New Way Now: Snowdrop transforms messy transactions into clear insights with AlloyDB and Gemini
26Featured in this video: Daniel Ritchie, Head of Product at Snowdrop, and Daniel Acosta, Head of Technology at Snowdrop. Snowdrop is a leading provider of advanced transaction enrichment solutions in EMEA and Asia Pacific, and is dedicated to empowering banks and payment providers.
Executive summary: Google Cloud partner Snowdrop enriches transactional data at scale with geolocation information, helping financial institutions match transactions to merchants. Using AlloyDB, Snowdrop created a scalable, unified data foundation for its Merchant Reconciliation System (MRS) API, enabling the company to embrace the latest AI technologies, including generative AI, to deliver accurate insights into every purchase. With AlloyDB, Snowdrop can process billions of transactions monthly, utilizing Google’s multimodal Gemini models to power advanced enrichment capabilities like vector search and automatic translation. Choosing solutions that already integrate with its existing product ecosystem, along with the flexibility to work with various models, including open models, has dramatically reduced Snowdrop’s development time — cutting months of data preparation down to just one week.
Challenge: With more than 2.1 billion transactions processed monthly, Snowdrop was looking to improve the quality and speed of its enrichments for customers. In addition, Snowdrop wanted to connect its data in a secure way to the latest AI technologies, such as gen AI, to create advanced enrichment capabilities that make it even easier to identify merchants.
Solution: With its MRS API already being powered by Google Cloud, Snowdrop chose AlloyDB — Google Cloud's 100% PostgreSQL-compatible database service — to enable more efficient data management and improve merchant information processing. AlloyDB provides enterprise-grade performance and availability for even the most demanding transactional workloads, helping Snowdrop establish a scalable, secure, and reliable data foundation for AI innovation. In addition, AlloyDB makes it easy to access and build with state-of-the-art models, whether Google models like Gemini, open models, or even custom models. Using Gemini, Snowdrop has enhanced its enrichment capabilities through vector search and automatic translation of information from global merchants operating in multiple languages and alphabets.
Results: Now, Snowdrop can easily scale when onboarding customers, creating custom solutions, or expanding into new regions, handling billions of transactions monthly and cutting development times from months down to weeks. In addition, Gemini’s multimodal capabilities have helped to enable vector search and automatic language translation, improving Snowdrop’s ability to match merchants and transactions more easily, quickly, and accurately.
Key takeaways and highlights from our interview with Daniel Ritchie, head of product at Snowdrop, and Daniel Acosta, head of technology at Snowdrop:
→ “We’re using Gemini to transliterate and enrich around locations. We have many vocabularies in Europe, and Gemini is good for this instruction and it saves a lot of time.”
→ “The benefits of using AlloyDB internally have been evident. It basically opens up new opportunities for us to work on new use cases. At the same time, it makes our developers contribute in a way that they were not able to contribute before.”
→ “In Europe, we have a lot of languages, a lot of different regions, and the governments are starting to put their own models in universities, and we can use those to provide better enrichment. The use of AlloyDB with open LLMs (large language models) can target specific problems. Avoiding vendor lock-in, also, is more proof-safe for the future because [of] regulatory changes. So, we choose whatever is best for the task.”
→ “Every client comes with a set of problems, on their data, on their regulations. The flexibility [of] Google Cloud allows us to meet the client’s needs. Kicking off a new region always took months to get the data to a place where we were comfortable. Now, we can do this in a matter of one week.”
Google Cloud products used: AlloyDB, Gemini
Learn more:
→ Snowdrop: Enriching transactional data at scale to foster global financial transparency and trust https://goo.gle/41DIZRa
→ Snowdrop launches its next generation of geo-based transaction enrichment services for a better digital banking experience https://goo.gle/4hHe9gc
→ Snowdrop Solutions https://goo.gle/4iwkQDc