Hong Kong’s top enrichment engine

gini’s enrichment engine has been trained on millions of transactions from thousands of app users over the years. As a result, our Hong Kong data enrichment quality is higher than any other provider in terms of accuracy, scope and speed.

Unmatched experience

We've been training our machine learning models for more than 3 years and counting
training our machine learning models
and counting
Our machine learning models have been trained on more than 4 million transactions and counting
transactions enriched
and counting
Our leading merchant database boasts 20,583 unique Hong Kong merchants and counting
Hong Kong merchants
and counting

Our powerful enrichment engine is the product of multiple data sources, public and private APIs, complex algorithms, human intelligence tasks and advanced machine learning models that have been refined continuously over the years.

This enables us to deliver an unrivalled level of quality attributes, structure and context to your raw transaction data in real time, saving you the years of effort it would take to label your data manually or train your own systems.

As a unique combination of data sources, complex APIs, algorithms and human intelligence tasks — improved and refined continuously over the years — gini’s powerful enrichment engine delivers an unrivalled level of quality attributes, structure and context to your raw transaction data.

Powered by machine learning, our enrichment engine saves you the years of effort it would take to label your data manually.

A leading Hong Kong financial institution compared gini to other vendors by running 50,000 transactions through each system and analysing the results.
gini came out on top, with:
gini achieved the highest percentage transactions enriched compared to other providers
Credit card
transactions enriched
gini achieved the highest percentage transactions enriched compared to other providers
EPS
transactions enriched

Better categories

From years of operating and refining one of the top personal finance apps in Asia, we’ve developed highly accurate, user-approved categories tailored to the Hong Kong market.
gini's categorisation engine has evolved and improved according to user preferences

User-approved

Informed by the spending habits of our app users, plus more than 1,000 descriptive tags, the quantity and type of categories we use has evolved to achieve optimal relevance, clarity and workability for the user.

Conglomerates often have different sub-brands with different categories

Greater accuracy

gini’s categories distinguish between each sub-brand of conglomerates, and include even small vendors not equipped for credit card payments.  

Plus, gini categorises merchants by their business names, rather than their legal company names which often differ greatly by industry. All this ensures a high degree of accuracy that cannot be achieved by any other provider.

MCC codes
USA-oriented
Set categories based on merchant industry
Conglomerates
Credit card accepting merchants only
Legal company name — may not accurately represent the merchant’s category
gini
HK-oriented
Evolving categories based on user preferences and descriptive tags
Conglomerates + sub-brands
All merchants
Trade name — accurately represents the merchant’s category
gini's data enrichment can be accessed on AWS Marketplace

Instant deployment

We designed our enrichment engine as an online software plug-in solution that can be  deployed to your own infrastructure in minutes — not months. 

  • One-click launch
  • Installed by Amazon Web Services
  • No additional infrastructure needed
  • No third-party data access
  • World-class security
Get in touch with the gini team today
Get in touch to arrange a demo