April 2025 Summaries

Data quality is crucial for every successful business initiative, but many organizations struggle with duplicate records despite investments in robust data infrastructure. Census Entity Resolution is a powerful new capability built directly into the Census platform that helps identify, merge, and resolve duplicate records across business data, addressing the fragmented approach to data quality and siloed processes. By using deterministic rules with fuzzy matching capabilities, Census Entity Resolution can detect duplicates within datasets and provide a simplified setup experience, seamless integration with existing data workflows, and lower costs compared to existing solutions like AWS's solution. The capability is designed to help organizations resolve duplicate records across their entire data ecosystem, eliminating chaos and improving marketing, sales, customer support, analytics, and other business processes.

Automating ticket categorization with Large Language Models (LLMs) can lead to faster resolutions, better resource planning, and improved product development prioritization. To set up this system, one must first define clear categories that make sense for the business, considering issue types, priority levels, and required teams. A well-crafted prompt is then used to instruct the LLM on how to categorize tickets, with refinements possible through testing and iteration. Once a working prompt is achieved, it can be implemented across the entire support system using various data tools or custom integrations, ensuring consistent data formats, feedback loops, and bias monitoring. Advanced techniques include sentiment analysis, auto-response suggestions, root cause identification, and tracking metrics to measure categorization accuracy. By implementing LLM-based ticket categorization, businesses can empower their agents with superpowers, focusing on solving complex problems and building relationships with customers.

Census's AI Columns are a powerful tool for applying standard sets of categories to large, complex datasets, making it easier to understand and navigate the data. Categories can be used to summarize data such as user feedback or firmographic data, and can help spot emerging trends quickly. LLMs like OpenAI, Claude, and Gemini can assist in reliably applying categories, and Census makes it easy to apply these techniques at scale. The process involves defining your categories, writing a prompt that includes the template variables, list of categories, instructions on how to use them, and any additional context, and then using preview to see how AI would categorize some of your records. Once saved and run, the data will be available in the dataset, and quality control can be performed by creating a new Basic Dataset with SQL. The results can be analyzed using SQL queries, and categorized records can be used in Segments, Syncs, or other workbench transformations. Advanced Mode allows for AI to generate categories if data is unclear, involving aggregating the data, generating a single block of text, and then using an LLM to recommend a set of categories.

The Embedded Segment Builder is a new capability introduced by Census, empowering SaaS platforms to provide their customers with sophisticated audience creation and activation capabilities. This builder allows users to build complex segments using an intuitive visual interface and instantly activate them across 200+ integrations without leaving the platform. The Embedded Segment Builder addresses challenges such as creating an intuitive interface for complex segment logic, managing segment computations and real-time previews at scale, maintaining integrations with hundreds of tools and platforms, and ensuring consistent segment definitions across systems. With this builder, customers can connect to any data source, create segments with real-time feedback, activate them anywhere, and implement it in 1-2 days with Census handling the complexity of computation, previews, and data syncing.

The company is expanding its offerings to support businesses in creating a fully connected source of truth, providing tools like CSV datasets to help them manage difficult-to-house data. The challenge of "rogue" data, where one-off information remains disconnected from workflows, is being addressed by giving every piece of data a proper home within the Census workspace. This is achieved through a seamless process that requires zero coding or data engineering resources. Real-world applications of CSV datasets include event marketing, sales enablement, partner management, and ad-hoc analysis. The company's commitment to supporting a true central source of truth acknowledges that businesses have edge cases and one-off data needs, providing structured flexibility to maintain governance while accommodating the realities of business.