Coat – SALE!!

Haldane · £172

Manteau long camel 70% laine

Rue Ondine · €199

CAMEL OVERCOAT LONG LINE

Arlo & Wren · £180
↓ System 1 · classify & normalise

COORD record

CategoryOuterwear › Coats › Overcoat
ColourCamel
FabricWool blend, 70%
FitOversized, long
MatchYour camel wool coat94%
↓ System 2 · decide with tools
price_gap("WR-1042")+9% vs median
suggest_markdown("WR-1042")−20% → £208

Fashion-native data layer that classifies and normalises every listing, so your agents decide with deterministic tools, not guesses.

Marketing catalogue of every competitor ad on Meta and TikTok, linked to the product and price it's pushing.

If“Every Monday, 8:00am”
AND“Competitor price cuts”>20%
Then“Post the trade briefing to #trading”

COORD · 8:00 AM

Morning. Before today’s trade meeting: Haldane cut outerwear −30% over the weekend, and 4 competitors are now on sale in coats. Your knitwear sits 19% above market. 3 moves are drafted for your review.

Automated briefings so the Monday trade meeting starts with what moved, what it means and what to approve.

Connected to COORD MCP · 10 tools

> ▍

MCP & API so Claude, ChatGPT or your own agents can query COORD's market data directly, with the same tools our agents use.

Similarity searchp95 44ms

38ms

50M+ vectors in ClickHouse

Pages crawled today

1,284,512

Crawling Haldane · Outerwear

Products embedded today

61,204

+24/s
768-dim

Fashion-scale search with COORD's embeddings on ClickHouse, matching 50M+ products in milliseconds.

Security

Safe by designSecure by default

We’re Cyber Essentials certified, GDPR compliant and working towards SOC 2 Type I. We also have AES‑256 encryption, row-level permissions, and audit logs.

SOC 2 Type I

Pending

GDPR

Cyber Essentials

Made to measure

Measure twice,mark down once.

Every buyer, merchandiser and agent works from the same measurements.

Shared definitions

One meaning of sell-through, full-price and margin across buying, merch and finance.

Roles & permissions

Give each team its own brands, categories and markets, and keep margin to who needs it.

Agent access

Agents sign in over OAuth with scoped, short-lived tokens, never a shared key.

Approvals & audit trail

Every agent action and price change is logged with its approver, ready for your SIEM.

Single sign-on

SAML or OIDC through Okta, Microsoft Entra or Google, with MFA built in.

Directory sync

Starters and leavers sync from your directory, so access follows your org chart.

Short on time?
We'll handle the setup

Skip the manual setup. A dedicated engineer handles integrations, data mapping, alerts and agent permissions. Live in under two weeks.

Book a demo
01

Kickoff call

We review your stack, data sources and KPIs, then agree on success criteria.

02

Implementation

We connect your systems, validate the data, and configure alerts and agent access.

03

Review call

We walk through the live workspace, resolve open questions and plan next steps.