How far does the data coverage go?

From the past 90 days through upcoming schedules. Estimated sales and units sold are provided per broadcast. Sales are not calculated per product.

Can we bring the same engine into our own service?

The attribute extraction, broadcast speech-to-text and OCR, and data pipeline behind DataHuB can be brought to your catalog too. Share your goals and data scope in an inquiry, and we'll recommend the right combination.

What can we see on screen?

Schedules and product data for 18 channels, category and attribute trends analyzed by AI, and estimated sales and captions per broadcast. Analysis views keep expanding.

Can general ecommerce businesses use it?

Yes. Brands, vendors, and ecommerce merchandisers who need to watch the home shopping market use it together. It's also used for product sourcing and market research.

Can we try it for free?

Sign up and start using the basic features right away. If you need deeper analysis or API access, we'll walk you through the plans.

DataHuB

Data from 18 Korean home shopping channels, on screen and by API

A commerce AI engine cleans up schedules, products, and estimated sales from 18 Korean home shopping and T-commerce channels every day. Read the market on screen, and pull the same data as is through the partner API.

Talk to our team Visit DataHuB

DataHuB live service screen: real-time broadcast schedules across home shopping channels

Scattered data can't back a decision

When schedule, sales, broadcast, and behavioral data sit apart, teams fall back on gut feel, and teams that need data collect it from scratch every time. Clean it up in one place and it's ready to use on screen and by API.

Conventional operations compiled by hand

Operations run on gut feel and manual work

Schedule and sales info compiled by hand, channel by channel
Competitor channel moves spotted too late
No way to go back to what was said on air
Teams that need data collect it anew every time

Operations backed by clean data

Schedules and product data from 18 channels on one screen
Trends by category and attribute
Sales pitches and offers from broadcast captions
The same data supplied as is through the partner API

A screen to read the market, an API to receive the data

On top of commerce data collected by Hsmoa, attribute extraction AI, sales estimation models, and broadcast speech-to-text add insights, and the data is delivered by API.

Schedule analysis

All 18 channels' schedules at a glance

See schedules and their patterns by time slot, channel, and category, and track what competing channels are doing.

Product/attribute trends

What's trending, by attribute

Read market trends at a level deeper than product names, using category and attribute data cleaned up by attribute extraction AI.

Estimated sales

How much each broadcast sold

See estimated sales and units sold per broadcast, and how they split across products in the same broadcast. These are model estimates, so read them as trends.

Broadcast content analysis

From on-air remarks to sales pitches

Converts broadcast audio to text to analyze remarks, keywords, and sales pitch patterns. Information inside video becomes searchable data.

Broadcast speech-to-text keyword analysis

Partner data API

The same data, straight into your systems

Schedules, products, estimated sales, and brands from 18 Korean home shopping and T-commerce channels, plus broadcast speech-to-text captions, are supplied through the partner REST API. Integration is server to server and secured to contracted IP addresses.

Built on data that's hard to replace

Commerce source data collected over years, combined with commerce-specific AI, gives DataHuB its depth.

OWNED DATA

Commerce source data collected by DataHuB

18 home shopping and T-commerce channels combined
Schedule and product data, plus broadcast video and audio archives
From the past 90 days to upcoming schedules

COMMERCE AI

Cleaned up in production by the GEO engine

Attribute extraction AI cleans up product context
Broadcast speech-to-text and OCR turn video into text
Depth general-purpose BI tools can't reach

DATA PLATFORM

Served on screen and by API

Teams read the market on screen
Companies that need data pull it through the partner API
Estimated sales and brands included as fields

IN PRODUCTION

Proven in daily work and external supply

In daily use by home shopping and commerce teams
Supplying data externally through the partner API
Analysis views expanding based on feedback

DataHuB is where the commerce GEO engine runs on real data

It collects data from 18 channels, a commerce AI engine cleans it up daily, and it's served on screen and by API. You can bring the same engine to your own catalog.

Commerce data pipeline

Collects and cleans up schedule, product, and broadcast data from 18 home shopping and T-commerce channels, keeping it ready for analysis at all times.

Attribute extraction AI

Extracts categories and attributes from product names, images, and descriptions, laying the foundation for trend analysis and API data.

Broadcast STT/OCR

Turns speech and captions in broadcast video into text, creating data you can search and summarize.

Collaboration can take many forms, from screen subscriptions and data supply through the partner API to bringing the same engine to your catalog. Tell us the data scope and purpose you need, and we'll suggest the right combination.

Talk to us about partnering

Frequently asked questions

Find answers to common questions before you get started.

Basic covers schedules and product lists from 18 Korean home shopping and T-commerce channels. Premium adds estimated sales, units sold, and brands per broadcast. Broadcast speech-to-text (caption) data is also available by API. Access is server to server only, and calls are accepted only from IP addresses registered in your contract.