<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Do We Need a New Role in Data Organizations — The Semantic Steward? in Data Governance</title>
    <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174523#M3056</link>
    <description>&lt;P&gt;&lt;SPAN&gt;Great perspective! I completely agree that a Semantic Consistency Agent could be a practical starting point. Automating semantic drift detection and lineage-based impact analysis would allow Semantic Stewards to focus on business alignment and governance decisions. &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;AI can identify inconsistencies, but human accountability is essential to establish trusted business meaning.&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; This could be a powerful step toward agent-driven semantic governance!&lt;/SPAN&gt;&lt;/P&gt;</description>
    <pubDate>Sat, 24 Oct 2026 22:58:29 GMT</pubDate>
    <dc:creator>vkondepati</dc:creator>
    <dc:date>2026-10-24T22:58:29Z</dc:date>
    <item>
      <title>Do We Need a New Role in Data Organizations — The Semantic Steward?</title>
      <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174346#M3053</link>
      <description>&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;As data platforms evolve from managing data to managing business meaning, is it time to introduce a new role: Semantic Steward?&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Traditionally, data governance has focused on data quality, ownership, security, lineage, and compliance.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;But with the rise of &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Layers, AI/BI, Databricks Unity Catalog, Metric Views, and conversational analytics through Genie&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt;, another governance challenge is becoming increasingly important:&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Who owns and governs the meaning of data?&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Consider a simple example:&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Different business teams may define &lt;/SPAN&gt;&lt;EM&gt;&lt;SPAN&gt;Revenue&lt;/SPAN&gt;&lt;/EM&gt;&lt;SPAN&gt;, &lt;/SPAN&gt;&lt;EM&gt;&lt;SPAN&gt;Active Customer&lt;/SPAN&gt;&lt;/EM&gt;&lt;SPAN&gt;, &lt;/SPAN&gt;&lt;EM&gt;&lt;SPAN&gt;Customer Churn&lt;/SPAN&gt;&lt;/EM&gt;&lt;SPAN&gt;, or &lt;/SPAN&gt;&lt;EM&gt;&lt;SPAN&gt;Net Sales&lt;/SPAN&gt;&lt;/EM&gt;&lt;SPAN&gt; differently.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The underlying data may be accurate, pipelines may be reliable, and access controls may be correctly configured.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Yet dashboards, business reports, and AI-generated responses can still produce conflicting answers because the business semantics are inconsistent.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;This raises an interesting question:&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Should semantic governance become a distinct responsibility within the enterprise data operating model?&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;/P&gt;&lt;H3&gt;&lt;SPAN&gt;Introducing the Semantic Steward&lt;/SPAN&gt;&lt;/H3&gt;&lt;P&gt;&lt;SPAN&gt;I see the Semantic Steward as a potential emerging role responsible for ensuring that business definitions, KPIs, metrics, dimensions, and semantic relationships remain consistent, trusted, reusable, and governed across the organization.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Potential responsibilities could include:&lt;/SPAN&gt;&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Governance:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Establish and maintain standardized business definitions, KPIs, measures, and dimensions.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Metric Ownership:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Coordinate approval, versioning, and change management for enterprise metrics.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Consistency:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Reduce conflicting definitions across BI platforms, data products, and AI applications.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Business-to-Technical Alignment:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Connect business terminology with physical data models and semantic implementations.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;AI Readiness:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Ensure AI assistants and analytics agents consume validated business context.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Lifecycle Management:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Maintain traceability, certification, change history, and impact analysis for semantic definitions.&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;/UL&gt;&lt;H3&gt;&lt;SPAN&gt;How Is This Different from a Data Steward?&lt;/SPAN&gt;&lt;/H3&gt;&lt;P&gt;&lt;SPAN&gt;A Data Steward typically focuses on the governance, quality, definition, and appropriate use of data assets.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;A Semantic Steward could specialize in governing &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;how those data assets are interpreted and translated into business meaning&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt;.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The responsibilities would naturally overlap, but the accountability could be different.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;For example:&lt;/SPAN&gt;&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Data Steward:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; Is customer data accurate, complete, classified, and governed?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Steward:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; What exactly qualifies as an active customer, and is that definition consistently applied across every analytics and AI experience?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;/UL&gt;&lt;H3&gt;&lt;SPAN&gt;Why This Matters for Databricks&lt;/SPAN&gt;&lt;/H3&gt;&lt;P&gt;&lt;SPAN&gt;As organizations adopt Unity Catalog semantics, Metric Views, AI/BI Dashboards, and Genie, business definitions increasingly become reusable enterprise assets rather than logic embedded separately in dashboards.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;This creates an opportunity to treat semantic definitions as governed data products, with clear ownership, review processes, and change controls.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Technology provides the foundation, but &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;who is accountable for semantic correctness?&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;/P&gt;&lt;H3&gt;&lt;SPAN&gt;I'd Love to Hear the Community's Perspective&lt;/SPAN&gt;&lt;/H3&gt;&lt;OL&gt;&lt;LI&gt;&lt;P&gt;&lt;SPAN&gt;Do you believe &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;Semantic Steward&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; should become a dedicated role in enterprise data organizations?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;SPAN&gt;Should this responsibility belong to existing Data Stewards, Analytics Engineers, or Business Domain Owners instead?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;SPAN&gt;How are organizations currently managing approvals and changes to enterprise KPI definitions?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;LI&gt;&lt;P&gt;&lt;SPAN&gt;As AI agents increasingly consume enterprise data, should semantic governance become a formal part of AI governance?&lt;/SPAN&gt;&lt;/P&gt;&lt;/LI&gt;&lt;/OL&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;My perspective:&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; As AI makes data consumption more accessible, governance must evolve beyond data accuracy to include &lt;/SPAN&gt;&lt;EM&gt;&lt;SPAN&gt;meaning accuracy&lt;/SPAN&gt;&lt;/EM&gt;&lt;SPAN&gt;.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Perhaps the next evolution of data governance is not only governing data, but also governing the business meaning behind it.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;STRONG&gt;&lt;SPAN&gt;What do you think? Is Semantic Steward a necessary new role, or simply an evolution of existing data governance responsibilities?&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;#Databricks #UnityCatalog #SemanticLayer #DataGovernance #DataStewardship #AIAnalytics #MetricViews #SemanticGovernance&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Thu, 22 Oct 2026 16:19:41 GMT</pubDate>
      <guid>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174346#M3053</guid>
      <dc:creator>vkondepati</dc:creator>
      <dc:date>2026-10-22T16:19:41Z</dc:date>
    </item>
    <item>
      <title>Re: Do We Need a New Role in Data Organizations — The Semantic Steward?</title>
      <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174474#M3054</link>
      <description>&lt;P&gt;Venkat, your point about different business teams defining 'Revenue' or 'Active Customer' differently is a classic source of friction. I've seen this in almost every implementation, and you're right that even with perfect data quality, inconsistent semantics undermine the whole platform.&lt;/P&gt;&lt;P&gt;I agree that the 'Semantic Steward' is a necessary function. Right now, I see this responsibility often falling into a grey area between analytics engineers and business domain owners, which usually results in inconsistent definitions across BI tools and AI agents.&lt;/P&gt;&lt;P&gt;Rather than waiting for a new organizational role, I think we can start bridging this gap today by using agents to automate the discovery phase. Imagine a 'Semantic Consistency Agent' that scans Unity Catalog, BI metadata, and code repositories to flag discrepancies in metric definitions. It could use lineage to trace where logic diverges and highlight conflicting definitions for 'Active Customer' before they reach a dashboard or an AI assistant.&lt;/P&gt;&lt;P&gt;This doesn't replace the steward, but it shifts the human effort from manual research to resolution. It allows the team to focus on the business context and consensus-building, while the agent handles the heavy lifting of identifying where definitions have drifted. It is a practical way to begin governing meaning before the organization formally defines the role.&lt;/P&gt;</description>
      <pubDate>Fri, 23 Oct 2026 19:37:45 GMT</pubDate>
      <guid>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174474#M3054</guid>
      <dc:creator>slowder</dc:creator>
      <dc:date>2026-10-23T19:37:45Z</dc:date>
    </item>
    <item>
      <title>Re: Do We Need a New Role in Data Organizations — The Semantic Steward?</title>
      <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174523#M3056</link>
      <description>&lt;P&gt;&lt;SPAN&gt;Great perspective! I completely agree that a Semantic Consistency Agent could be a practical starting point. Automating semantic drift detection and lineage-based impact analysis would allow Semantic Stewards to focus on business alignment and governance decisions. &lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;AI can identify inconsistencies, but human accountability is essential to establish trusted business meaning.&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt; This could be a powerful step toward agent-driven semantic governance!&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Sat, 24 Oct 2026 22:58:29 GMT</pubDate>
      <guid>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174523#M3056</guid>
      <dc:creator>vkondepati</dc:creator>
      <dc:date>2026-10-24T22:58:29Z</dc:date>
    </item>
    <item>
      <title>Re: Do We Need a New Role in Data Organizations — The Semantic Steward?</title>
      <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174562#M3057</link>
      <description>&lt;P&gt;Exactly: use agents as a starting point, but decide up front which responsibilities require the human touch and optimize that experience!&lt;/P&gt;</description>
      <pubDate>Sun, 25 Oct 2026 17:32:38 GMT</pubDate>
      <guid>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174562#M3057</guid>
      <dc:creator>slowder</dc:creator>
      <dc:date>2026-10-25T17:32:38Z</dc:date>
    </item>
    <item>
      <title>Re: Do We Need a New Role in Data Organizations — The Semantic Steward?</title>
      <link>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174606#M3058</link>
      <description>&lt;P&gt;Hi&amp;nbsp;&lt;a href="https://community.databricks.com/t5/user/viewprofilepage/user-id/187220"&gt;@vkondepati&lt;/a&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;1. Is a new role necessary?&lt;BR /&gt;Not necessarily in every organization. A dedicated Semantic Steward makes sense when multiple business domains and AI applications rely on shared metrics. Smaller organizations could assign this responsibility to existing roles.&lt;/P&gt;&lt;P&gt;2.Who should own semantic governance?&lt;BR /&gt;Business domain owners should define what a metric means, while data stewards and analytics engineers help translate that definition into governed, tested implementations.&lt;/P&gt;&lt;P&gt;3.Should semantic governance be part of AI governance?&lt;BR /&gt;Yes. Access control determines what an AI assistant can access, but governed semantic definitions help determine whether it interprets the data correctly.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;One aspect i wnated to understand is when a shared KPI definition changes, how would you ensure that the update is consistently reflected across Unity Catalog Metric Views, BI dashboards, and AI assistants without creating conflicting results during the transition? Would you manage semantic definitions as version-controlled artifacts with automated validation and impact analysis, similar to CI/CD for data pipelines?&lt;/P&gt;</description>
      <pubDate>Mon, 26 Oct 2026 07:31:45 GMT</pubDate>
      <guid>https://community.databricks.com/t5/data-governance/do-we-need-a-new-role-in-data-organizations-the-semantic-steward/m-p/174606#M3058</guid>
      <dc:creator>Aravind_Reddy</dc:creator>
      <dc:date>2026-10-26T07:31:45Z</dc:date>
    </item>
  </channel>
</rss>

