Universal Search, also referred to as Semantic Search, is now available (early access) within Veza. As the first release, it allows users to locate queries and dashboards with ease and accuracy, moving beyond the limitations of traditional keyword targeted searches. PS – in the upcoming releases, Universal Search will allow users to search user access reviews (UAR) campaigns, access profiles, integrations, and across every product and capability.
Understanding the “Why” Behind a Search
A keyword based targeted search is designed for precision and is highly effective when you know the specific terms in a query’s name or description. But what happens when your goal is broader? For example, a user looking for “disabled users accessing snowflake” is expressing an intent—to find insights about a specific risk. A search that looks for a literal text match is doing its job perfectly, but it may not connect that search to a query named, for instance, “Terminated Identities” The user’s goal isn’t just to match words, but to find content that matches their conceptual need.
On the contrary if you search for the same in the traditional Query listing search interface you may not get your desired output.
Universal Search: Understanding User Intent
Universal Search fundamentally transforms this experience for our users. It’s an Access AI capability designed to understand the intent and contextual meaning behind your query. Instead of just matching keywords, it strives to grasp what you are truly seeking, even if the precise terms aren’t used.
Behind the scenes, our Access AI models build a sophisticated contextual map of your Veza environment. Through a process that generates unique vector embeddings, the models analyze your queries and dashboards—their titles, descriptions, and filter logic—to assign each item a precise position on this map based on its meaning.
When you perform a search, the AI plots your request onto this same map, instantly identifying the queries and dashboards that are conceptually closest. This method allows the search engine to uncover highly relevant results that keyword-matching would miss, delivering a far more effective and intuitive experience.
Key Benefits and Capabilities:
Intuitive Discovery: Users can now search using natural language, describing what they’re looking for, rather than needing to remember exact query names. For example, the search “disabled users accessing snowflake” will now intelligently surface relevant queries related to disabled users and their Snowflake access.
Comprehensive Results: Users will see related queries and dashboards that may be pertinent to their search, facilitating faster discovery and a broader understanding of available queries and insights.
Efficient Searching with Minimal Input: Universal Search can provide relevant results even with brief or less specific search terms.
Understanding Nuance: It also allows users to comprehend common scenarios, including negations (e.g., “users not in X group”) and access patterns to discover relevant queries and dashboards.
By understanding intent and meaning, Universal Search empowers users to navigate their data landscape more effectively, uncover insights faster, and ultimately gain more value from the platform. This new capability – powered by Access AI – represents a significant step forward in making insights discovery within Veza more powerful and user-friendly. In the coming weeks – we will extend Access AI powered Universal Search to all our products – please help share feedback, we would be grateful!
