TalksAWS re:Invent 2025 - Accelerating E-commerce Insights with Snowflake Intelligence (AIM113)

AWS re:Invent 2025 - Accelerating E-commerce Insights with Snowflake Intelligence (AIM113)

Accelerating E-commerce Insights with Snowflake Intelligence

Addressing the Gap Between Business and Data Teams

  • Enterprises face a critical problem with the disparities between how business teams and data teams operate and work together.
  • Data teams spend a lot of time creating static reports that are often outdated by the time the insights are delivered, making it difficult for business teams to act on them.
  • Existing solutions are rigid and struggle to provide the "why" behind the data, limiting the ability to make informed decisions.
  • There is a need to bridge the gap between business users and data, enabling everyone to ask deep questions, get faster insights, and make more confident decisions.

Introducing Snowflake Intelligence

  • Snowflake Intelligence is an enterprise intelligence agent that empowers every employee to ask deep questions, get verified answers, and leverage generative AI to solve problems.
  • It enables:
    • Deep analysis and reasoning to answer questions like "What should I do next?" and "What are the headwinds for my business?"
    • Transparent and verified answers with full traceability to the data sources.
    • Enterprise-ready security, governance, and data quality within the Snowflake platform.

Key Features of Snowflake Intelligence

  1. Dynamic Discovery:

    • Provides clear explanations of how the agent reasoned and arrived at the decisions.
    • Enables users to test assumptions, explore data, and develop comprehensive answers.
  2. Trusted Insights:

    • Generates verified answers with full traceability to the original data sources.
    • Reformulates questions, provides explanations, and exposes the underlying SQL queries.
    • Ensures transparency and observability for trustworthy decision-making.
  3. Comprehensive Data Analysis:

    • Allows users to leverage all their data, including structured and unstructured, to answer complex, forward-looking questions.
    • Connects to various data sources, including the Snowflake Marketplace, to provide a holistic view.
    • Enables deep reasoning and analysis across the entire organization's data.

Desai's Journey with Snowflake Intelligence

  • Desai, an e-commerce analytics platform, faced challenges with their previous BI solution, where business users became increasingly reliant on the data team to access insights.
  • They explored building an AI-powered "e-commerce analyst" to remove the barrier and empower their customers to directly access and analyze their data.
  • After evaluating various AI frameworks, Desai found Snowflake Intelligence to be a promising solution that aligned with their requirements.

Implementing Snowflake Intelligence

  1. Semantic Views:

    • Leveraged their existing DBT model documentation to create semantic views in Snowflake.
    • Extended the views with additional metadata, such as aliases, synonyms, and sample values, to provide context for the AI agent.
  2. Search Services:

    • Implemented Cortex Search Services to enable fuzzy matching and improve the user experience for freeform product discovery questions.
  3. Agent Configuration:

    • Defined the agent instructions, including the model and other configurations, to tailor the intelligence agent to their e-commerce domain.
  4. Deployment Pipeline:

    • Automated the provisioning of the Snowflake Intelligence stack, including semantic views, search services, and the agent, as part of their deployment process.

Benefits and Impact for Desai

  • Reduced the time and effort required for business users to access insights, enabling them to directly ask and answer complex questions.
  • Provided transparency and confidence in the data through the "thinking steps" and traceability features of Snowflake Intelligence.
  • Integrated the Snowflake Intelligence agent seamlessly into their own application, creating a branded "Luma AI" experience for their customers.
  • Observed a 75% reduction in support requests as users could independently extract insights and take action.
  • Enabled features like segmentation, campaign activation, and scheduling that were previously difficult to implement.

Future Roadmap

  • Customizing the agent instructions and context on a per-brand basis to better accommodate the nuances of different e-commerce businesses.
  • Empowering users to infuse their own domain knowledge and brand-specific instructions into the agent.
  • Developing scheduling and alerting capabilities to enable the agent to proactively monitor and report on key business metrics.

Your Digital Journey deserves a great story.

Build one with us.

Cookies Icon

These cookies are used to collect information about how you interact with this website and allow us to remember you. We use this information to improve and customize your browsing experience, as well as for analytics.

If you decline, your information won’t be tracked when you visit this website. A single cookie will be used in your browser to remember your preference.