May 2026 Summaries

The emergence of customer context layers marks a shift in AI development, focusing on real-time, identity-resolved data to enhance the performance of AI agents. Unlike traditional data layers that support analytics, these layers provide immediate context to customer-facing AI agents, enabling them to act accurately based on current customer behaviors. This infrastructure collects and processes behavioral data across all customer interactions, ensuring that AI agents can make informed decisions in real-time. The importance of such layers is underscored by the fact that most companies lack a robust data foundation, which is crucial for trustworthy AI performance. By integrating context layers with existing data platforms like Snowflake and Databricks, companies can improve both customer-facing and analytics agents. This real-time context is critical as consumer expectations of AI continue to rise, with businesses needing to match these expectations to maintain trust and competitiveness.

Snowplow has introduced a range of AI-driven tools to streamline data workflow processes, including the Snowplow Assistant and the Snowplow MCP server. The Snowplow Assistant, embedded in the Snowplow Console, allows teams to design tracking, manage pipelines, and resolve data quality issues through conversational interfaces, utilizing natural language to draft data structures and generate tracker code. It operates within the user's existing permissions and requires explicit confirmation for data modifications, ensuring data security and privacy. Meanwhile, the Snowplow MCP server extends the Assistant's capabilities to various AI tools like Claude and Codex, supporting remote operations through OAuth authentication. Additionally, the Snowplow CLI MCP server facilitates local management of tracking plans as code. These innovations mark the beginning of a broader initiative to integrate AI as a central component of Snowplow's data management solutions.

Identity resolution is the process of unifying various user identifiers, such as cookies and device IDs, into a single customer profile. There are two main types: batch identity resolution, which updates profiles on a scheduled basis and is suitable for retrospective analysis, and real-time identity resolution, which updates profiles as events occur, enabling immediate personalization and decision-making. The choice between batch and real-time impacts how effectively personalization, AI, and activation systems can operate, with real-time resolution offering a significant advantage for live, context-sensitive applications. While batch solutions are cost-effective for non-time-sensitive tasks like cohort analysis or lifetime value modeling, real-time solutions are crucial for applications requiring up-to-the-moment data, such as dynamic personalization or AI-driven customer interactions. Snowplow Identities, for instance, offers real-time identity resolution that operates within the data pipeline, ensuring that user profiles are instantly updated and accurate for immediate use in various business applications.

An identity graph is a sophisticated data structure designed to unify all identifiers generated by an individual, such as cookies, emails, device IDs, and IP addresses, into a single customer profile, effectively addressing the challenges posed by multi-device user interactions. Unlike simpler models that flatten customer data into single rows, identity graphs use nodes to represent identifiers and edges to depict the relationships between them, allowing for efficient merging of profiles based on deterministic and probabilistic matching techniques. These graphs are crucial for accurate data analytics, personalized marketing, and AI-driven decision-making, as they ensure that fragmented customer journeys across various devices and sessions are correctly resolved into cohesive profiles. While they can be built in-house using SQL or purchased as part of a Customer Data Platform (CDP), the most effective approach is integrating them into real-time data pipelines, allowing for immediate resolution and personalization capabilities. As privacy regulations like GDPR require strict compliance, well-designed identity graphs ensure that identifier deletion requests are propagated across all connected nodes, maintaining user privacy and data control.