New Way Now: Dun & Bradstreet cuts data workflows to minutes; boosts product response times by 60%
33Featured in this video: Michael Manos, Chief Technology Officer of Dun & Bradstreet
Executive summary: Dun & Bradstreet, a leading provider of business data and analytics, is bringing its centuries-old data company into the AI era, migrating 150 products into Google Cloud. Leveraging services like Dataproc, BigQuery, and Bigtable, Dun & Bradstreet modernized its data pipelines, which ingest and process five exabytes of raw data nightly. This resulted in significant improvements: core data quality checks were reduced from hours to minutes, and data publishing time was cut in half. The move has not only boosted website and data product responses times by over 60% but also paved the way for AI innovation. Utilizing Vertex AI and Google’s multimodal Gemini models, Dun & Bradstreet launched its advanced gen AI assistant, Chat D&B™ , which enables customers to ask questions conversationally and surface actionable insights from across its vast universe of data.
Challenge: Dun & Bradstreet wanted to accelerate its robust data pipelines to improve the timeliness of its data while enhancing its ability to deliver actionable insights on business records from over 580 million organizations in more than 250 markets across the globe.
Solution: In partnership with Google Cloud, Dun & Bradstreet migrated 150 products, modernizing its data pipelines, which ingest and process five exabytes of raw data nightly, with services like Dataproc, BigQuery, and Bigtable. Migrating to Google Cloud has helped Dun & Bradstreet significantly increase the speed of data flows, reducing quality check processes from hours to minutes and cutting the time it takes to publish new data in half. This strong data foundation also enables Dun & Bradstreet to leverage the full power of Google Cloud’s ecosystem, including cutting-edge data and AI technologies like generative AI. Utilizing Vertex AI and Gemini, Dun & Bradstreet launched ChatD&B, its advanced gen AI assistant, which enables customers to ask questions conversationally, surface actionable insights, and quickly take action on them — from contacting the right prospects to pulling out fraud and risk information and much more.
Results: With Google Cloud, Dun & Bradstreet has increased the timeliness and quality of the data it provides — all while boosting its website and product response times by more than 60%. In addition, Dun & Bradstreet is now primed to embrace gen AI for enhancing other areas of its business, such as implementing Gemini Code Assist to boost developer productivity and reduce the time it takes to deliver new innovations.
By the numbers snapshot:
150 different products and applications moved to Google Cloud
11-hour global data quality checks reduced to minutes with Dataproc
50% reduction in total data pipeline publishing time with BigQuery
Hours-long data ingestion processes reduced to minutes with Bigtable
62% overall improvement on response time for websites and data products
2x faster development cycles achieved with Gemini Code Assist, cutting multi-sprint tasks to one or two sprints
Key takeaways and highlights from our interview with Michael Manos, Chief Technology Officer of Dun & Bradstreet:
→ “We have moved 150 different products to Google Cloud. We consume on a nightly basis about five exabytes of data in its raw form. Using Dataproc, we saw some of [our] quality checks go from 11 hours on a global set of data down to minutes. BigQuery has allowed us to reduce the total pipeline publishing of our data end to end by 50% across the globe. Bigtable has reduced our overall load jobs from hours to minutes. All these things have had an amazing increase in performance — our customers have seen a 62% improvement on the response time of our websites and the products they use.”
→ “With Gemini and Vertex AI, we were able to lock in quickly a great model and launch the first D&B commercial generative AI platform. Chat D&B allows you to interface with the entire universe of data that we have to target businesses and you can then tie into the sales and marketing databases that we have to pull out contact information, fraud and risk information, and get those results interactively and integrate it into your CRM or automatically generate email campaigns.”
→ “The adoption of generative AI internally has been a game changer for us. Utilizing things like Gemini Code Assist, we have seen sprints that would normally take three or four sprints to happen, get cut down to one or two sprints. In terms of the development cycles, it's been incredibly powerful.”
Google Cloud products used: Vertex AI, Gemini, Dataproc, BigQuery, Bigtable, Gemini Code Assist
Learn more:
→ Dun & Bradstreet Launches ChatD&B™ , Its Advanced Gen AI Assistanhttps://investor.dnb.com/news/news-details/2024/Dun--Bradstreet-Launches-ChatDB-Its-Advanced-Gen-AI-Assistant/default.aspx..